Gaussian mixture models for signal mapping and positioning

Maps of RSS from a wireless transmitter can be used for positioning or for planning wireless infrastructure. The RSS values measured at a single point are not always the same, but follow some distribution, which vary from point to point. In existing approaches in the literature this variation is neglected or its mapping requires making many measurements at every point, which makes the measurement collection very laborious. We propose to use GMs for modeling joint distributions of the position and the RSS value. The proposed model is more versatile than methods found in the literature as it models the joint distribution of RSS measurements and the location space. This allows us to model the distributions of RSS values in every point of space without making many measurement in every point. In addition, GMs allow us to compute conditional probabilities and posteriors of position in closed form. The proposed models can model any RSS attenuation pattern, which is useful for positioning in multifloor buildings. Our tests with WLAN signals show that positioning with the proposed algorithm provides accurate position estimates. We conclude that the proposed algorithm can provide useful information about distributions of RSS values for different applications.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Research group: Automation and Systems Theory, Aalto University, University of Liverpool, Universidad Antonio de Nebrija, Uppsala University
Contributors: Raitoharju, M., García-Fernández, F., Hostettler, R., Piché, R., Särkkä, S.
Publication date: 1 Mar 2020
Peer-reviewed: Yes
Early online date: 10 Oct 2019

Publication information

Journal: Signal Processing
Volume: 168
Article number: 107330
ISSN (Print): 0165-1684
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Software, Signal Processing, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering
Keywords: Gaussian mixtures, Indoor positioning, RSS, Signal mapping, Statistical modeling
URLs: 
Source: Scopus
Source ID: 85073693932

Research output: Contribution to journalArticleScientificpeer-review

Bayesian receiver operating characteristic metric for linear classifiers

We propose a novel classifier accuracy metric: the Bayesian Area Under the Receiver Operating Characteristic Curve (CBAUC). The method estimates the area under the ROC curve and is related to the recently proposed Bayesian Error Estimator. The metric can assess the quality of a classifier using only the training dataset without the need for computationally expensive cross-validation. We derive a closed-form solution of the proposed accuracy metric for any linear binary classifier under the Gaussianity assumption, and study the accuracy of the proposed estimator using simulated and real-world data. These experiments confirm that the closed-form CBAUC is both faster and more accurate than conventional AUC estimators.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Research group: Computational Systems Biology, Computing Sciences, AI Virtanen Institute for Molecular Sciences, University of Eastern Finland
Contributors: Hassan, S. S., Huttunen, H., Niemi, J., Tohka, J.
Number of pages: 8
Pages: 52-59
Publication date: 1 Dec 2019
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition Letters
Volume: 128
ISSN (Print): 0167-8655
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
Keywords: Bayesian error estimation, Classification, Receiver operating characteristic curve

Bibliographical note

EXT="Tohka, Jussi"

Source: Scopus
Source ID: 85071016385

Research output: Contribution to journalArticleScientificpeer-review

Fast fourier color constancy and grayness index for ISPA illumination estimation challenge

We briefly introduce two submissions to the Illumination Estimation Challenge, in the Int'l Workshop on Color Vision, affiliated to the 11th Int'l Symposium on Image and Signal Processing and Analysis. The fourier-transform-based submission is ranked 3rd, and the statistical Gray-pixel-based one ranked 6th.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, South China University of Technology
Contributors: Qian, Y., Chen, K., Yu, H.
Number of pages: 3
Pages: 352-354
Publication date: 17 Oct 2019

Host publication information

Title of host publication: ISPA 2019 - 11th International Symposium on Image and Signal Processing and Analysis
Publisher: IEEE
Editors: Loncaric, S., Bregovic, R., Carli, M., Subasic, M.
ISBN (Electronic): 9781728131405

Publication series

Name: International Symposium on Image and Signal Processing and Analysis, ISPA
Volume: 2019-September
ISSN (Print): 1845-5921
ISSN (Electronic): 1849-2266
ASJC Scopus subject areas: Computational Theory and Mathematics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Color constancy, FFCC, Gray pixel, Illumination

Bibliographical note

EXT="Chen, Ke"

Source: Scopus
Source ID: 85074428933

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

An approval of MPPT based on pv cell’s simplified equivalent circuit during fast-shading conditions

The partial shading conditions significantly affect the functionality of solar power plants despite the presence of multiple maximum power point tracking systems. The primary cause of this problem is the presence of local maxima in the power–current and/or power–voltage characteristic curves that restrict the functionality of the conventional maximum power point tracking systems. The present article proposes a modified algorithm based on the simplified equivalent circuit of solar cells to improve the functionality of traditional maximum power point tracking systems. This algorithm provides a method for regularly monitoring the photo-current of each solar module. The upper and lower boundaries of the regulating parameter such as current or voltage are decided very precisely, which is helpful to find the location of the global maximum. During a sequential search, the control system accurately determines the lower and upper boundaries of the global maximum. Simultaneously, the maximum power point tracking system increases the photovoltaic current up to one of these boundaries and applies one of the conventional algorithms. Additionally, the control system regularly monitors the photovoltaic characteristics and changes the limits of regulating parameter concerning any change in global maximum location. This proposed method is fast and precise to locate the global maximum boundaries and to track global maximum even under fast-changing partial shading conditions. The improved performance and overall efficiency are validated by simulation study for variable solar irradiance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Automation Technology and Mechanical Engineering, Ariel University Center of Samaria
Contributors: Rajput, S., Averbukh, M., Yahalom, A., Minav, T.
Number of pages: 14
Publication date: 1 Sep 2019
Peer-reviewed: Yes

Publication information

Journal: Electronics (Switzerland)
Volume: 8
Issue number: 9
Article number: 1060
ISSN (Print): 2079-9292
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Signal Processing, Hardware and Architecture, Computer Networks and Communications, Electrical and Electronic Engineering
Keywords: Equivalent circuit, Global maximum, Maximum power point tracking, Partial shading, Photovoltaic system, Solar module
Electronic versions: 
Source: Scopus
Source ID: 85075081647

Research output: Contribution to journalArticleScientificpeer-review

Generalization of the K-SVD algorithm for minimization of β-divergence

In this paper, we propose, describe, and test a modification of the K-SVD algorithm. Given a set of training data, the proposed algorithm computes an overcomplete dictionary by minimizing the β-divergence (β>=1) between the data and its representation as linear combinations of atoms of the dictionary, under strict sparsity restrictions. For the special case β=2, the proposed algorithm minimizes the Frobenius norm and, therefore, for β=2 the proposed algorithm is equivalent to the original K-SVD algorithm. We describe the modifications needed and discuss the possible shortcomings of the new algorithm. The algorithm is tested with random matrices and with an example based on speech separation.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Universitat Politècnica de València
Contributors: Garcia-Molla, V. M., San Juan, P., Virtanen, T., Vidal, A. M., Alonso, P.
Number of pages: 7
Pages: 47-53
Publication date: 1 Sep 2019
Peer-reviewed: Yes

Publication information

Journal: Digital Signal Processing: A Review Journal
Volume: 92
ISSN (Print): 1051-2004
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Beta-divergence, K-SVD, Matching pursuit algorithms, NMF, Nonnegative K-SVD
Source: Scopus
Source ID: 85065411279

Research output: Contribution to journalArticleScientificpeer-review

Performance analysis of single-query 6-DoF camera pose estimation in self-driving setups

In this work, we consider the problem of single-query 6-DoF camera pose estimation, i.e. estimating the position and orientation of a camera by using reference images and a point cloud. We perform a systematic comparison of three state-of-the-art strategies for 6-DoF camera pose estimation: feature-based, photometric-based and mutual-information-based approaches. Two standard datasets with self-driving setups are used for experiments, and the performance of the studied methods is evaluated in terms of success rate, translation error and maximum orientation error. Building on the analysis of the results, we evaluate a hybrid approach that combines feature-based and mutual-information-based pose estimation methods to benefit from their complementary properties for pose estimation. Experiments show that (1) in cases with large appearance change between query and reference, the hybrid approach outperforms feature-based and mutual-information-based approaches by an average increment of 9.4% and 8.7% in the success rate, respectively; (2) in cases where query and reference images are captured at similar imaging conditions, the hybrid approach performs similarly as the feature-based approach, but outperforms both photometric-based and mutual-information-based approaches with a clear margin; (3) the feature-based approach is consistently more accurate than mutual-information-based and photometric-based approaches when at least 4 consistent matching points are found between the query and reference images.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Universidad Industrial de Santander, Czech Technical University in Prague
Contributors: Fu, J., Pertuz, S., Matas, J., Kämäräinen, J.
Pages: 58-73
Publication date: Sep 2019
Peer-reviewed: Yes

Publication information

Journal: Computer Vision and Image Understanding
Volume: 186
ISSN (Print): 1077-3142
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition
Keywords: 3D point cloud, Camera pose estimation, Hybrid method, Mutual information, Photometric matching, Self driving car

Bibliographical note

EXT="Matas, Jiri"

Source: Scopus
Source ID: 85067195521

Research output: Contribution to journalArticleScientificpeer-review

A perspective on wireless M-bus for smart electricity grids

The Internet of Things (IoT) enables long-range outdoor networks, such as smart grid and municipal lighting, as well as short-range indoor systems for smart homes, residential security, and energy management. Wireless connectivity and standardized communication protocols become an essential technology baseline for these diverse IoT applications. The focus of this work is wireless connectivity for smart metering systems. One of the recent protocols in this field is Wireless M-BUS, which is being widely utilized for remote metering applications across Europe. Therefore, in this paper, we detail a novel multi-platform framework designed to serve as a data generator for the protocol in question. The developed software allows to construct Wireless M-Bus telegrams with a high level of detail according to the EN 13757-4 specification and then schedule them for periodic transmission. The evaluation of the data generator is done in real scenario by using previously developed prototype equipped with IQRF TR72DA communication module acting as a smart meter with implemented software framework. As a result, the evaluation of communication distance between the developed Wireless MBus prototype and commercial gateway was tested in case of indoor scenario at Brno University of Technology, Faculty of Electrical Engineering and Communication.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electrical Engineering, Brno University of Technology, Department of Telecommunication
Contributors: Masek, P., Stusek, M., Zeman, K., Mozny, R., Ometov, A., Hosek, J.
Number of pages: 6
Pages: 730-735
Publication date: 1 Jul 2019

Host publication information

Title of host publication: 2019 42nd International Conference on Telecommunications and Signal Processing, TSP 2019
Publisher: IEEE
Editor: Herencsar, N.
ISBN (Electronic): 9781728118642
ASJC Scopus subject areas: Signal Processing, Computer Networks and Communications, Information Systems and Management
Keywords: Data generator, Energy efficient wireless communication, Low power communication, Machine-to-Machine, Smart Grid, Wireless M-BUS

Bibliographical note

EXT="Stusek, Martin"

Source: Scopus
Source ID: 85071067560

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Design and implementation of a multi-sensor newborn EEG seizure and background model with inter-channel field characterization

This paper presents a novel multi-sensor non-stationary EEG model; it is obtained by combining state of the art mono-sensor newborn EEG simulators, a multilayer newborn head model comprised of four homogeneous concentric spheres, a multi-sensor propagation scheme based on array processing and optical dispersion to calculate inter-channel attenuation and delay, and lastly, a multi-variable optimization paradigm using particle swarm optimization and Monte-Carlo simulations to validate the model for optimal conditions. Multi-sensor EEG of 7 newborns, comprised of seizure and background epochs, are analyzed using time-space, time-frequency, power maps and multi-sensor causality techniques. The outcomes of these methods are validated by medical insights and serve as a backbone for any assumptions and as performance benchmarks for the model to be evaluated against. The results obtained with the developed model show 85.7% averaged time-frequency correlation (which is the selected measure for similarity with real EEG)with 5.9% standard deviation, and the averaged error obtained is 34.6% with 8% standard deviation. The resulting performances indicate that the proposed model provides a suitable matching fit with real EEG in terms of their probability density function, inter-sensor attenuation and translation, and multi-sensor causality. They also demonstrate the model flexibility to generate new unseen samples by utilizing user-defined parameters, making it suitable for other relevant applications.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Qatar University, University of Queensland
Contributors: Al-Sa'd, M. F., Boashash, B.
Number of pages: 29
Pages: 71-99
Publication date: 1 Jul 2019
Peer-reviewed: Yes

Publication information

Journal: Digital Signal Processing: A Review Journal
Volume: 90
ISSN (Print): 1051-2004
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: EEG analysis, Multi-channel EEG, Multi-sensor propagation, Time-frequency processing, Time-space analysis
Source: Scopus
Source ID: 85065143207

Research output: Contribution to journalArticleScientificpeer-review

Input magnitude data setting in error-reduction algorithm for one-dimensional discrete phase retrieval problem

In this paper we discuss how does the input magnitude data setting influence the behavior of error-reduction algorithm in the case of the one-dimensional discrete phase retrieval problem. We present experimental results related to the convergence or stagnation of the algorithm. We also discuss the issue of the zeros distribution of the solution, when the solution of the problem exists.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, FETTI, Technical University of Cluj-NapocaUniversitatea Tehnica din Cluj-Napoca
Contributors: Rusu, C., Astola, J.
Publication date: 1 Jul 2019

Host publication information

Title of host publication: ISSCS 2019 - International Symposium on Signals, Circuits and Systems
Publisher: IEEE
Article number: 8801743
ISBN (Electronic): 9781728138961
ASJC Scopus subject areas: Computer Vision and Pattern Recognition, Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering

Bibliographical note

EXT="Rusu, Corneliu"

Source: Scopus
Source ID: 85071848180

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

On the performance of narrow-band internet of things (NB-IoT) for delay-tolerant services

Narrowband IoT (NB-IoT) stands for a radio access technology standardized by the 3GPP organization in Release 13 to enable a large set of use-cases for massive Machine-type Communications (mMTCs). Compared to legacy human-oriented 4G (LTE) communication systems, NB-IoT has game-changing features in terms of extended coverage, enhanced power saving modes, and a reduced set of available functionality. At the end of the day, these features allow for connectivity of devices in challenging positions, enabling long battery life and reducing device complexity. This article addresses the development of the universal testing device for delay-tolerant services allowing for in-depth verification of NB-IoT communication parameters. The presented outputs build upon our long-term cooperation with the Vodafone Czech Republic a.s. company.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electrical Engineering, Brno University of Technology, Department of Telecommunication
Contributors: Mozny, R., Masek, P., Stusek, M., Zeman, K., Ometov, A., Hosek, J.
Number of pages: 6
Pages: 637-642
Publication date: 1 Jul 2019

Host publication information

Title of host publication: 2019 42nd International Conference on Telecommunications and Signal Processing, TSP 2019
Publisher: IEEE
Editor: Herencsar, N.
ISBN (Electronic): 9781728118642

Publication series

Name: 2019 42nd International Conference on Telecommunications and Signal Processing, TSP 2019
ASJC Scopus subject areas: Signal Processing, Computer Networks and Communications, Information Systems and Management
Keywords: IoT, mMTC, LPWAN, NB-IoT, Network services

Bibliographical note

EXT="Stusek, Martin"

Source: Scopus
Source ID: 85071063087

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

User Experience Study of 360° Music Videos on Computer Monitor and Virtual Reality Goggles

360° videos are increasingly used for media and entertainment, but the best practices for editing them are not yet well established. In this paper, we present a study in which we investigated the user experience of 360° music videos viewed on computer monitor and VR goggles. The research was conducted in the form of a laboratory experiment with 20 test participants. During the within-subject study, participants watched and evaluated four versions of the same 360° music video with a different cutting rate. Based on the results, an average cutting rate of 26 seconds delivered the highest-quality user experience both for computer monitor and VR goggles. The cutting rate matched with participants' mental models, and there was enough time to explore the environment without getting bored. Faster cutting rates made the users nervous, and a video consisting of a single shot was considered to be too static and boring.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Tampere University, Tampere University of Applied Sciences
Contributors: Holm, J., Väänänen, K., Remans, M. M. R.
Number of pages: 7
Pages: 81-87
Publication date: 1 Jul 2019

Host publication information

Title of host publication: Information Visualization - Biomedical Visualization and Geometric Modelling and Imaging, IV 2019
Publisher: IEEE
Editors: Banissi, E., Ursyn, A., McK. Bannatyne, M. W., Datia, N., Pires, J. M., Francese, R., Sarfraz, M., Wyeld, T. G., Bouali, F., Venturin, G., Azzag, H., Lebbah, M., Trutschl, M., Cvek, U., Muller, H., Nakayama, M., Kernbach, S., Caruccio, L., Risi, M., Erra, U., Vitiello, A., Rossano, V.
ISBN (Electronic): 9781728128382

Publication series

Name: Proceedings of the International Conference on Information Visualisation
ISSN (Print): 1093-9547
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition
Keywords: 360° video, cutting rate, hmd, music video, virtual reality, virtual reality goggles

Bibliographical note

jufoid=58079
EXT="Holm, Jukka"
INT=comp,"Remans, Mohammad Mushfiqur Rahman"

Source: Scopus
Source ID: 85072286445

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Socially inspired relaying and proactive mode selection in mmWave vehicular communications

As the Internet of Vehicles matures and acquires its social flavor, novel wireless connectivity enablers are being demanded for reliable data transfer in high-rate applications. The recently ratified New Radio communications technology operates in millimeter-wave (mmWave) spectrum bands and offers sufficient capacity for bandwidth-hungry services. However, seamless operation over mmWave is difficult to maintain on the move, since such extremely high frequency radio links are susceptible to unexpected blockage by various obstacles, including vehicle bodies. As a result, proactive mode selection, that is, migration from infrastructure- to vehicle-based connections and back, is becoming vital to avoid blockage situations. Fortunately, the very social structure of interactions between the neighboring smart cars and their passengers may be leveraged to improve session continuity by relaying data via proximate vehicles. This paper conceptualizes the socially inspired relaying scenarios, conducts underlying mathematical analysis, continues with a detailed 3-D modeling to facilitate proactive mode selection, and concludes by discussing a practical prototype of a vehicular mmWave platform.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Electrical Engineering, University of California, Los Angeles
Contributors: Moltchanov, D., Kovalchukov, R., Gerasimenko, M., Andreev, S., Koucheryavy, Y., Gerla, M.
Number of pages: 12
Pages: 5172-5183
Publication date: 1 Jun 2019
Peer-reviewed: Yes

Publication information

Journal: IEEE Internet of Things Journal
Volume: 6
Issue number: 3
ISSN (Print): 2327-4662
Original language: English
ASJC Scopus subject areas: Signal Processing, Information Systems, Hardware and Architecture, Computer Science Applications, Computer Networks and Communications
Keywords: Internet of Things, Millimeter wave (mmWave) communication, Social network services, Vehicular ad hoc networks
Electronic versions: 
Source: Scopus
Source ID: 85067875266

Research output: Contribution to journalArticleScientificpeer-review

1-D Convolutional Neural Networks for Signal Processing Applications

1D Convolutional Neural Networks (CNNs) have recently become the state-of-the-art technique for crucial signal processing applications such as patient-specific ECG classification, structural health monitoring, anomaly detection in power electronics circuitry and motor-fault detection. This is an expected outcome as there are numerous advantages of using an adaptive and compact 1D CNN instead of a conventional (2D) deep counterparts. First of all, compact 1D CNNs can be efficiently trained with a limited dataset of 1D signals while the 2D deep CNNs, besides requiring 1D to 2D data transformation, usually need datasets with massive size, e.g., in the »Big Data» scale in order to prevent the well-known »overfitting» problem. 1D CNNs can directly be applied to the raw signal (e.g., current, voltage, vibration, etc.) without requiring any pre- or post-processing such as feature extraction, selection, dimension reduction, denoising, etc. Furthermore, due to the simple and compact configuration of such adaptive 1D CNNs that perform only linear 1D convolutions (scalar multiplications and additions), a real-time and low-cost hardware implementation is feasible. This paper reviews the major signal processing applications of compact 1D CNNs with a brief theoretical background. We will present their state-of-the-art performances and conclude with focusing on some major properties. Keywords - 1-D CNNs, Biomedical Signal Processing, SHM.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Research group: Multimedia Research Group - MRG, Qatar University, Electrical and Electronics Engineering Department, Izmir University of Economics
Contributors: Kiranyaz, S., Ince, T., Abdeljaber, O., Avci, O., Gabbouj, M.
Number of pages: 5
Pages: 8360-8364
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering

Bibliographical note

EXT="Kiranyaz, Serkan"
EXT="Ince, Turker"

Source: Scopus
Source ID: 85068995333

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Convex Energy Optimization of Streaming Applications for MPSoCs

The energy efficiency of modern MPSoCs is enhanced by complex hardware features such as Dynamic Voltage and Frequency Scaling (DVFS) and Dynamic Power Management (DPM). This paper introduces a new method, based on convex problem solving, that determines the most energy efficient operating point in terms of frequency and number of active cores in an MPSoC. The solution can challenge the popular approaches based on never-idle (or As-Slow-As-Possible (ASAP)) and race-to-idle (or As-Fast-As-Possible (AFAP)) principles. Experimental data are reported using a Samsung Exynos 5410 MPSoC and show a reduction in energy of up to 27 % when compared to ASAP and AFAP.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Universite de Rennes
Contributors: Nogues, E., Mercat, A., Arrestier, F., Pelcat, M., Menard, D.
Number of pages: 5
Pages: 1557-1561
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Source: Scopus
Source ID: 85068997223

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Deep Learning for Audio Signal Processing

Given the recent surge in developments of deep learning, this paper provides a review of the state-of-the-art deep learning techniques for audio signal processing. Speech, music, and environmental sound processing are considered side-by-side, in order to point out similarities and differences between the domains, highlighting general methods, problems, key references, and potential for cross fertilization between areas. The dominant feature representations (in particular, log-mel spectra and raw waveform) and deep learning models are reviewed, including convolutional neural networks, variants of the long short-term memory architecture, as well as more audio-specific neural network models. Subsequently, prominent deep learning application areas are covered, i.e., audio recognition (automatic speech recognition, music information retrieval, environmental sound detection, localization and tracking) and synthesis and transformation (source separation, audio enhancement, generative models for speech, sound, and music synthesis). Finally, key issues and future questions regarding deep learning applied to audio signal processing are identified.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Aalborg University, Google Inc, University of Toulon, Austrian Research Institute for Artificial Intelligence
Contributors: Purwins, H., Li, B., Virtanen, T., Schlüter, J., Chang, S. Y., Sainath, T.
Number of pages: 14
Pages: 206-219
Publication date: 1 May 2019
Peer-reviewed: Yes

Publication information

Journal: IEEE Journal on Selected Topics in Signal Processing
Volume: 13
Issue number: 2
ISSN (Print): 1932-4553
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: audio enhancement, automatic speech recognition, connectionist temporal memory, Deep learning, environmental sounds, music information retrieval, source separation
Source: Scopus
Source ID: 85065962111

Research output: Contribution to journalArticleScientificpeer-review

Deep Reinforcement Learning for Financial Trading Using Price Trailing

Developing accurate financial analysis tools can be useful both for speculative trading, as well as for analyzing the behavior of markets and promptly responding to unstable conditions ensuring the smooth operation of the financial markets. This led to the development of various methods for analyzing and forecasting the behaviour of financial assets, ranging from traditional quantitative finance to more modern machine learning approaches. However, the volatile and unstable behavior of financial markets forbids the accurate prediction of future prices, reducing the performance of these approaches. In contrast, in this paper we propose a novel price trailing method that goes beyond traditional price forecasting by reformulating trading as a control problem, effectively overcoming the aforementioned limitations. The proposed method leads to developing robust agents that can withstand large amounts of noise, while still capturing the price trends and allowing for taking profitable decisions.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Aristotle University of Thessaloniki, School of Informatics
Contributors: Zarkias, K. S., Passalis, N., Tsantekidis, A., Tefas, A.
Number of pages: 5
Pages: 3067-3071
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Deep Reinforcement Learning, Financial Markets, Price Forecasting, Trading

Bibliographical note

EXT="Tefas, Anastasios"

Source: Scopus
Source ID: 85068989312

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Deep Temporal Logistic Bag-of-features for Forecasting High Frequency Limit Order Book Time Series

Forecasting time series has several applications in various domains. The vast amount of data that are available nowadays provide the opportunity to use powerful deep learning approaches, but at the same time pose significant challenges of high-dimensionality, velocity and variety. In this paper, a novel logistic formulation of the well-known Bag-of-Features model is proposed to tackle these challenges. The proposed method is combined with deep convolutional feature extractors and is capable of accurately modeling the temporal behavior of time series, forming powerful forecasting models that can be trained in an end-to-end fashion. The proposed method was extensively evaluated using a large-scale financial time series dataset, that consists of more than 4 million limit orders, outperforming other competitive methods.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, School of Informatics, Aristotle University of Thessaloniki, Aarhus Universitet
Contributors: Passalis, N., Tefas, A., Kanniainen, J., Gabbouj, M., Iosifidis, A.
Number of pages: 5
Pages: 7545-7549
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Limit Order Book, Temporal Bag-of-Features, Time series forecasting

Bibliographical note

EXT="Tefas, Anastasios"
EXT="Iosifidis, Alexandros"

Source: Scopus
Source ID: 85069000578

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Low-latency Deep Clustering for Speech Separation

This paper proposes a low algorithmic latency adaptation of the deep clustering approach to speaker-independent speech separation. It consists of three parts: a) the usage of long-short-term-memory (LSTM) networks instead of their bidirectional variant used in the original work, b) using a short synthesis window (here 8 ms) required for low-latency operation, and, c) using a buffer in the beginning of audio mixture to estimate cluster centres corresponding to constituent speakers which are then utilized to separate speakers within the rest of the signal. The buffer duration would serve as an initialization phase after which the system is capable of operating with 8 ms algorithmic latency. We evaluate our proposed approach on two-speaker mixtures from Wall Street Journal (WSJ0) corpus. We observe that the use of LSTM yields around one dB lower SDR as compared to the baseline bidirectional LSTM in terms of source to distortion ratio (SDR). Moreover, using an 8 ms synthesis window instead of 32 ms degrades the separation performance by around 2.1 dB as compared to the baseline. Finally, we also report separation performance with different buffer durations noting that separation can be achieved even for buffer duration as low as 300 ms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Research group: Audio research group - ARG
Contributors: Wang, S., Naithani, G., Virtanen, T.
Number of pages: 5
Pages: 76-80
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Deep clustering, Low latency, Monaural speech separation

Bibliographical note

int=comp,"Wang, Shanshan"

Source: Scopus
Source ID: 85068960960

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Open framework for mammography-based breast cancer risk assessment

In recent years, several studies have established a relationship between mammographic parenchymal patterns and breast cancer risk. However, there is a lack of publicly available data and software for objective comparison and clinical validation. This paper presents an open and adaptable implementation (OpenBreast v1.0) of a fully-Automatic computerized framework for mammographic image analysis for breast cancer risk assessment. OpenBreast implements mammographic image analysis in four stages: breast segmentation, detection of region-of-interests, feature extraction and risk scoring. For each stage, we provide implementations of several state-of-The-Art methods. The pipeline is tested on a set of 305 full-field digital mammography images corresponding to 84 patients (51 cases and 49 controls) from the breast cancer digital repository (BCDR). OpenBreast achieves a competitive AUC of 0.846 in breast cancer risk assessment. In addition, used jointly with widely accepted risk factors such as patient age and breast density, mammographic image analysis using OpenBreast shows a statistically significant improvement in performance with an AUC of 0.876 (\mathrm{p}<0.001). Our framework will be made publicly available and it is easy to incorporate new methods.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Research group: Vision, Universidad Industrial de Santander, Tampere University, Brigham and Women's Hospital
Contributors: Pertuz, S., Torres, G. F., Tamimi, R., Kämäräinen, J.
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE EMBS International Conference on Biomedical and Health Informatics, BHI 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781728108483
ASJC Scopus subject areas: Artificial Intelligence, Signal Processing, Information Systems and Management, Biomedical Engineering, Health Informatics, Radiology Nuclear Medicine and imaging
Keywords: Breast cancer, Mammography, Parenchymal analysis, Risk assessment, Texture analysis

Bibliographical note

EXT="Pertuz, Said"

Source: Scopus
Source ID: 85073005187

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Random Forest Oriented Fast QTBT Frame Partitioning

Block partition structure is a critical module in video coding scheme to achieve significant gap of compression performance. Under the exploration of future video coding standard by the Joint Video Exploration Team (JVET), named Versatile Video Coding (VVC), a new Quad Tree Binary Tree (QTBT) block partition structure has been introduced. In addition to the QT block partitioning defined by High Efficiency Video Coding (HEVC) standard, new horizontal and vertical BT partitions are enabled, which drastically increases the encoding time compared to HEVC. In this paper, we propose a fast QTBT partitioning scheme based on a Machine Learning approach. Complementary to techniques proposed in literature to reduce the complexity of HEVC Quad Tree (QT) partitioning, the propose solution uses Random Forest classifiers to determine for each block which partition modes between QT and BT is more likely to be selected. Using uncertainty zones of classifier decisions, the proposed complexity reduction technique is able to reduce in average by 30% the encoding time of JEM-v7.0 software in Random Access configuration with only 0.57% Bjontegaard Delta Rate (BD-BR) increase.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Universite de Rennes, MMP
Contributors: Amestoy, T., Mercat, A., Hamidouche, W., Bergeron, C., Menard, D.
Number of pages: 5
Pages: 1837-1841
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Complexity Reduction, JEM, Machine Learning, QTBT, Random Forest, Video Compression, VVC
Source: Scopus
Source ID: 85068966990

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Time Difference of Arrival Estimation of Speech Signals Using Deep Neural Networks with Integrated Time-frequency Masking

The Time Difference of Arrival (TDoA) of a sound wavefront impinging on a microphone pair carries spatial information about the source. However, captured speech typically contains dynamic non-speech interference sources and noise. Therefore, the TDoA estimates fluctuate between speech and interference. Deep Neural Networks (DNNs) have been applied for Time-Frequency (TF) masking for Acoustic Source Localization (ASL) to filter out non-speech components from a speaker location likelihood function. However, the type of TF mask for this task is not obvious. Secondly, the DNN should estimate the TDoA values, but existing solutions estimate the TF mask instead. To overcome these issues, a direct formulation of the TF masking as a part of a DNN-based ASL structure is proposed. Furthermore, the proposed network operates in an online manner, i.e., producing estimates frame-by-frame. Combined with the use of recurrent layers it exploits the sequential progression of speaker related TDoAs. Training with different microphone spacings allows model re-use for different microphone pair geometries in inference. Real-data experiments with smartphone recordings of speech in interference demonstrate the network's generalization capability.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences
Contributors: Pertilä, P., Parviainen, M.
Number of pages: 5
Pages: 436-440
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Acoustic Source Localization, Microphone Arrays, Recurrent Neural Networks, Time-Frequency Masking
Electronic versions: 
Source: Scopus
Source ID: 85068957469

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Variance Preserving Initialization for Training Deep Neuromorphic Photonic Networks with Sinusoidal Activations

Photonic neuromorphic hardware can provide significant performance benefits for Deep Learning (DL) applications by accelerating and reducing the energy requirements of DL models. However, photonic neuromorphic architectures employ different activation elements than those traditionally used in DL, slowing down the convergence of the training process for such architectures. An initialization scheme that can be used to efficiently train deep photonic networks that employ quadratic sinusoidal activation functions is proposed in this paper. The proposed initialization scheme can overcome these limitations, leading to faster and more stable training of deep photonic neural networks. The ability of the proposed method to improve the convergence of the training process is experimentally demonstrated using two different DL architectures and two datasets.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Aristotle University of Thessaloniki
Contributors: Passalis, N., Mourgias-Alexandris, G., Tsakyridis, A., Pleros, N., Tefas, A.
Number of pages: 5
Pages: 1483-1487
Publication date: 1 May 2019

Host publication information

Title of host publication: 2019 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2019 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781479981311
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Neuromorphic Hardware, Photonic Neural Networks, Sinusoidal Activations

Bibliographical note

EXT="Tefas, Anastasios"

Source: Scopus
Source ID: 85064389224

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Automatic detection of water inside concrete slabs using ground penetrating radar

Ground Penetrating Radar (GPR) is generally used as a non-destructive method of inspection for structures and for finding defects in concrete slabs. In this paper, GPR is used in the detection of water inside the cavities of concrete hollow core slabs. We propose an algorithm that determines the water level inside the concrete slab by analyzing the time delays of the reflections originating from inside the cavity. The algorithm is based on utilizing prior knowledge about the geometry of the hollow core slab. The presence of water was successfully detected and an estimate for the height of the water surface was obtained with a GPR system operating with a central frequency of 2.7 GHz. Based on the experiments, the proposed method holds promise in providing a robust and accurate method for the detection of water inside the concrete slabs. Results, possible future research and analysis of the feasibility of GPR systems in water detection are presented and discussed.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences, Research group: MMDM
Contributors: Khan, M. A., Vehmas, R., Visa, A.
Publication date: 1 Apr 2019

Host publication information

Title of host publication: 2019 IEEE Radar Conference, RadarConf 2019
Publisher: IEEE
ISBN (Electronic): 9781728116792
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Instrumentation

Bibliographical note

jufoid=57477

Source: Scopus
Source ID: 85073122917

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Exploiting Multipath Terahertz Communications for Physical Layer Security in beyond 5G Networks

Terahertz (THz) band communications, capable of achieving the theoretical capacity of up to several terabits-per-second, are one of the attractive enablers for beyond 5G wireless networks. THz systems will use extremely directional narrow beams, allowing not only to extend the communication range but also to partially secure the data already at the physical layer. The reason is that, in most cases, the Attacker has to be located within the transmitter beam in order to eavesdrop the message. However, even the use of very narrow beams results in the considerably large area around the receiver, where the Attacker can capture all the data. In this paper, we study how to decrease the message eavesdropping probability by leveraging the inherent multi-path nature of the THz communications. We particularly propose sharing the data transmission over multiple THz propagation paths currently available between the communicating entities. We show that, at a cost of the slightly reduced link capacity, the message eavesdropping probability in the described scheme decreases significantly even when several Attackers operate in a cooperative manner. The proposed solution can be utilized for the transmission of the sensitive data, as well as to secure the key exchange in THz band networks beyond 5G.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electrical Engineering, Research group: Wireless Communications and Positioning, State University of New York
Contributors: Petrov, V., Moltchanov, D., Jornet, J. M., Koucheryavy, Y.
Number of pages: 8
Pages: 865-872
Publication date: 1 Apr 2019

Host publication information

Title of host publication: INFOCOM 2019 - IEEE Conference on Computer Communications Workshops, INFOCOM WKSHPS 2019
Publisher: IEEE
ISBN (Electronic): 9781728118789
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Information Systems and Management, Safety, Risk, Reliability and Quality, Computer Networks and Communications
Source: Scopus
Source ID: 85073261652

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Convolutional low-resolution fine-grained classification

Successful fine-grained image classification methods learn subtle details between visually similar (sub-)classes, but the problem becomes significantly more challenging if the details are missing due to low resolution. Encouraged by the recent success of Convolutional Neural Network (CNN) architectures in image classification, we propose a novel resolution-aware deep model which combines convolutional image super-resolution and convolutional fine-grained classification into a single model in an end-to-end manner. Extensive experiments on multiple benchmarks demonstrate that the proposed model consistently performs better than conventional convolutional networks on classifying fine-grained object classes in low-resolution images.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Vision
Contributors: Cai, D., Chen, K., Qian, Y., Kämäräinen, J.
Pages: 166-171
Publication date: Mar 2019
Peer-reviewed: Yes
Early online date: 2017

Publication information

Journal: Pattern Recognition Letters
Volume: 119
ISSN (Print): 0167-8655
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
Keywords: Deep learning, Fine-grained image classification, Super resolution convoluational neural networks
Source: Scopus
Source ID: 85032974725

Research output: Contribution to journalArticleScientificpeer-review

Energy Detection-Based Spectrum Sensing over Fisher-Snedecor F Fading Channels

This paper investigates the performance of energy detection-based spectrum sensing over Fisher-Snedecor F fading channels. To this end, an analytical expression for the corre- sponding average detection probability is firstly derived and then this is extended to account for collaborative spectrum sensing. The complementary receiver operating characteristics (ROC) are analyzed for different conditions of the average signal-to- noise ratio (SNR), time-bandwidth product, multipath fading, shadowing and number of collaborating users. It is shown that the energy detection performance is strongly linked to the severity of the multipath fading and amount of shadowing, whereby even small variations in either of these physical phenomena significantly impact the detection probability. Also, the versatile modeling capability of the Fisher-Snedecor F distribution is veridfied in the context of energy detection based spectrum sensing as it provides considerably more accurate characterization than the conventional Rayleigh fading model. To confirm the validity of the analytical results presented in this paper, we compare them with the results of some simulations.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Research group: Wireless Communications and Positioning, Electrical Engineering, Queen's University, Belfast, Northern Ireland, Centre for Wireless Innovation, Khalifa University, Department of Chemistry and Bioengineering, University of Surrey, German Jordanian University, Aristotle University of Thessaloniki
Contributors: Yoo, S. K., Cotton, S. L., Sofotasios, P. C., Muhaidat, S., Badarneh, O. S., Karagiannidis, G. K.
Publication date: 20 Feb 2019

Host publication information

Title of host publication: 2018 IEEE Global Communications Conference
Publisher: IEEE
Article number: 8647778
ISBN (Electronic): 9781538647271
ASJC Scopus subject areas: Information Systems and Management, Renewable Energy, Sustainability and the Environment, Safety, Risk, Reliability and Quality, Signal Processing, Modelling and Simulation, Instrumentation, Computer Networks and Communications
Source: Scopus
Source ID: 85063500863

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Improved Session Continuity in 5G NR with Joint Use of Multi-Connectivity and Guard Bandwidth

The intermittent millimeter-wave radio links as a result of human-body blockage are an inherent feature of the 5G New Radio (NR) technology by 3GPP. To improve session continuity in these emerging systems, two mechanisms have recently been proposed, namely, multi-connectivity and guard bandwidth. The former allows to establish multiple spatially-diverse connections and switch between them dynamically, while the latter reserves a fraction of system bandwidth for sessions changing their state from non-blocked to blocked, which ensures that the ongoing sessions have priority over the new ones. In this paper, we assess the joint performance of these two schemes for the user- and system-centric metrics of interest. Our numerical results reveal that the multi-connectivity operation alone may not suffice to increase the ongoing session drop probability considerably. On the other hand, the use of guard bandwidth significantly improves session continuity by somewhat compromising new session drop probability and system resource utilization. Surprisingly, the 5G NR system implementing both these techniques inherits their drawbacks. However, complementing it with an initial AP selection procedure effectively alleviates these limitations by maximizing the system resource utilization, while still providing sufficient flexibility to enable the desired trade-off between new and ongoing session drop probabilities.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electrical Engineering, Department of Chemistry and Bioengineering, Peoples’ Friendship University of Russia
Contributors: Kovalchukov, R., Moltchanov, D., Begishev, V., Samuylov, A., Andreev, S., Koucheryavy, Y., Samouylov, K.
Publication date: 20 Feb 2019

Host publication information

Title of host publication: 2018 IEEE Global Communications Conference, GLOBECOM 2018
Publisher: IEEE
ISBN (Electronic): 9781538647271
ASJC Scopus subject areas: Information Systems and Management, Renewable Energy, Sustainability and the Environment, Safety, Risk, Reliability and Quality, Signal Processing, Modelling and Simulation, Instrumentation, Computer Networks and Communications
Source: Scopus
Source ID: 85063532495

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

ALMARVI System Solution for Image and Video Processing in Healthcare, Surveillance and Mobile Applications

ALMARVI is a collaborative European research project funded by Artemis involving 16 industrial as well as academic partners across 4 countries, working together to address various computational challenges in image and video processing in 3 application domains: healthcare, surveillance and mobile. This paper is an editorial for a special issue discussing the integrated system created by the partners to serve as a cross-domain solution for the project. The paper also introduces the partner articles published in this special issue to discuss the various technological developments achieved within ALMARVI spanning all system layers, from hardware to applications. We illustrate the challenges faced within the project based on use cases from the three targeted application domains, and how these can address the 4 main project objectives addressing 4 challenges faced by high performance image and video processing systems: massive data rate, low power consumption, composability and robustness. We present a system stack composed of algorithms, design frameworks and platforms as a solution to these challenges. Finally, the use cases from the three different application domains are mapped on the system stack solution and are evaluated based on their performance for each of the 4 ALMARVI objectives.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, Delft University of Technology, Philips Healthcare Nederland
Contributors: Al-Ars, Z., van der Vlugt, S., Jääskeläinen, P., van der Linden, F.
Pages: 1-7
Publication date: Jan 2019
Peer-reviewed: Yes
Early online date: 2018

Publication information

Journal: Journal of Signal Processing Systems
Volume: 91
Issue number: 1
ISSN (Print): 1939-8018
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Source: Scopus
Source ID: 85057058836

Research output: Contribution to journalArticleScientificpeer-review

Digital Predistortion for 5G Small Cell: GPU Implementation and RF Measurements

In this paper, we present a high data rate implementation of a digital predistortion (DPD) algorithm on a modern mobile multicore CPU containing an on-chip GPU. The proposed implementation is capable of running in real-time, thanks to the execution of the predistortion stage inside the GPU, and the execution of the learning stage on a separate CPU core. This configuration, combined with the low complexity DPD design, allows for more than 400 Msamples/s sample rates. This is sufficient for satisfying 5G new radio (NR) base station radio transmission specifications in the sub-6 GHz bands, where signal bandwidths up to 100 MHz are specified. The linearization performance is validated with RF measurements on two base station power amplifiers at 3.7 GHz, showing that the 5G NR downlink emission requirements are satisfied.

General information

Publication status: E-pub ahead of print
MoE publication type: A1 Journal article-refereed
Organisations: Electrical Engineering, Computing Sciences, Research area: Computer engineering, Research group: Wireless Communications and Positioning, University of Vaasa (UVA), Tampere University
Contributors: Pascual Campo, P., Lampu, V., Meirhaeghe, A., Boutellier, J., Anttila, L., Valkama, M.
Number of pages: 12
Publication date: 2019
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
ISSN (Print): 1939-8018
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: 5G, Digital predistortion (DPD), GPU, High data rate, Real-time

Bibliographical note

INT=comp,"Meirhaeghe, Alexandre"

Source: Scopus
Source ID: 85077054281

Research output: Contribution to journalArticleScientificpeer-review

The Direction Cosine Matrix Algorithm in Fixed-point: Implementation and Analysis

Inertial navigation allows tracking and updating the position and orientation of a moving object based on accelerometer and gyroscope data without external positioning aid, such as GPS. Therefore, inertial navigation is an essential technique for, e.g., indoor positioning. As inertial navigation is based on integration of acceleration vector components, computation errors accumulate and make the position and orientation estimate drift. Even though maximum computation precision is desired, also efficiency needs consideration in the age of Internet-of-Things, to enable deployment of inertial navigation based applications to the smallest devices. This work formulates the Direction Cosine Matrix update algorithm, a central component for inertial navigation, in fixed-point and analyzes its precision and computation load compared to a regular floating-point implementation. The results show that the fixed-point version maintains very high precision, while requiring no floating point hardware for operation. The paper presents execution time results on three very different embedded processors.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Computing Sciences
Contributors: Meirhaeghe, A., Boutellier, J., Collin, J.
Publication date: 2019

Host publication information

Title of host publication: ICASSP 2019 - 2019 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher: IEEE
ISBN (Print): 978-1-4799-8132-8
ISBN (Electronic): 978-1-4799-8131-1
ASJC Scopus subject areas: Signal Processing

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Video coding of dynamic 3D point cloud data

Due to the increased popularity of augmented (AR) and virtual (VR) reality experiences, the interest in representing the real world in an immersive fashion has never been higher. Distributing such representations enables users all over the world to freely navigate in never seen before media experiences. Unfortunately, such representations require a large amount of data, not feasible for transmission on today's networks. Thus, efficient compression technologies are in high demand. This paper proposes an approach to compress 3D video data utilizing 2D video coding technology. The proposed solution was developed to address the needs of "tele-immersive" applications, such as VR, AR, or mixed reality with "Six Degrees of Freedom" capabilities. Volumetric video data is projected on 2D image planes and compressed using standard 2D video coding solutions. A key benefit of this approach is its compatibility with readily available 2D video coding infrastructure. Furthermore, objective and subjective evaluation shows significant improvement in coding efficiency over reference technology. The proposed solution was contributed and evaluated in international standardization. Although it is was not selected as the winning proposal, as very similar solution has been selected developed since then.

General information

Publication status: E-pub ahead of print
MoE publication type: A1 Journal article-refereed
Organisations: Computing Sciences, Nokia Technologies
Contributors: Schwarz, S., Sheikhipour, N., Fakour Sevom, V., Hannuksela, M. M.
Number of pages: 12
Publication date: 2019
Peer-reviewed: Yes

Publication information

Journal: APSIPA Transactions on Signal and Information Processing
Article number: e31
ISSN (Print): 2048-7703
Original language: English
ASJC Scopus subject areas: Signal Processing, Information Systems
Keywords: AR, Immersive media, Point cloud coding, Volumetric video
Source: Scopus
Source ID: 85076929574

Research output: Contribution to journalArticleScientificpeer-review

Cascade of Boolean detector combinations

This paper considers a scenario when we have multiple pre-trained detectors for detecting an event and a small dataset for training a combined detection system. We build the combined detector as a Boolean function of thresholded detector scores and implement it as a binary classification cascade. The cascade structure is computationally efficient by providing the possibility to early termination. For the proposed Boolean combination function, the computational load of classification is reduced whenever the function becomes determinate before all the component detectors have been utilized. We also propose an algorithm, which selects all the needed thresholds for the component detectors within the proposed Boolean combination. We present results on two audio-visual datasets, which prove the efficiency of the proposed combination framework. We achieve state-of-the-art accuracy with substantially reduced computation time in laughter detection task, and our algorithm finds better thresholds for the component detectors within the Boolean combination than the other algorithms found in the literature.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing
Contributors: Mahkonen, K., Virtanen, T., Kämäräinen, J.
Publication date: Dec 2018
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Image and Video Processing
Volume: 2018
Article number: 61
ISSN (Print): 1687-5176
Ratings: 
  • Scopus rating (2018): CiteScore 2.1 SJR 0.342 SNIP 1.002
Original language: English
ASJC Scopus subject areas: Signal Processing, Information Systems, Electrical and Electronic Engineering
Keywords: Binary classification, Boolean combination, Classification cascade
Source: Scopus
Source ID: 85050598957

Research output: Contribution to journalArticleScientificpeer-review

Efficient Solving of Markov Decision Processes on GPUs Using Parallelized Sparse Matrices

Markov Decision Processes (MDPs) provide important capabilities for facilitating the dynamic adaptation of hardware and software configurations to the environments in which they operate. However, the use of MDPs in embedded signal processing systems is limited because of the large computational demands for solving this class of system models. This paper presents Sparse Parallel Value Iteration (SPVI), a new algorithm for solving large MDPs on resource-constrained embedded systems that are equipped with mobile GPUs. SPVI leverages recent advances in parallel solving of MDPs and adds sparse linear algebra techniques to significantly outperform the state-of-the-art. The method and its application are described in detail, and demonstrated with case studies that are implemented on an NVIDIA Tegra K1 System On Chip (SoC). The experimental results show execution time improvements in the range of 65 % -78% for several applications. SPVI also lifts restrictions required by other MDP solver approaches, making it more widely compatible with large classes of optimization problems.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Research area: Computer engineering, Computing Sciences, University of Maryland, Department of Electrical and Computer Engineering, Georgia Institute of Technology
Contributors: Sapio, A., Bhattacharyya, S. S., Wolf, M.
Number of pages: 6
Pages: 13-18
Publication date: Dec 2018

Host publication information

Title of host publication: 2018 Conference on Design and Architectures for Signal and Image Processing, DASIP 2018
Publisher: IEEE COMPUTER SOCIETY PRESS
ISBN (Electronic): 9781538682371

Publication series

Name: Conference on Design and Architectures for Signal and Image Processing, DASIP
ISSN (Print): 2164-9766
ASJC Scopus subject areas: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Keywords: CUDA, GPU, Markov decision processes, MDP, Sparsity, Value iteration

Bibliographical note

jufoid=71852

Source: Scopus
Source ID: 85061388518

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Acoustic scene classification: An overview of dcase 2017 challenge entries

We present an overview of the challenge entries for the Acoustic Scene Classification task of DCASE 2017 Challenge. Being the most popular task of the challenge, acoustic scene classification entries provide a wide variety of approaches for comparison, with a wide performance gap from top to bottom. Analysis of the submissions confirms once more the popularity of deep-learning approaches and mel frequency representations. Statistical analysis indicates that the top ranked system performed significantly better than the others, and that combinations of top systems are capable of reaching close to perfect performance on the given data.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG
Contributors: Mesaros, A., Heittola, T., Virtanen, T.
Number of pages: 5
Pages: 411-415
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Acoustic scene classification, Audio classb ification, DCASE challenge
Source: Scopus
Source ID: 85057432865

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

An active learning method using clustering and committee-based sample selection for sound event classification

This paper proposes an active learning method to control a labeling process for efficient annotation of acoustic training material, which is used for training sound event classifiers. The proposed method performs K-medoids clustering over an initially unlabeled dataset, and medoids as local representatives, are presented to an annotator for manual annotation. The annotated label on a medoid propagates to other samples in its cluster for label prediction. After annotating the medoids, the annotation continues to the unexamined sounds with mismatched prediction results from two classifiers, a nearest-neighbor classifier and a model-based classifier, both trained with annotated data. The annotation on the segments with mismatched predictions are ordered by the distance to the nearest annotated sample, farthest first. The evaluation is made on a public environmental sound dataset. The labels obtained through a labeling process controlled by the proposed method are used to train a classifier, using supervised learning. Only 20% of the data needs to be manually annotated with the proposed method, to achieve the accuracy with all the data annotated. In addition, the proposed method clearly outperforms other active learning algorithms proposed for sound event classification through all the experiments, simulating varying fraction of data that is manually labeled.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG
Contributors: Shuyang, Z., Heittola, T., Virtanen, T.
Number of pages: 5
Pages: 116-120
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Active learning, Committee-based sample selection, K-medoids clustering, Sound event classification
Electronic versions: 
Source: Scopus
Source ID: 85057424847

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Deep neural network based speech separation optimizing an objective estimator of intelligibility for low latency applications

Mean square error (MSE) has been the preferred choice as loss function in the current deep neural network (DNN) based speech separation techniques. In this paper, we propose a new cost function with the aim of optimizing the extended short time objective intelligibility (ESTOI) measure. We focus on applications where low algorithmic latency (≤ 10 ms) is important. We use long short-term memory networks (LSTM) and evaluate our proposed approach on four sets of two-speaker mixtures from extended Danish hearing in noise (HINT) dataset. We show that the proposed loss function can offer improved or at par objective intelligibility (in terms of ESTOI) compared to an MSE optimized baseline while resulting in lower objective separation performance (in terms of the source to distortion ratio (SDR)). We then proceed to propose an approach where the network is first initialized with weights optimized for MSE criterion and then trained with the proposed ESTOI loss criterion. This approach mitigates some of the losses in objective separation performance while preserving the gains in objective intelligibility.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG, Oticon A/S
Contributors: Naithani, G., Nikunen, J., Bramslow, L., Virtanen, T.
Number of pages: 5
Pages: 386-390
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Deep neural networks, Low latency, Speech intelligibility, Speech separation
Source: Scopus
Source ID: 85057383332

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Multichannel NMF for source separation with ambisonic signals

This paper proposes a novel method for separation of sound sources with ambisonic signals using multichannel non-negative matrix factorization (MNMF) for source spectrogram estimation. We present a novel frequency-independent spatial covariance matrix (SCM) model for spherical harmonic (SH) domain signals which makes the MNMF parameter estimation framework computationally feasible up to 3rd order SH signals. The evaluation is done with simulated SH domain mixtures by measuring the separation performance using objective criteria and comparing the proposed method against SH domain beamforming. The proposed method improves average separation performance over beamforming with post-filtering when using 1st and 2nd order SH signals while at higher orders performance among all tested methods is similar.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Aalto University
Contributors: Nikunen, J., Politis, A.
Number of pages: 5
Pages: 251-255
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Ambisonics, Multichannel NMF, Source separation
Source: Scopus
Source ID: 85057417393

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Time-frequency masking strategies for single-channel low-latency speech enhancement using neural networks

This paper presents a low-latency neural network based speech enhancement system. Low-latency operation is critical for speech communication applications. The system uses the time-frequency (TF) masking approach to retain speech and remove the non-speech content from the observed signal. The ideal TF mask are obtained by supervised training of neural networks. As the main contribution different neural network models are experimentally compared to investigate computational complexity and speech enhancement performance. The proposed system is trained and tested on noisy speech data where signal-to-noise ratio (SNR) ranges from -5 dB to +5 dB and the results show significant reduction of non-speech content in the resulting signal while still meeting a low-latency operation criterion, which is here considered to be less than 20 ms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG, Huawei European Research Center
Contributors: Parviainen, M., Pertila, P., Virtanen, T., Grosche, P.
Number of pages: 5
Pages: 51-55
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Neural networks, Speech enhancement
Electronic versions: 
Source: Scopus
Source ID: 85057354159

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Using sequential information in polyphonic sound event detection

To detect the class, and start and end times of sound events in real world recordings is a challenging task. Current computer systems often show relatively high frame-wise accuracy but low event-wise accuracy. In this paper, we attempted to merge the gap by explicitly including sequential information to improve the performance of a state-of-the-art polyphonic sound event detection system. We propose to 1) use delayed predictions of event activities as additional input features that are fed back to the neural network; 2) build N-grams to model the co-occurrence probabilities of different events; 3) use se-quentialloss to train neural networks. Our experiments on a corpus of real world recordings show that the N-grams could smooth the spiky output of a state-of-the-art neural network system, and improve both the frame-wise and the event-wise metrics.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG
Contributors: Huang, G., Heittola, T., Virtanen, T.
Number of pages: 5
Pages: 291-295
Publication date: 2 Nov 2018

Host publication information

Title of host publication: 16th International Workshop on Acoustic Signal Enhancement, IWAENC 2018
Publisher: IEEE
ISBN (Electronic): 9781538681510
ASJC Scopus subject areas: Signal Processing, Acoustics and Ultrasonics
Keywords: Language modelling, Polyphonic sound event detection, Sequential information
Source: Scopus
Source ID: 85057368244

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Skew-t Filter and Smoother with Improved Covariance Matrix Approximation

Filtering and smoothing algorithms for linear discrete-time state-space models with skew-t-distributed measurement noise are proposed. The algorithms use a variational Bayes based posterior approximation with coupled location and skewness variables to reduce the error caused by the variational approximation. Although the variational update is done suboptimally using an expectation propagation algorithm, our simulations show that the proposed method gives a more accurate approximation of the posterior covariance matrix than an earlier proposed variational algorithm. Consequently, the novel filter and smoother outperform the earlier proposed robust filter and smoother and other existing low-complexity alternatives in accuracy and speed. We present both simulations and tests based on real-world navigation data, in particular GPS data in an urban area, to demonstrate the performance of the novel methods. Moreover, the extension of the proposed algorithms to cover the case where the distribution of the measurement noise is multivariate skew-t is outlined. Finally, the paper presents a study of theoretical performance bounds for the proposed algorithms.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Automation and Hydraulic Engineering, Linkiping University, Linki-ping University
Contributors: Nurminen, H., Ardeshiri, T., Piche, R., Gustafsson, F.
Pages: 5618-5633
Publication date: Nov 2018
Peer-reviewed: Yes
Early online date: 13 Aug 2018

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 66
Issue number: 21
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2018): CiteScore 6.28 SJR 1.477 SNIP 2.55
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Approximation algorithms, Biological system modeling, Covariance matrices, Cramer-Rao lower bound, expectation propagation, Gaussian distribution, Kalman filter, Noise measurement, robust filtering, RTS smoother, Signal processing algorithms, skew t, Smoothing methods, t-distribution, truncated normal distribution, variational Bayes
Electronic versions: 
Source: Scopus
Source ID: 85051630249

Research output: Contribution to journalArticleScientificpeer-review

Toward Efficient Execution of RVC-CAL Dataflow Programs on Multicore Platforms

The increasing number of cores in System on Chips (SoC) has introduced challenges in software parallelization. As an answer to this, the dataflow programming model offers a concurrent and reusability promoting approach for describing applications. In this work, a runtime for executing Dataflow Process Networks (DPN) on multicore platforms is proposed. The main difference between this work and existing methods is letting the operating system perform Central processing unit (CPU) load-balancing freely, instead of limiting thread migration between processing cores through CPU affinity. The proposed runtime is benchmarked on desktop and server multicore platforms using five different applications from video coding and telecommunication domains. The results show that the proposed method offers significant improvements over the state-of-art, in terms of performance and reliability.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, Research area: Computer engineering, Univ of Oulu
Contributors: Hautala, I., Boutellier, J., Nyländen, T., Silvén, O.
Number of pages: 11
Pages: 1507-1517
Publication date: Nov 2018
Peer-reviewed: Yes
Early online date: 9 Feb 2018

Publication information

Journal: Journal of Signal Processing Systems
Volume: 90
Issue number: 11
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2018): CiteScore 1 SJR 0.203 SNIP 0.497
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: Dataflow Process Networks, Multicore, Orcc, RVC-CAL
Source: Scopus
Source ID: 85041532591

Research output: Contribution to journalArticleScientificpeer-review

Data-driven study of synchronous population activity in generic spiking neuronal networks: How much do we capture using the minimal model for the considered phenomena?

General information

Publication status: Published
Organisations: Faculty of Biomedical Sciences and Engineering
Contributors: Acimovic, J., Teppola, H., Mäki-Marttunen, T. M., Linne, M.
Number of pages: 1
Pages: 68-69
Publication date: 29 Oct 2018
Peer-reviewed: Yes

Publication information

Journal: BMC Neuroscience
Volume: 19
Issue number: Suppl 2
ISSN (Print): 1471-2202
Ratings: 
  • Scopus rating (2018): CiteScore 2.56 SJR 1.017 SNIP 0.814
Original language: English
ASJC Scopus subject areas: Neuroscience (miscellaneous), Signal Processing
Keywords: spiking neuronal network, neuronal activity, spontaneous activity, data-driven computational model, model fitting, short-term synaptic dynamics

Research output: Contribution to journalMeeting AbstractScientificpeer-review

Deep Learning Case Study for Automatic Bird Identification

An automatic bird identification system is required for offshore wind farms in Finland. Indubitably, a radar is the obvious choice to detect flying birds, but external information is required for actual identification. We applied visual camera images as external data. The proposed system for automatic bird identification consists of a radar, a motorized video head and a single-lens reflex camera with a telephoto lens. A convolutional neural network trained with a deep learning algorithm is applied to the image classification. We also propose a data augmentation method in which images are rotated and converted in accordance with the desired color temperatures. The final identification is based on a fusion of parameters provided by the radar and the predictions of the image classifier. The sensitivity of this proposed system, on a dataset containing 9312 manually taken original images resulting in 2.44 × 106 augmented data set, is 0.9463 as an image classifier. The area under receiver operating characteristic curve for two key bird species is 0.9993 (the White-tailed Eagle) and 0.9496 (The Lesser Black-backed Gull), respectively. We proposed a novel system for automatic bird identification as a real world application. We demonstrated that our data augmentation method is suitable for image classification problem and it significantly increases the performance of the classifier.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Data-analytics and Optimization, Mathematics
Contributors: Niemi, J., Tanttu, J.
Number of pages: 15
Publication date: 29 Oct 2018
Peer-reviewed: Yes

Publication information

Journal: Applied Sciences (Switzerland)
Volume: 8
Issue number: 11
Article number: 2089
ISSN (Print): 2076-3417
Ratings: 
  • Scopus rating (2018): SJR 0.379 SNIP 0.985
Original language: English
ASJC Scopus subject areas: Artificial Intelligence, Signal Processing, Computer Vision and Pattern Recognition
Keywords: Machine learning, Deep learning, Convolutional neural networks, Classification, data augmentation, intelligent surveillance systems

Research output: Contribution to journalArticleScientificpeer-review

Eigen Posture Based Fall Risk Assessment System Using Kinect

Postural Instability (PI) is a major reason for fall in geriatric population as well as for people with diseases or disorders like Parkinson's, stroke etc. Conventional stability indicators like Berg Balance Scale (BBS) require clinical settings with skilled personnel's interventions to detect PI and finally classify the person into low, mid or high fall risk categories. Moreover these tests demand a number of functional tasks to be performed by the patient for proper assessment. In this paper a machine learning based approach is developed to determine fall risk with minimal human intervention using only Single Limb Stance exercise. The analysis is done based on the spatiotemporal dynamics of skeleton joint positions obtained from Kinect sensor. A novel posture modeling method has been applied for feature extraction along with some traditional time domain and metadata features to successfully predict the fall risk category. The proposed unobstrusive, affordable system is tested over 224 subjects and is able to achieve 75% mean accuracy on the geriatric and patient population.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Tata Consultancy Services India
Contributors: Tripathy, S. R., Chakravarty, K., Sinha, A.
Number of pages: 4
Pages: 1-4
Publication date: 26 Oct 2018

Host publication information

Title of host publication: 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
Volume: 2018-July
Publisher: IEEE
Article number: 8513263
ISBN (Electronic): 9781538636466
ASJC Scopus subject areas: Signal Processing, Biomedical Engineering, Computer Vision and Pattern Recognition, Health Informatics
Keywords: BBS, Eigenpose, EMD, Fall risk, Index Terms-Kinect
Source: Scopus
Source ID: 85056666030

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Identification of Parkinson's Disease Utilizing a Single Self-recorded 20-step Walking Test Acquired by Smartphone's Inertial Measurement Unit

Parkinson's disease (PD) is a degenerative and long-term disorder of the central nervous system, which often causes motor symptoms, e.g., tremor, rigidity, and slowness. Currently, the diagnosis of PD is based on patient history and clinical examination. Technology-derived decision support systems utilizing, for example, sensor-rich smartphones can facilitate more accurate PD diagnosis. These technologies could provide less obtrusive and more comfortable remote symptom monitoring. The recent studies showed that motor symptoms of PD can reliably be detected from data gathered via smartphones. The current study utilized an open-access dataset named 'mPower' to assess the feasibility of discriminating PD from non-PD by analyzing a single self-administered 20-step walking test. From this dataset, 1237 subjects (616 had PD) who were age and gender matched were selected and classified into PD and non-PD categories. Linear acceleration (ACC) and gyroscope (GYRO) were recorded by built-in sensors of smartphones. Walking bouts were extracted by thresholding signal magnitude area of the ACC signals. Features were computed from both ACC and GYRO signals and fed into a random forest classifier of size 128 trees. The classifier was evaluated deploying 100-fold cross-validation and provided an accumulated accuracy rate of 0.7 after 10k validations. The results show that PD and non-PD patients can be separated based on a single short-lasting self-administered walking test gathered by smartphones' built-in inertial measurement units.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Faculty of Biomedical Sciences and Engineering, Research group: Personal Health Informatics-PHI, Unit of Neurology, Satakunta Central Hospital
Contributors: Mehrang, S., Jauhiainen, M., Pietilä, J., Puustinen, J., Ruokolainen, J., Nieminen, H.
Number of pages: 4
Pages: 2913-2916
Publication date: 26 Oct 2018

Host publication information

Title of host publication: 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
Volume: 2018-July
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 8512921
ISBN (Electronic): 9781538636466
ASJC Scopus subject areas: Signal Processing, Biomedical Engineering, Computer Vision and Pattern Recognition, Health Informatics
Source: Scopus
Source ID: 85056600537

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

The Accuracy of Atrial Fibrillation Detection from Wrist Photoplethysmography. A Study on Post-Operative Patients

Atrial fibrillation (AF) is the most common type of cardiac arrhythmia. Although not life-threatening itself, AF significantly increases the risk of stroke and myocardial infarction. Current tools available for screening and monitoring of AF are inadequate and an unobtrusive alternative, suitable for long-term use, is needed. This paper evaluates an atrial fibrillation detection algorithm based on wrist photoplethysmographic (PPG) signals. 29 patients recovering from surgery in the post-anesthesia care unit were monitored. 15 patients had sinus rhythm (SR, 67.5± 10.7 years old, 7 female) and 14 patients had AF (74.8± 8.3 years old, 8 female) during the recordings. Inter-beat intervals (IBI) were estimated from PPG signals. As IBI estimation is highly sensitive to motion or other types of noise, acceleration signals and PPG waveforms were used to automatically detect and discard unreliable IBI. AF was detected from windows of 20 consecutive IBI with 98.45±6.89% sensitivity and 99.13±1.79% specificity for 76.34±19.54% of the time. For the remaining time, no decision was taken due to the lack of reliable IBI. The results show that wrist PPG is suitable for long term monitoring and AF screening. In addition, this technique provides a more comfortable alternative to ECG devices.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Faculty of Biomedical Sciences and Engineering, Research group: Sensor Technology and Biomeasurements (STB), PulseOn SA, Tampere University Hospital
Contributors: Tarniceriu, A., Harju, J., Yousefi, Z. R., Vehkaoja, A., Parak, J., Yli-Hankala, A., Korhonen, I.
Number of pages: 4
Pages: 4844-4847
Publication date: 26 Oct 2018

Host publication information

Title of host publication: 40th Annual International Conference of the IEEE Engineering in Medicine and Biology Society, EMBC 2018
Volume: 2018-July
Publisher: IEEE
Article number: 8513197
ISBN (Electronic): 9781538636466
ASJC Scopus subject areas: Signal Processing, Biomedical Engineering, Computer Vision and Pattern Recognition, Health Informatics

Bibliographical note

INT=tut-bmt, "Yousefi, Zeinab Rezaei"

Source: Scopus
Source ID: 85056672654

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Benchmark database for fine-grained image classification of benthic macroinvertebrates

Managing the water quality of freshwaters is a crucial task worldwide. One of the most used methods to biomonitor water quality is to sample benthic macroinvertebrate communities, in particular to examine the presence and proportion of certain species. This paper presents a benchmark database for automatic visual classification methods to evaluate their ability for distinguishing visually similar categories of aquatic macroinvertebrate taxa. We make publicly available a new database, containing 64 types of freshwater macroinvertebrates, ranging in number of images per category from 7 to 577. The database is divided into three datasets, varying in number of categories (64, 29, and 9 categories). Furthermore, in order to accomplish a baseline evaluation performance, we present the classification results of Convolutional Neural Networks (CNNs) that are widely used for deep learning tasks in large databases. Besides CNNs, we experimented with several other well-known classification methods using deep features extracted from the data.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Qatar University, University of Jyvaskyla, Finnish Environment Institute
Contributors: Raitoharju, J., Riabchenko, E., Ahmad, I., Iosifidis, A., Gabbouj, M., Kiranyaz, S., Tirronen, V., Ärje, J., Kärkkäinen, S., Meissner, K.
Number of pages: 11
Pages: 73-83
Publication date: 1 Oct 2018
Peer-reviewed: Yes

Publication information

Journal: Image and Vision Computing
Volume: 78
ISSN (Print): 0262-8856
Ratings: 
  • Scopus rating (2018): CiteScore 3.58 SJR 0.633 SNIP 1.541
Original language: English
ASJC Scopus subject areas: Signal Processing, Computer Vision and Pattern Recognition
Keywords: Benthic macroinvertebrates, Biomonitoring, Convolutional Neural Networks, Deep learning, Fine-grained classification
Source: Scopus
Source ID: 85052861257

Research output: Contribution to journalArticleScientificpeer-review

Guest Editorial Special Issue on Multimedia Big Data in Internet of Things

General information

Publication status: Published
MoE publication type: B1 Article in a scientific magazine
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG, Beijing University of Posts and Telecommunications, University of Technology Sydney, IBM Research
Contributors: Ma, H., Yu, S., Gabbouj, M., Mueller, P.
Number of pages: 3
Pages: 3405-3407
Publication date: Oct 2018
Peer-reviewed: No

Publication information

Journal: IEEE Internet of Things Journal
Volume: 5
Issue number: 5
Article number: 8534720
ISSN (Print): 2327-4662
Ratings: 
  • Scopus rating (2018): CiteScore 11.33 SJR 1.396 SNIP 3.874
Original language: English
ASJC Scopus subject areas: Signal Processing, Information Systems, Hardware and Architecture, Computer Science Applications, Computer Networks and Communications
Source: Scopus
Source ID: 85056768996

Research output: Contribution to journalEditorialScientific

LoTTA: Energy-Efficient Processor for Always-on Applications

Various use cases in the era of Internet-of-Things (IoT) demand processor devices to have low energy consumption in order to maximize the battery life. In addition to energy constraints, there is often a need to both swiftly execute control-oriented code to provide low reaction times and to occasionally perform real time signal processing tasks efficiently. As a response to these requirements, we propose LoTTA, an extremely energy-efficient exposed datapath core. Its transport-triggered programming model helps in lowering the execution latency via low cost data forwarding. Control efficiency is achieved by an optimized control unit with zero delay slot branches and predicated execution. An instruction register file is included for frequently executed program hot spots to reduce the instruction stream energy consumption. These features allow the processor to execute CHStone and EEMBC CoreMark benchmarks on average with 19% fewer cycles compared to a 6-stage LM32, a traditional RISC core with similar datapath resources. The core consumes 53% less energy on average compared to the RISC core. When including the instruction stream overheads, in the best case, LoTTA saves 79% energy, and on average 40%.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research area: Computer engineering
Contributors: Multanen, J., Kultala, H., Jääskeläinen, P., Viitanen, T., Tervo, A., Takala, J.
Publication date: Oct 2018

Host publication information

Title of host publication: 2018 IEEE International Workshop on Signal Processing Systems (SiPS)
Publisher: IEEE
ISBN (Print): 978-1-5386-6319-6
ISBN (Electronic): 978-1-5386-6318-9
ASJC Scopus subject areas: Computer Science(all), Hardware and Architecture, Signal Processing
Keywords: Energy efficiency, low power, Computer architecture, IoT, always-on, Processor

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Data-driven study of synchronous popula on ac vity in generic spiking neuronal networks: How much do we capture using the minimal model for the considered phenomena?

General information

Publication status: Published
Organisations: Faculty of Biomedical Sciences and Engineering, Research group: Computational Neuro Science-CNS, Research group: Computational Neuro Science-CNS
Contributors: Acimovic, J., Teppola, H., Mäki-Marttunen, T. M., Linne, M.
Publication date: 20 Sep 2018
Peer-reviewed: Unknown
Event: Paper presented at Brain and Mind Symposium 2018, Helsinki, Finland.
ASJC Scopus subject areas: Cellular and Molecular Neuroscience, Neuroscience (miscellaneous), Signal Processing

Research output: Other conference contributionPaper, poster or abstractScientific

A joint target localization and classification framework for sensor networks

In this paper, we propose a joint framework for target localization and classification using a single generalized model for non-imaging based multi-modal sensor data. For target localization, we exploit both sensor data and estimated dynamics within a local neighborhood. We validate the capabilities of our framework by using a multi-modal dataset, which includes ground truth GPS information (e.g., time and position) and data from co-located seismic and acoustic sensors. Experimental results show that our framework achieves better classification accuracy compared to recent fusion algorithms using temporal accumulation and achieves more accurate target localizations than multilateration.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research area: Computer engineering, U.S. Army Research Laboratory, University of Maryland
Contributors: Lee, K., Riggan, B. S., Bhattacharyya, S. S.
Number of pages: 5
Pages: 3076-3080
Publication date: 10 Sep 2018

Host publication information

Title of host publication: 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Volume: 2018-April
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 8462641
ISBN (Print): 9781538646588
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Classification, Localization, Sensor fusion, Sensor networks, Tracking

Bibliographical note

JUFOID=57409

Source: Scopus
Source ID: 85054211699

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Estimation of time-varying room impulse responses of multiple sound sources from observed mixture and isolated source signals

This paper proposes a method for online estimation of time-varying room impulse responses (RIR) between multiple isolated sound sources and a far-field mixture. The algorithm is formulated as adaptive convolutive filtering in short-time Fourier transform (STFT) domain. We use the recursive least squares (RLS) algorithm for estimating the filter parameters due to its fast convergence rate, which is required for modeling rapidly changing RIRs of moving sound sources. The proposed method allows separation of reverberated sources from the far-field mixture given that their close-field signals are available. The evaluation is based on measuring unmixing performance (removal of reverberated source) using objective separation criteria calculated between the ground truth recording of the preserved sources and the unmixing result obtained with the proposed algorithm. We compare online and offline formulations for the RIR estimation and also provide evaluation with blind source separation algorithm only operating on the mixture signal.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG
Contributors: Nikunen, J., Virtanen, T.
Number of pages: 5
Pages: 421-425
Publication date: 10 Sep 2018

Host publication information

Title of host publication: 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Volume: 2018-April
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 8462535
ISBN (Print): 9781538646588

Publication series

Name:  Proceedings of the IEEE International Conference on Acoustics, Speech, and Signal Processing
ISSN (Electronic): 2379-190X
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Adaptive filtering, Informed source separation, Online room impulse response estimation, Source unmixing

Bibliographical note

jufoid=57409

Source: Scopus
Source ID: 85054234348

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Low-energy graph fourier basis functions span salient objects

There is an emerging interest aiming at defining principles for signals on general graphs, which are analogous to the basic principles in traditional signal processing. One example is the Graph Fourier Transform which aims at decomposing a graph signal into its components based on a set of basis functions with corresponding graph frequencies. It has been observed that most of the important information of a graph signal is contained inside the low frequency band, which leads to several applications such as denoising, compression, etc. In this paper, we show that the low frequency basis functions span the salient regions in an image, which can also be considered as important regions. Motivated by this, we present a novel simple and unsupervised method to utilize a number of low-energy basis functions and show that it improves the performance of seven state-of-the-art salient object detection methods in five datasets under four different evaluation criteria, with only minor exceptions.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG
Contributors: Malik, J., Aytekin, C., Gabbouj, M.
Number of pages: 5
Pages: 1548-1552
Publication date: 10 Sep 2018

Host publication information

Title of host publication: 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Volume: 2018-April
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 8462672
ISBN (Print): 9781538646588
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Graph fourier transform, Graph signal processing, Salient object detection

Bibliographical note

jufoid=57409

Source: Scopus
Source ID: 85054244934

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Statistical Evaluation of Visual Quality Metrics for Image Denoising

This paper studies the problem of full reference visual quality assessment of denoised images with a special emphasis on images with low contrast and noise-like texture. Denoising of such images together with noise removal often results in image details loss or smoothing. A new test image database, FLT, containing 75 noise-free 'reference' images and 300 filtered ('distorted') images is developed. Each reference image, corrupted by an additive white Gaussian noise, is denoised by the BM3D filter with four different values of threshold parameter (four levels of noise suppression). After carrying out a perceptual quality assessment of distorted images, the mean opinion scores (MOS) are obtained and compared with the values of known full reference quality metrics. As a result, the Spearman Rank Order Correlation Coefficient (SROCC) between PSNR values and MOS has a value close to zero, and SROCC between values of known full-reference image visual quality metrics and MOS does not exceed 0.82 (which is reached by a new visual quality metric proposed in this paper). The FLT dataset is more complex than earlier datasets used for assessment of visual quality for image denoising. Thus, it can be effectively used to design new image visual quality metrics for image denoising.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Computational Imaging-CI, National Aerospace University
Contributors: Egiazarian, K., Ponomarenko, M., Lukin, V., Ieremeiev, O.
Number of pages: 5
Pages: 6752-6756
Publication date: 10 Sep 2018

Host publication information

Title of host publication: 2018 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2018 - Proceedings
Volume: 2018-April
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 8462294
ISBN (Print): 9781538646588
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: BM3D, Full-reference metrics, Image denoising, Image visual quality assessment

Bibliographical note

EXT="Lukin, Vladimir"
JUFOID=57409

Source: Scopus
Source ID: 85054084165

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

2D Video Coding of Volumetric Video Data

Due to the increased popularity of augmented and virtual reality experiences, the interest in representing the real world in an immersive fashion has never been higher. Distributing such representations enables users all over the world to freely navigate in never seen before media experiences. Unfortunately, such representations require a large amount of data, not feasible for transmission on today's networks. Thus, efficient compression technologies are in high demand. This paper proposes an approach to compress 3D video data utilizing 2D video coding technology. The proposed solution was developed to address the needs of 'tele-immersive' applications, such as virtual (VR), augmented (AR) or mixed (MR) reality with Six Degrees of Freedom (6DoF) capabilities. Volumetric video data is projected on 2D image planes and compressed using standard 2D video coding solutions. A key benefit of this approach is its compatibility with readily available 2D video coding infrastructure. Furthermore, objective and subjective evaluation shows significant improvement in coding efficiency over reference technology.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Nokia Technologies
Contributors: Schwarz, S., Hannuksela, M. M., Fakour-Sevom, V., Sheikhi-Pour, N.
Number of pages: 5
Pages: 61-65
Publication date: 5 Sep 2018

Host publication information

Title of host publication: 2018 Picture Coding Symposium, PCS 2018 - Proceedings
Publisher: IEEE
Article number: 8456265
ISBN (Print): 9781538641606
ASJC Scopus subject areas: Signal Processing, Media Technology

Bibliographical note

INT=sgn,"Sheikhi-Pour, Nahid"

Source: Scopus
Source ID: 85053915056

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Inertial Odometry on Handheld Smartphones

Building a complete inertial navigation system using the limited quality data provided by current smartphones has been regarded challenging, if not impossible. This paper shows that by careful crafting and accounting for the weak information in the sensor samples, smartphones are capable of pure inertial navigation. We present a probabilistic approach for orientation and use-case free inertial odometry, which is based on double-integrating rotated accelerations. The strength of the model is in learning additive and multiplicative IMU biases online. We are able to track the phone position, velocity, and pose in realtime and in a computationally lightweight fashion by solving the inference with an extended Kalman filter. The information fusion is completed with zero-velocity updates (if the phone remains stationary), altitude correction from barometric pressure readings (if available), and pseudo-updates constraining the momentary speed. We demonstrate our approach using an iPad and iPhone in several indoor dead-reckoning applications and in a measurement tool setup.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Artificial Intelligence and Vision - AIV, Aalto University
Contributors: Solin, A., Cortes, S., Rahtu, E., Kannala, J.
Number of pages: 8
Pages: 1361-1368
Publication date: 5 Sep 2018

Host publication information

Title of host publication: 2018 21st International Conference on Information Fusion, FUSION 2018
Publisher: IEEE
Article number: 8455482
ISBN (Print): 9780996452762
ASJC Scopus subject areas: Computer Vision and Pattern Recognition, Signal Processing, Statistics, Probability and Uncertainty, Instrumentation
Source: Scopus
Source ID: 85054102788

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

User Positioning in mmW 5G Networks Using Beam-RSRP Measurements and Kalman Filtering

In this paper, we exploit the 3D-beamforming features of multiantenna equipment employed in fifth generation (5G) networks, operating in the millimeter wave (mmW) band, for accurate positioning and tracking of users. We consider sequential estimation of users' positions, and propose a two-stage extended Kalman filter (EKF) that is based on reference signal received power (RSRP) measurements. In particular, beamformed downlink (DL) reference signals (RSs) are transmitted by multiple base stations (BSs) and measured by user equipments (UEs) employing receive beamforming. The so-obtained beam-RSRP (BRSRP) measurements are reported to the BSs where the corresponding directions of departure (DoDs) are sequentially estimated by a novel EKF. Such angle estimates from multiple BSs are subsequently fused on a central entity into 3D position estimates of UEs by means of another (second-stage) EKF. The proposed positioning scheme is scalable since the computational burden is shared among different network entities, namely transmission/reception points (TRPs) and 5G-NR Node B (gNB), and may be accomplished with the signalling currently specified for 5G. We assess the performance of the proposed algorithm on a realistic outdoor 5G deployment with a detailed ray tracing propagation model based on the METIS Madrid map. Numerical results with a system operating at 39 GHz show that sub-meter 3D positioning accuracy is achievable in future mmW 5G networks.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Huawei Technologies Oy (Finland) Co., Ltd
Contributors: Rastorgueva-Foi, E., Costa, M., Koivisto, M., Leppänen, K., Valkama, M.
Number of pages: 7
Pages: 1150-1156
Publication date: 5 Sep 2018

Host publication information

Title of host publication: 2018 21st International Conference on Information Fusion, FUSION 2018
Publisher: IEEE
Article number: 8455289
ISBN (Print): 978-1-5386-4330-3
ISBN (Electronic): 978-0-9964527-6-2
ASJC Scopus subject areas: Computer Vision and Pattern Recognition, Signal Processing, Statistics, Probability and Uncertainty, Instrumentation
Keywords: 5G networks, beamforming, direction-of-departure, extended Kalman filter, line-of-sight, localization, location-awareness, positioning, RSRP, tracking
Source: Scopus
Source ID: 85054063725

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Anisotropic Spatiotemporal Regularization in Compressive Video Recovery by Adaptively Modeling the Residual Errors as Correlated Noise

Many approaches to compressive video recovery proceed iteratively, treating the difference between the previous estimate and the ideal video as residual noise to be filtered. We go beyond the common white-noise modeling by adaptively modeling the residual as stationary spatiotemporally correlated noise. This adaptive noise model is updated at each iteration and is highly anisotropic in space and time; we leverage it with respect to the transform spectra of a motion-compensated video denoiser. Experimental results demonstrate that our proposed adaptive correlated noise model outperforms state-of-the-art methods both quantitatively and qualitatively.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Signal and Image Restoration-RST
Contributors: Eslahi, N., Foi, A.
Publication date: 27 Aug 2018

Host publication information

Title of host publication: 2018 IEEE 13th Image, Video, and Multidimensional Signal Processing Workshop, IVMSP 2018 - Proceedings
Publisher: IEEE
Article number: 8448455
ISBN (Print): 9781538609514
ASJC Scopus subject areas: Signal Processing, Media Technology
Electronic versions: 
Source: Scopus
Source ID: 85053868569

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Multi-source localization using a DOA Kernel based spatial covariance model and complex nonnegative matrix factorization

This paper presents an algorithm for multiple source localization using a beamforming-inspired spatial covariance model (SCM) and complex non-negative matrix factorization (CNMF). In this work, we assume that the source signals are known in advance whereas the mixing filter is modeled by the weighted sum of direction of arrival (DOA) kernels which encode the phase and the amplitude differences between microphones for every possible source direction. The direction of arrival (i.e. azimuth and elevation) for each source is estimated using CNMF. The proposed system is evaluated for DOA estimation task using two datasets covering a large number of configurations (number of channels, number of simultaneous sources, reverberation time, microphones spacing, source types and angular positions of the sources). Finally, a comparison to other state-of-the-art methods is performed, showing the robustness of the proposed method.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Audio research group - ARG, Universidad de Jaen
Contributors: Carabias-Orti, J. J., Cabanas-Molero, P., Vera-Candeas, P., Nikunen, J.
Number of pages: 5
Pages: 440-444
Publication date: 27 Aug 2018

Host publication information

Title of host publication: 2018 IEEE 10th Sensor Array and Multichannel Signal Processing Workshop, SAM 2018
Publisher: IEEE
Article number: 8448664
ISBN (Print): 9781538647523

Publication series

Name: Proceedings of the IEEE Sensor Array and Multichannel Signal Processing Workshop
ISSN (Print): 1551-2282
ISSN (Electronic): 2151-870X
ASJC Scopus subject areas: Signal Processing, Control and Systems Engineering, Electrical and Electronic Engineering
Keywords: Direction of arrival, Nonnegative matrix factorization, Source localization, Spatial covariance matrix, Time difference of arrival

Bibliographical note

EXT="Carabias-Orti, J. J."

Source: Scopus
Source ID: 85053623218

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Data Clustering Based on Community Structure in Mutual k-Nearest Neighbor Graph

Data clustering is a fundamental machine learning problem. Community structure is common in social and biological networks. In this article we propose a novel data clustering algorithm that uses this phenomenon in mutual k - nearest neighbor (MKNN) graph constructed from the input dataset. We use the authentic scores-a metric that measures the strength of an edge in a social network graph-to rank all the edges in the MKNN graph. By removing the edges gradually in the order of their authentic scores, we collapse the MKNN graph into components to find the clusters. The proposed method has two major advantages comparing to other popular data clustering algorithms. First, it is robust to the noise in the data. Second, it finds clusters of arbitrary shape. We evaluated our algorithm on synthetic noisy datasets, synthetic 2D datasets and real-world image datasets. Results on the noisy datasets show that the proposed algorithm clearly outperforms the competing algorithms in terms of Normalized Mutual Information (NMI) scores. The proposed algorithm is the only one that does not fail on any data in the the synthetic 2D dataset, which are specifically designed to show the limitations of the clustering algorithms. When testing on the real-world image datasets, the best NMI scores achieved by the proposed algorithm is more than any other competing algorithm. The proposed algorithm has computational complexity of O(k3n+kn\log (kn)) and space complexity of O(kn), which is better than or equivalent to the most popular clustering algorithms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG, Qatar University
Contributors: Zhang, H., Kiranyaz, S., Gabbouj, M.
Number of pages: 7
Pages: 262-268
Publication date: 20 Aug 2018

Host publication information

Title of host publication: 2018 41st International Conference on Telecommunications and Signal Processing, TSP 2018
Publisher: IEEE
Article number: 8441226
ISBN (Print): 9781538646953
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing
Keywords: authentic score, data clustering, graph

Bibliographical note

EXT="Kiranyaz, Serkan"

Source: Scopus
Source ID: 85053518955

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Performance Evaluation of UAV-Assisted mmWave Operation in Mobility-Enabled Urban Deployments

To meet the capacity demand of future cellular networks, which is expected to increase substantially by 2020, using mm Wave band has been suggested. Although this higher frequency band has larger amounts of spectrum and will lead to much higher capacity in 5G networks, it has some propagation constraints such as limited coverage and sensitivity to Line of Sight (LoS) blockage. A possible solution to these issues could be small cell densification combined with utilizing access points carried by Unmanned Aerial Vehicles (UAVs). This paper presents a performance evaluation of the UAV-assisted mmWave network in urban environments using ns-3 simulations.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Emerging Technologies for Nano-Bio-Info-Cogno
Contributors: Khosravi, Z., Gerasimenko, M., Andreev, S., Koucheryavy, Y.
Number of pages: 4
Pages: 150-153
Publication date: 20 Aug 2018

Host publication information

Title of host publication: 2018 41st International Conference on Telecommunications and Signal Processing, TSP 2018
Publisher: IEEE
Article number: 8441321
ISBN (Print): 9781538646953
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing
Keywords: 5G, drone cell, mm Wave, ns3

Bibliographical note

INT=elt,"Khosravi, Zeinab"

Source: Scopus
Source ID: 85053557557

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Improvement of GPS and BeiDou extended orbit predictions with CNNs

This paper presents a method for improving the accuracy of extended GNSS satellite orbit predictions with convolutional neural networks (CNN). Satellite orbit predictions are used in self-assisted GNSS to reduce the Time to First Fix of a satellite positioning device. We describe the models we use to predict the satellite orbit and present the improvement method that uses CNN. The CNN estimates future prediction errors of our model and these estimates are used to correct our orbit predictions. We also describe how the neural network can be implemented into our prediction algorithm. In tests with GPS and BeiDou data, the method significantly improves orbit prediction accuracy. For example, the 68% error quantile of 7 day orbit prediction errors of GPS satellites was reduced by 45% on average.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Automation and Hydraulic Engineering, Research group: Automation and Systems Theory, Mathematics, Research group: Positioning
Contributors: Pihlajasalo, J., Leppäkoski, H., Ali-Löytty, S., Piché, R.
Number of pages: 6
Pages: 54-59
Publication date: 10 Aug 2018

Host publication information

Title of host publication: 26th European Navigation Conference, ENC 2018 : Gothenburg, Sweden, 14-17 May, 2018
Publisher: IEEE
Article number: 8433244
ISBN (Print): 9781538649626
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Control and Optimization
Electronic versions: 

Bibliographical note

INT=mat,"Jaakko Pihlajasalo"

Source: Scopus
Source ID: 85052494723

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Nonlocality-Reinforced Convolutional Neural Networks for Image Denoising

We introduce a paradigm for nonlocal sparsity reinforced deep convolutional neural network denoising. It is a combination of a local multiscale denoising by a convolutional neural network (CNN) based denoiser and a nonlocal denoising based on a nonlocal filter (NLF), exploiting the mutual similarities between groups of patches. CNN models are leveraged with noise levels that progressively decrease at every iteration of our framework, while their output is regularized by a nonlocal prior implicit within the NLF. Unlike complicated neural networks that embed the nonlocality prior within the layers of the network, our framework is modular, and it uses standard pretrained CNNs together with standard nonlocal filters. An instance of the proposed framework, called NN3D, is evaluated over large grayscale image datasets showing state-of-the-art performance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Noiseless Imaging Ltd
Contributors: Cruz, C., Foi, A., Katkovnik, V., Egiazarian, K.
Number of pages: 5
Pages: 1216-1220
Publication date: 1 Aug 2018
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 25
Issue number: 8
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2018): CiteScore 4.37 SJR 0.785 SNIP 1.88
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering, Applied Mathematics
Keywords: BM3D, convolutional neural network, image denoising, nonlocal filters
Source: Scopus
Source ID: 85049097929

Research output: Contribution to journalArticleScientificpeer-review

Summarization of User-Generated Sports Video by Using Deep Action Recognition Features

Automatically generating a summary of sports video poses the challenge of detecting interesting moments, or highlights, of a game. Traditional sports video summarization methods leverage editing conventions of broadcast sports video that facilitate the extraction of high-level semantics. However, user-generated videos are not edited, and thus traditional methods are not suitable to generate a summary. In order to solve this problem, this work proposes a novel video summarization method that uses players' actions as a cue to determine the highlights of the original video. A deep neural network-based approach is used to extract two types of action-related features and to classify video segments into interesting or uninteresting parts. The proposed method can be applied to any sports in which games consist of a succession of actions. Especially, this work considers the case of Kendo (Japanese fencing) as an example of a sport to evaluate the proposed method. The method is trained using Kendo videos with ground truth labels that indicate the video highlights. The labels are provided by annotators possessing different experience with respect to Kendo to demonstrate how the proposed method adapts to different needs. The performance of the proposed method is compared with several combinations of different features, and the results show that it outperforms previous summarization methods.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, University of Tokyo, Osaka University, Graduate School of Information Science, Univ of Oulu
Contributors: Tejero-de-Pablos, A., Nakashima, Y., Sato, T., Yokoya, N., Linna, M., Rahtu, E.
Pages: 2000-2011
Publication date: Aug 2018
Peer-reviewed: Yes
Early online date: 15 Jan 2018

Publication information

Journal: IEEE Transactions on Multimedia
Volume: 20
Issue number: 8
ISSN (Print): 1520-9210
Ratings: 
  • Scopus rating (2018): CiteScore 7.41 SJR 1.222 SNIP 2.951
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Computer Science Applications, Electrical and Electronic Engineering
Keywords: 3D convolutional neural networks, action recognition, Cameras, deep learning, Feature extraction, Games, Hidden Markov models, long short-term memory, Semantics, Sports video summarization, Three-dimensional displays, user-generated video
Source: Scopus
Source ID: 85041689127

Research output: Contribution to journalArticleScientificpeer-review

Adjusting university education with workspace training and self-education

Close to 100% employment of students and easy access to abundance of information on Internet has essentially changed student's learning practices and their earlier knowledge background, especially on rapidly progressing field of Software Engineering. On workplace they have to use technologies, which are used in practice of their employing enterprise, but often do not understand the scientific and/or technological principles on which these technologies are based. They seek explanations on Internet, but information on Internet is often low quality, one-sided and presented with business targets on mind - to get more users to technologies developed and sold by a business enterprise. Thus university has to explain basic principles of technologies what students already know and have used and correct some popular beliefs, which are supported by software vendors and based their business interests. Non-formal sources of knowledge - workplace training and Internet - do not reduce teacher's task, but force teachers constantly study all new which appears in this field, thus increase teachers workload. Students increasing use of non-formal sources of knowledge imply need for flipping the process - instead of teaching students are set to learn from provided detailed tutorials. Use of Internet and work has made self-study, seeking information from Internet sources very customary for current students, thus such flipping worked very well in a game programming course provided by the first author.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research group: Software Engineering and Intelligent Systems, Tallinn University of Technology, University of Lapland
Contributors: Henno, J., Jaakkola, H., Makela, J.
Number of pages: 8
Pages: 701-708
Publication date: 28 Jun 2018

Host publication information

Title of host publication: 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics, MIPRO 2018
Publisher: IEEE
ISBN (Electronic): 9789532330977
ASJC Scopus subject areas: Electrical and Electronic Engineering, Hardware and Architecture, Signal Processing, Computer Networks and Communications
Keywords: CSS, game programming, HTML5, Internet, IT, JavaScript, SE, university education, working students
Source: Scopus
Source ID: 85050228879

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Role of the user in information systems development

We have focused our paper on the aspects important in adapting an Information System (IS) to the user's cultural background. We are interested both in the factors related to IS development and in the use of IS. Increasingly, ISs are being developed and used in a global context. We have perceived differences in expectations of functionalities, architecture, structural properties, information search practices, web-based system properties, and user interfaces. One conclusion would be that a high quality IS reflects user behavior in its use context. In that case, the system has to model its user one way or another. Until now, the topic has been handled without meaningful effort to model user behavior. Current publications cover a wide variety of rules on how to take into account cultural differences in the IS context. In this paper, our aim is to study the current state-of-the-art of user modeling - modeling the human being as an IS user. We start with general aspects related to the role of the user in IS development and alternatives to adaptable systems. The findings are applicable in the educational context as well. More and more, the use of computers and ISs is becoming an essential part of studies: the use of MOOCs (Massively Open Online Courses) as a part or replacement for traditional face-to-face classes; flipped learning methodology emphasizing the significance of self-learning; and blended learning, including quite often computerized study content. Our focus is on the global context, in which students represent different cultures and the IS is globally available.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research group: Software Engineering and Intelligent Systems, Computer Science Institute, Tallinn University of Technology, University of Lapland
Contributors: Jaakkola, H., Thalheim, B., Henno, J., Mäkelä, J., Keto, H.
Number of pages: 8
Pages: 625-632
Publication date: 28 Jun 2018

Host publication information

Title of host publication: 2018 41st International Convention on Information and Communication Technology, Electronics and Microelectronics, MIPRO 2018
Publisher: IEEE
ISBN (Electronic): 9789532330977
ASJC Scopus subject areas: Electrical and Electronic Engineering, Hardware and Architecture, Signal Processing, Computer Networks and Communications
Keywords: adaption, context, human behavior modeling, information system, information systems development, requirements engineering, user, user adaptable, user modeling
Electronic versions: 
Source: Scopus
Source ID: 85050202384

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Full-duplex radio technology for simultaneously detecting and preventing improvised explosive device activation

Originating from civilian/commercial wireless networks, the progressive concept of same frequency simultaneous transmission and reception (SF-STAR), a.k.a. in-band full-duplex operation, has high potential also at the future battlefield. The prospects of a military full-duplex radio (MFDR) are not limited to enhancing the spectral efficiency of tactical communications, which would already be a significant advancement considering the universal congestion of electromagnetic spectrum. Perhaps even more importantly, armed forces could gain a major technical advantage by employing multifunction MFDRs that are capable of jointly conducting signals intelligence, electronic warfare, and tactical communications owing to their SF-STAR capability. This study focuses on one specific promising application, where a radio transceiver performs spectrum monitoring and signal surveillance for potential hostile transmissions when simultaneously performing an electronic attack against opposing forces' receivers at the same frequency band. In particular, we demonstrate by experiments in a laboratory environment that the MFDR technology can be successfully used for detecting an attempt to control remotely an improvised explosive device while also preventing its activation by transmitting a jamming signal.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering
Contributors: Riihonen, T., Korpi, D., Turunen, M., Valkama, M.
Number of pages: 4
Pages: 1-4
Publication date: 27 Jun 2018

Host publication information

Title of host publication: 2018 International Conference on Military Communications and Information Systems, ICMCIS 2018
Publisher: IEEE
ISBN (Electronic): 9781538645598

Bibliographical note

INT=elt,"Turunen, Matias"

Source: Scopus
Source ID: 85050255592

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Stabilization to trajectories for parabolic equations

Both internal and boundary feedback exponential stabilization to trajectories for semilinear parabolic equations in a given bounded domain are addressed. The values of the controls are linear combinations of a finite number of actuators which are supported in a small region. A condition on the family of actuators is given which guarantees the local stabilizability of the control system. It is shown that a linearization-based Riccati feedback stabilizing controller can be constructed. The results of numerical simulations are presented and discussed.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Mathematics, Johann Radon Institute for Computational and Applied Mathematics
Contributors: Phan, D., Rodrigues, S. S.
Publication date: 1 Jun 2018
Peer-reviewed: Yes

Publication information

Journal: Mathematics of Control, Signals, and Systems
Volume: 30
Issue number: 2
Article number: 11
ISSN (Print): 0932-4194
Ratings: 
  • Scopus rating (2018): CiteScore 0.95 SJR 0.386 SNIP 0.701
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Signal Processing, Control and Optimization, Applied Mathematics
Keywords: Feedback stabilization to trajectories, Semilinear parabolic equations
Source: Scopus
Source ID: 85050079985

Research output: Contribution to journalArticleScientificpeer-review

Delay-Accuracy Trade-off in Opportunistic Time-of-Arrival Localization

While designing a positioning network, the localization performance is traditionally the main concern. However, collection of measurements together with channel access methods require a nonzero time, causing a delay experienced by network nodes. This fact is usually neglected in positioning-related literature. In terms of the delay-accuracy trade-off, broadcast schemes have an advantage over unicast, provided nodes can be properly synchronized. In this letter, we analyze the delay-accuracy trade-off, for localization schemes in which the position estimates are obtained based on broadcasted ranging signals. We find that for dense networks, the trade-off is the same for cooperative and noncooperative networks, and cannot exceed a certain threshold value.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Research group: System-on-Chip for GNSS, Wireless Communications and Cyber-Physical Embedded Computing, Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Research group: System-on-Chip for GNSS, Wireless Communications and Cyber-Physical Embedded Computing, Chalmers University of Technology
Contributors: Daniel, O., Wymeersch, H., Nurmi, J.
Pages: 763-767
Publication date: Jun 2018
Peer-reviewed: Yes
Early online date: 14 Apr 2018

Publication information

Journal: IEEE Signal Processing Letters
Volume: 25
Issue number: 6
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2018): CiteScore 4.37 SJR 0.785 SNIP 1.88
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering

Research output: Contribution to journalArticleScientificpeer-review

Digital Predistortion for Hybrid MIMO Transmitters

This article investigates digital predistortion (DPD) linearization of hybrid beamforming large-scale antenna transmitters. We propose a novel DPD processing and learning technique for an antenna sub-array, which utilizes a combined signal of the individual power amplifier (PA) outputs in conjunction with a decorrelation-based learning rule. In effect, the proposed approach results in minimizing the nonlinear distortions in the direction of the intended receiver. This feature is highly desirable, since emissions in other directions are naturally weak due to beamforming. The proposed parameter learning technique requires only a single observation receiver, and therefore supports simple hardware implementation. It is also shown to clearly outperform the current state-of-the-art technique which utilizes only a single PA for learning. Analysis of the feedback network amplitude and phase imbalances reveals that the technique is robust even to high levels of such imbalances. Finally, we also show that the array system out-of-band emissions are well-behaving in all spatial directions, and essentially below those of the corresponding single-antenna transmitter, due to the combined effects of the DPD and beamforming.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Lund University
Contributors: Abdelaziz, M., Anttila, L., Brihuega, A., Tufvesson, F., Valkama, M.
Pages: 445-454
Publication date: Jun 2018
Peer-reviewed: Yes
Early online date: 7 Apr 2018

Publication information

Journal: IEEE Journal on Selected Topics in Signal Processing
Volume: 12
Issue number: 3
ISSN (Print): 1932-4553
Ratings: 
  • Scopus rating (2018): CiteScore 7.1 SJR 1.404 SNIP 2.516
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: 5G, digital predistortion, hybrid beamforming, large-array transmitters, out-of-band emissions, power amplifiers
Source: Scopus
Source ID: 85045212003

Research output: Contribution to journalArticleScientificpeer-review

Sample-based regularization for support vector machine classification

In this paper, we propose a new regularization scheme for the well-known Support Vector Machine (SVM) classifier that operates on the training sample level. The proposed approach is motivated by the fact that Maximum Margin-based classification defines decision functions as a linear combination of the selected training data and, thus, the variations on training sample selection directly affect generalization performance. We show that the exploitation of the proposed regularization scheme is well motivated and intuitive. Experimental results show that the proposed regularization scheme outperforms standard SVM in human action recognition tasks as well as classical recognition problems.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing
Contributors: Tran, D. T., Waris, M. A., Gabbouj, M., Iosifidis, A.
Number of pages: 6
Pages: 1-6
Publication date: 8 Mar 2018

Host publication information

Title of host publication: Proceedings of the 7th International Conference on Image Processing Theory, Tools and Applications, IPTA 2017
Publisher: IEEE
ISBN (Electronic): 9781538618417
ASJC Scopus subject areas: Signal Processing, Radiology Nuclear Medicine and imaging
Keywords: Dropout, kernel methods, Regularization, Support Vector Machine

Bibliographical note

INT=sgn,"Tran, Dat Thanh"

Source: Scopus
Source ID: 85050655029

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Subaperture image segmentation for lossless compression

The paper proposes an image segmentation method for lossless compression of plenoptic images. Each light-field image captured by the plenoptic camera is processed to obtain a stack of subaperture images. Each subaperture image is encoded by using a gradient-base detector which classifies the image edges and designs refined contexts for an improved prediction and segmentation. The paper's main contribution is a new segmentation method which generates a preliminary segmentation, either by scaling the intensity differences or by using a quantum cut based algorithm, and merges it with an edge ranking-based segmentation. The results show around 2% improved performance compared to the state-of-the-art for a dataset of 118 plenoptic images.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, University of Electronic Science and Technology of China
Contributors: Schiopu, I., Gabbouj, M., Iosifidis, A., Zeng, B., Liu, S.
Number of pages: 6
Pages: 1-6
Publication date: 8 Mar 2018

Host publication information

Title of host publication: Proceedings of the 7th International Conference on Image Processing Theory, Tools and Applications, IPTA 2017
Publisher: IEEE
ISBN (Electronic): 9781538618417
ASJC Scopus subject areas: Signal Processing, Radiology Nuclear Medicine and imaging
Keywords: image segmentation, Lossless compression, plenoptic image, quantum cut segmentation
Source: Scopus
Source ID: 85050764396

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Analysis of common rail pressure signal of dual-fuel large industrial engine for identification of injection duration of pilot diesel injectors

In this paper, we address the problem of identification of injection duration of common rail (CR) diesel pilot injectors of dual-fuel engines. In these pilot injectors, the injected volume is small and the repeatability of injections and identification of drifts of injectors are important factors, which need to be taken into account in order to achieve good repeatability (shot-to-shot with every cylinder) and therefore a well-balanced engine and furthermore reduced overall wear. This information can then be used for calibration and diagnostics purposes to guarantee engine longevity facilitated by consistent operating conditions throughout the life of the unit. A diagnostics method based on analysis of CR pressure with experimental results is presented in this paper. Using the developed method, the relative duration of injection events can be identified for multiple injectors. We use the phenomenon of drop in rail pressure due to an injection event as a feature of the injection process. The method is based on filtered CR pressure data during and after the injection event. First, the pressure signal during injection is extracted after control of each injection event. After that, the signal is normalized and filtered. Then a derivative of the filtered signal is calculated. Change in the derivative of the filtered signal larger than a predefined threshold indicates an injection event that can be detected and its relative duration can be identified. We present the experimental results and demonstrate the efficacy of the proposed methods using two different
types of pressure sensors. We are able to properly identify a change of ≥10 μs (2%, 500 μs) in injection time. This shows that the developed method detects drifts in injection duration and the magnitude of drift. This information can be used for adaptive control of injection duration, so that finally the injected fuel volume is the same as the original.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Automation and Hydraulic Engineering
Contributors: Krogerus, T., Hyvönen, M., Huhtala, K.
Pages: 1-9
Publication date: Mar 2018
Peer-reviewed: Yes
Early online date: 6 Dec 2017

Publication information

Journal: Fuel
Volume: 216
ISSN (Print): 0016-2361
Ratings: 
  • Scopus rating (2018): CiteScore 5.8 SJR 1.745 SNIP 2.012
Original language: English
ASJC Scopus subject areas: Mechanical Engineering, Signal Processing, Modelling and Simulation, Applied Mathematics
Keywords: Analysis , Dual-fuel engine , Diesel , Common rail , Injector , Rail pressure
Electronic versions: 
URLs: 

Research output: Contribution to journalArticleScientificpeer-review

Deep multiresolution color constancy

In this paper, a computational color constancy method is proposed via estimating the illuminant chromaticity in a scene by pooling from many local estimates. To this end, first, for each image in a dataset, we form an image pyramid consisting of several scales of the original image. Next, local patches of certain size are extracted from each scale in this image pyramid. Then, a convolutional neural network is trained to estimate the illuminant chromaticity per-patch. Finally, two more consecutive trainings are conducted, where the estimation is made per-image via taking the mean (1st training) and median (2nd training) of local estimates. The proposed method is shown to outperform the state-of-the-art in a widely used color constancy dataset.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Intel Finland
Contributors: Aytekin, C., Nikkanen, J., Gabbouj, M.
Number of pages: 5
Pages: 3735-3739
Publication date: 20 Feb 2018

Host publication information

Title of host publication: 2017 IEEE International Conference on Image Processing, ICIP 2017 - Proceedings
Publisher: IEEE COMPUTER SOCIETY PRESS
ISBN (Electronic): 9781509021758
ASJC Scopus subject areas: Software, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Color constancy, Deep learning, Illuminant chromaticity estimation, Local estimation, Multi-resolution

Bibliographical note

jufoid=57423

Source: Scopus
Source ID: 85045299547

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

360-Degree video streaming and its subjective quality

Traditional challenges for deploying end-to-end streaming systems are made harder when considering 360-degree media content. One of these challenges relates to the lack of commonly accepted standardized methodologies for subjective 360-degree video quality assessment, especially oriented towards streaming services. The contribution of this paper falls in the area of subjective assessment of 360-degree video. - From traditional standardized test methodologies originally designed for 2D/3D video, we tailored a methodology more oriented towards Virtual Reality (VR) streaming services. The methodology inherits a lot from existing ITU standards for video subjective quality evaluation. The additions incorporate the special properties of 360-video, namely omnidirectionality, as opposed to traditional video. - With this goal in mind, a new metric called Similarity Ring Metric (SRM) is introduced. It measures the degree of similarity in watching patterns of a single subject or between different subjects for several subjective assessment tests. This metric enables an inclusion or rejection criteria for test results in subjective assessment sessions. We also present visual fatigue results related to a subjective quality experiment of 360-degree video.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Nokia Technologies
Contributors: Curcio, I. D., Toukomaa, H., Naik, D.
Number of pages: 23
Pages: 1-23
Publication date: 5 Feb 2018

Host publication information

Title of host publication: SMPTE 2017 Annual Technical Conference and Exhibition, SMPTE 2017
Publisher: SMPTE
ISBN (Electronic): 9781614829591
ASJC Scopus subject areas: Organizational Behavior and Human Resource Management, Computer Networks and Communications, Signal Processing, Information Systems and Management, Media Technology
Keywords: 360 degrees video, Omnidirectional video, subjective assessment, subjective quality evaluation, test methodology, virtual reality streaming

Bibliographical note

EXT="Curcio, Igor D.D."

Source: Scopus
Source ID: 85050404607

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Detection and Classification of Acoustic Scenes and Events: Outcome of the DCASE 2016 Challenge

Public evaluation campaigns and datasets promote active development in target research areas, allowing direct comparison of algorithms. The second edition of the challenge on Detection and Classification of Acoustic Scenes and Events (DCASE 2016) has offered such an opportunity for development of state-of-the-art methods, and succeeded in drawing together a large number of participants from academic and industrial backgrounds. In this paper, we report on the tasks and outcomes of the DCASE 2016 challenge. The challenge comprised four tasks: acoustic scene classification, sound event detection in synthetic audio, sound event detection in real-life audio, and domestic audio tagging. We present in detail each task and analyse the submitted systems in terms of design and performance. We observe the emergence of deep learning as the most popular classification method, replacing the traditional approaches based on Gaussian mixture models and support vector machines. By contrast, feature representations have not changed substantially throughout the years, as mel frequency-based representations predominate in all tasks. The datasets created for and used in DCASE 2016 are publicly available and are a valuable resource for further research.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Audio research group - ARG, School of Mathematical Sciences, Ecole Centrale de Nantes, IRCCyN UMR CNRS 6597 1, University of Surrey
Contributors: Mesaros, A., Heittola, T., Benetos, E., Foster, P., Lagrange, M., Virtanen, T., Plumbley, M. D.
Pages: 379-393
Publication date: Feb 2018
Peer-reviewed: Yes
Early online date: 28 Nov 2017

Publication information

Journal: IEEE/ACM Transactions on Audio Speech and Language Processing
Volume: 26
Issue number: 2
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2018): CiteScore 5.06 SJR 1.045 SNIP 2.382
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Instrumentation, Acoustics and Ultrasonics, Linguistics and Language, Electrical and Electronic Engineering, Speech and Hearing
Keywords: Acoustic scene classification, Acoustics, audio datasets, Event detection, Hidden Markov models, pattern recognition, sound event detection, Speech, Speech processing, Tagging
Electronic versions: 
Source: Scopus
Source ID: 85037645922

Research output: Contribution to journalArticleScientificpeer-review

Damped Posterior Linearization Filter

In this letter, we propose an iterative Kalman type algorithm based on posterior linearization. The proposed algorithm uses a nested loop structure to optimize the mean of the estimate in the inner loop and update the covariance, which is a computationally more expensive operation, only in the outer loop. The optimization of the mean update is done using a damped algorithm to avoid divergence. Our simulations show that the proposed algorithm is more accurate than existing iterative Kalman filters.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Automation and Hydraulic Engineering, Lab of Electrical and Automation, University of Liverpool
Contributors: Raitoharju, M., Svensson, L., Garcia-Fernandez, A. F., Piche, R.
Publication date: 2018
Peer-reviewed: Yes
Early online date: 13 Feb 2018

Publication information

Journal: IEEE Signal Processing Letters
Volume: 25
Issue number: 4
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2018): CiteScore 4.37 SJR 0.785 SNIP 1.88
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering, Applied Mathematics
Keywords: Bayesian state estimation, Computational modeling, Convergence, Cost function, estimation, Kalman filters, Noise measurement, nonlinear, Signal processing algorithms
Electronic versions: 

Bibliographical note

EXT="Raitoharju, Matti"

Source: Scopus
Source ID: 85042110206

Research output: Contribution to journalArticleScientificpeer-review

Estimation and Mitigation of Channel Non-Reciprocity in Massive MIMO

Time-division duplex (TDD) based massive MIMO systems rely on the reciprocity of the wireless propagation channels when calculating the downlink precoders based on uplink pilots. However, the effective uplink and downlink channels incorporating the analog radio front-ends of the base station (BS) and user equipments (UEs) exhibit non-reciprocity due to non-identical behavior of the individual transmit and receive chains. When downlink precoder is not aware of such channel non-reciprocity (NRC), system performance can be significantly degraded due to NRC induced interference terms. In this work, we consider a general TDD-based massive MIMO system where frequency-response mismatches at both the BS and UEs, as well as the mutual coupling mismatch at the BS large-array system all coexist and induce channel NRC. Based on the NRC-impaired signal models, we first propose a novel iterative estimation method for acquiring both the BS and UE side NRC matrices and then also propose a novel NRC-aware downlink precoder design which utilizes the obtained estimates. Furthermore, an efficient pilot signaling scheme between the BS and UEs is introduced in order to facilitate executing the proposed estimation method and the NRC-aware precoding technique in practical systems. Comprehensive numerical results indicate substantially improved spectral efficiency performance when the proposed NRC estimation and NRC-aware precoding methods are adopted, compared to the existing state-of-the-art methods.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning
Contributors: Raeesi, O., Gokceoglu, A., Valkama, M.
Publication date: 2018
Peer-reviewed: Yes
Early online date: 10 Mar 2018

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 66
Issue number: 10
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2018): CiteScore 6.28 SJR 1.477 SNIP 2.55
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Beamforming, channel non-reciprocity, channel state information, frequency-response mismatch, linear precoding, massive MIMO, mutual coupling, time division duplexing (TDD)
Electronic versions: 
Source: Scopus
Source ID: 85043450743

Research output: Contribution to journalArticleScientificpeer-review

Model-Based Dynamic Scheduling for Multicore Signal Processing

This paper presents a model-based design method and a corresponding new software tool, the HTGS Model-Based Engine (HMBE), for designing and implementing dataflow-based signal processing applications on multi-core architectures. HMBE provides complementary capabilities to HTGS (Hybrid Task Graph Scheduler), a recently-introduced software tool for implementing scalable workflows for high performance computing applications on compute nodes with high core counts and multiple GPUs. HMBE integrates model-based design approaches, founded on dataflow principles, with advanced design optimization techniques provided in HTGS. This integration contributes to (a) making the application of HTGS more systematic and less time consuming, (b) incorporating additional dataflow-based optimization capabilities with HTGS optimizations, and (c) automating significant parts of the HTGS-based design process using a principled approach. In this paper, we present HMBE with an emphasis on the model-based design approaches and the novel dynamic scheduling techniques that are developed as part of the tool. We demonstrate the utility of HMBE via two case studies: an image stitching application for large microscopy images and a background subtraction application for multispectral video streams.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, Research area: Computer engineering, University of Maryland, National Institute of Standards and Technology
Contributors: Wu, J., Blattner, T., Keyrouz, W., Bhattacharyya, S. S.
Number of pages: 14
Pages: 1-14
Publication date: 2018
Peer-reviewed: Yes
Early online date: 2018

Publication information

Journal: Journal of Signal Processing Systems
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2018): CiteScore 1 SJR 0.203 SNIP 0.497
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: Dataflow, Memory management, Multicore platforms, Scheduling
Source: Scopus
Source ID: 85054798661

Research output: Contribution to journalArticleScientificpeer-review

Modeling and estimation of signal-dependent and correlated noise

The additive white Gaussian noise (AWGN) model is ubiquitous in signal processing. This model is often justified by central-limit theorem (CLT) arguments. However, whereas the CLT may support a Gaussian distribution for the random errors, it does not provide any justification for the assumed additivity and whiteness. As a matter of fact, data acquired in real applications can seldom be described with good approximation by the AWGN model, especially because errors are typically correlated and not additive. Failure to model accurately the noise leads to inaccurate analysis, ineffective filtering, and distortion or even failure in the estimation. This chapter provides an introduction to both signal-dependent and correlated noise and to the relevant models and basic methods for the analysis and estimation of these types of noise. Generic one-parameter families of distributions are used as the essential mathematical setting for the observed signals. The distribution families covered as leading examples include Poisson, mixed Poisson–Gaussian, various forms of signal-dependent Gaussian noise (including multiplicative families and approximations of the Poisson family), as well as doubly censored heteroskedastic Gaussian distributions. We also consider various forms of noise correlation, encompassing pixel and readout cross-talk, fixed-pattern noise, column/row noise, etc., as well as related issues like photo-response and gain nonuniformity. The introduced models and methods are applicable to several important imaging scenarios and technologies, such as raw data from digital camera sensors, various types of radiation imaging relevant to security and to biomedical imaging.

General information

Publication status: Published
MoE publication type: A3 Part of a book or another research book
Organisations: Signal Processing, Research group: Signal and Image Restoration-RST, University of São Paulo
Contributors: Azzari, L., Borges, L. R., Foi, A.
Number of pages: 36
Pages: 1-36
Publication date: 2018

Host publication information

Title of host publication: Denoising of Photographic Images and Video : Fundamentals, Open Challenges and New Trends
Publisher: SPRINGER-VERLAG LONDON LTD
ISBN (Print): 978-3-319-96028-9

Publication series

Name: Advances in Computer Vision and Pattern Recognition
ISSN (Print): 2191-6586
ISSN (Electronic): 2191-6594
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
URLs: 
Source: Scopus
Source ID: 85053409603

Research output: Chapter in Book/Report/Conference proceedingChapterScientificpeer-review

Multi-view predictive latent space learning

In unsupervised circumstances, multi-view learning seeks a shared latent representation by taking the consensus and complementary principles into account. However, most existing multi-view unsupervised learning approaches do not explicitly lay stress on the predictability of the latent space. In this paper, we propose a novel multi-view predictive latent space learning (MVP) model and apply it to multi-view clustering and unsupervised dimension reduction. The latent space is forced to be predictive by maximizing the correlation between the latent space and feature space of each view. By learning a multi-view graph with adaptive view-weight learning, MVP effectively combines the complementary information from multi-view data. Experimental results on benchmark datasets show that MVP outperforms the state-of-the-art multi-view clustering and unsupervised dimension reduction algorithms.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Tianjin University
Contributors: Yuan, J., Gao, K., Zhu, P., Egiazarian, K.
Publication date: 2018
Peer-reviewed: Yes
Early online date: 2018

Publication information

Journal: Pattern Recognition Letters
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2018): CiteScore 3.73 SJR 0.662 SNIP 1.64
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
Keywords: Multi-view learning, Predictive latent space learning, Unsupervised clustering, Unsupervised dimension reduction
Source: Scopus
Source ID: 85049094619

Research output: Contribution to journalArticleScientificpeer-review

Probabilistic saliency estimation

In this paper, we model the salient object detection problem under a probabilistic framework encoding the boundary connectivity saliency cue and smoothness constraints into an optimization problem. We show that this problem has a closed form global optimum solution, which estimates the salient object. We further show that along with the probabilistic framework, the proposed method also enjoys a wide range of interpretations, i.e. graph cut, diffusion maps and one-class classification. With an analysis according to these interpretations, we also find that our proposed method provides approximations to the global optimum to another criterion that integrates local/global contrast and large area saliency cues. The proposed unsupervised approach achieves mostly leading performance compared to the state-of-the-art unsupervised algorithms over a large set of salient object detection datasets including around 17k images for several evaluation metrics. Furthermore, the computational complexity of the proposed method is favorable/comparable to many state-of-the-art unsupervised techniques.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG
Contributors: Aytekin, C., Iosifidis, A., Gabbouj, M.
Number of pages: 14
Pages: 359-372
Publication date: 2018
Peer-reviewed: Yes
Early online date: 20 Sep 2017

Publication information

Journal: Pattern Recognition
Volume: 74
ISSN (Print): 0031-3203
Ratings: 
  • Scopus rating (2018): CiteScore 7.35 SJR 1.363 SNIP 3.012
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
Keywords: Diffusion maps, One-class classification, Probabilistic model, Saliency, Salient object detection, Spectral graph cut
Source: Scopus
Source ID: 85032271491

Research output: Contribution to journalArticleScientificpeer-review

Separation of Moving Sound Sources Using Multichannel NMF and Acoustic Tracking

In this paper we propose a method for separation of moving sound sources. The method is based on first tracking the sources and then estimation of source spectrograms using multichannel non-negative matrix factorization (NMF) and extracting the sources from the mixture by single-channel Wiener filtering. We propose a novel multichannel NMF model with time-varying mixing of the sources denoted by spatial covariance matrices (SCM) and provide update equations for optimizing model parameters minimizing squared Frobenius norm. The SCMs of the model are obtained based on estimated directions of arrival of tracked sources at each time frame. The evaluation is based on established objective separation criteria and using real recordings of two and three simultaneous moving sound sources. The compared methods include conventional beamforming and ideal ratio mask separation. The proposed method is shown to exceed the separation quality of other evaluated blind approaches according to all measured quantities. Additionally, we evaluate the method's susceptibility towards tracking errors by comparing the separation quality achieved using annotated ground truth source trajectories.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Audio research group - ARG
Contributors: Nikunen, J., Diment, A., Virtanen, T.
Pages: 281-295
Publication date: 2018
Peer-reviewed: Yes
Early online date: 16 Nov 2017

Publication information

Journal: IEEE/ACM Transactions on Audio Speech and Language Processing
Volume: 26
Issue number: 2
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2018): CiteScore 5.06 SJR 1.045 SNIP 2.382
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Instrumentation, Acoustics and Ultrasonics, Linguistics and Language, Electrical and Electronic Engineering, Speech and Hearing
Keywords: acoustic source tracking, Acoustics, Array signal processing, Direction-of-arrival estimation, Estimation, Mathematical model, microphone arrays, Microphones, moving sound sources, Sound source separation, Spectrogram, time-varying mixing model
Source: Scopus
Source ID: 85035748523

Research output: Contribution to journalArticleScientificpeer-review

Multilinear class-specific discriminant analysis

There has been a great effort to transfer linear discriminant techniques that operate on vector data to high-order data, generally referred to as Multilinear Discriminant Analysis (MDA) techniques. Many existing works focus on maximizing the inter-class variances to intra-class variances defined on tensor data representations. However, there has not been any attempt to employ class-specific discrimination criteria for the tensor data. In this paper, we propose a multilinear subspace learning technique suitable for applications requiring class-specific tensor models. The method maximizes the discrimination of each individual class in the feature space while retains the spatial structure of the input. We evaluate the efficiency of the proposed method on two problems, i.e. facial image analysis and stock price prediction based on limit order book data.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Aarhus Universitet, Laboratory of Signal Processing
Contributors: Thanh Tran, D., Gabbouj, M., Iosifidis, A.
Number of pages: 6
Pages: 131-136
Publication date: 1 Dec 2017
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition Letters
Volume: 100
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2017): CiteScore 3 SJR 0.662 SNIP 1.556
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Artificial Intelligence
Keywords: Class-specific discriminant learning, Face verification, Multilinear discriminant analysis, Stock price prediction

Bibliographical note

INT=sgn,"Thanh Tran, Dat"

Source: Scopus
Source ID: 85032300703

Research output: Contribution to journalArticleScientificpeer-review

Sparse approximations in complex domain based on BM3D modeling

In this paper the concept of sparsity for complex-valued variables is introduced in the following three types: directly in complex domain and for two real-valued pairs phase/amplitude and real/imaginary parts of complex variables. The nonlocal block-matching technique is used for sparsity implementation and filter design for each type of sparsity. These filters are complex domain generalizations of the Block Matching 3D collaborative (BM3D) filter based on the high-order singular value decomposition (HOSVD) in order to generate group-wise adaptive analysis/synthesis transforms. Complex domain denoising is developed and studied as a test-problem for comparison of the designed filters as well as the different types of sparsity modeling.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Computational Imaging-CI
Contributors: Katkovnik, V., Ponomarenko, M., Egiazarian, K.
Number of pages: 13
Pages: 96-108
Publication date: 1 Dec 2017
Peer-reviewed: Yes

Publication information

Journal: Signal Processing
Volume: 141
ISSN (Print): 0165-1684
Ratings: 
  • Scopus rating (2017): CiteScore 3.94 SJR 0.94 SNIP 1.947
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Software, Signal Processing, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering
Keywords: Block matching, Complex domain, Denoising, Elsevier article, Phase imaging, Sample document, Sparsity
Source: Scopus
Source ID: 85020311730

Research output: Contribution to journalArticleScientificpeer-review

Model-based dynamic scheduling for multicore implementation of image processing systems

In this paper, we present a new software tool, called HTGS Model-based Engine (HMBE), for the design and implementation of multicore signal processing applications. HMBE provides complementary capabilities to HTGS (Hybrid Task Graph Scheduler), which is a recently-introduced software tool for implementing scalable workflows for high performance computing applications. HMBE integrates advanced design optimization techniques provided in HTGS with model-based approaches that are founded on dataflow principles. Such integration contributes to (a) making the application of HTGS more systematic and less time consuming, (b) incorporating additional dataflow-based optimization capabilities with HTGS optimizations, and (c) automating significant parts of the HTGS-based design process. In this paper, we present HMBE with an emphasis on novel dynamic scheduling techniques that are developed as part of the tool. We demonstrate the utility of HMBE through a case study involving an image stitching application for large scale microscopy images.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, University of Maryland, National Institute of Standards and Technology
Contributors: Wu, J., Blattner, T., Keyrouz, W., Bhattacharyya, S. S.
Publication date: 14 Nov 2017

Host publication information

Title of host publication: 2017 IEEE International Workshop on Signal Processing Systems, SiPS 2017
Publisher: IEEE
Article number: 8110003
ISBN (Electronic): 9781538604465
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics, Hardware and Architecture
Source: Scopus
Source ID: 85040564128

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Big Media Data Analysis

In this editorial a short introduction to the special issue on Big Media Data Analysis is given. The scope of this Editorial is to briefly present methodologies, tasks and applications of big media data analysis and to introduce the papers of the special issue. The special issue includes six papers that span various media analysis application areas like generic image description, medical image and video analysis, distance calculation acceleration and data collection.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG, Aarhus Universitet, University of Milan Bicocca, Department of Informatics, Aristotle University of Thessaloniki
Contributors: Iosifidis, A., Tefas, A., Pitas, I., Gabbouj, M.
Number of pages: 4
Pages: 105-108
Publication date: 1 Nov 2017
Peer-reviewed: Yes

Publication information

Journal: Signal Processing: Image Communication
Volume: 59
ISSN (Print): 0923-5965
Ratings: 
  • Scopus rating (2017): CiteScore 3.06 SJR 0.551 SNIP 1.499
Original language: English
ASJC Scopus subject areas: Software, Signal Processing, Computer Vision and Pattern Recognition, Electrical and Electronic Engineering
Keywords: Big Media Data, Data analytics, Deep learning, Machine learning, Statistical learning

Bibliographical note

EXT="Tefas, Anastasios"

Source: Scopus
Source ID: 85033445526

Research output: Contribution to journalEditorialScientificpeer-review

An optimized embedded target detection system using acoustic and seismic sensors

Detection of targets using low power embedded devices has important applications in border security and surveillance. In this paper, we build on recent algorithmic advances in sensor fusion, and present the design and implementation of a novel, multi-mode embedded signal processing system for detection of people and vehicles using acoustic and seismic sensors. Here, by "multi-mode", we mean that the system has available a complementary set of configurations that are optimized for different trade-offs. The multimode capability delivered by the proposed system is useful to supporting long lifetime (long term, energy-efficient "standby" operation), while also supporting optimized accuracy during critical time periods (e.g., when a potential threat is detected). In our target detection system, we apply a strategically-configured suite of single-and dual-modality signal processing techniques together with dataflow-based design optimization for energyefficient, real-time implementation. Through experiments using a Raspberry Pi platform, we demonstrate the capability of our target detection system to provide efficient operational tradeoffs among detection accuracy, energy efficiency, and processing speed.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research area: Computer engineering, University of Maryland, U.S. Army Research Laboratory, Department of Electrical and Computer Engineering
Contributors: Lee, K., Riggan, B. S., Bhattacharyya, S. S.
Number of pages: 5
Pages: 986-990
Publication date: 23 Oct 2017

Host publication information

Title of host publication: 25th European Signal Processing Conference, EUSIPCO 2017
Publisher: IEEE
ISBN (Electronic): 9780992862671

Bibliographical note

jufoid=55867

Source: Scopus
Source ID: 85041468751

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Full search equivalent fast block matching using orthonormal tree-structured haar transform

The goal of block matching is to find small parts (blocks) of an image that are similar to a given pattern (template). A lot of full search (FS) equivalent algorithms are based on transforms. However, the template size is limited to be a power-of-two. In this paper, we consider a fast block matching algorithm based on orthonormal tree-structured Haar transform (OTSHT) which makes it possible to use a template with arbitrary size. We evaluated the pruning performance, computational complexity, and design of tree. The pruning performance is compared to the algorithm based on orthonormal Haar transform (OHT).

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Tokyo Institute of Technology
Contributors: Ito, I., Egiazarian, K.
Number of pages: 6
Pages: 177-182
Publication date: 18 Oct 2017

Host publication information

Title of host publication: ISPA 2017 - 10th International Symposium on Image and Signal Processing and Analysis
Publisher: IEEE COMPUTER SOCIETY PRESS
ISBN (Electronic): 9781509040117
ASJC Scopus subject areas: Computational Theory and Mathematics, Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Signal Processing

Bibliographical note

jufoid=57665

Source: Scopus
Source ID: 85037808698

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Sparse Overcomplete Denoising: Aggregation Versus Global Optimization

Denoising is often addressed via sparse coding with respect to an overcomplete dictionary. There are two main approaches when the dictionary is composed of translates of an orthonormal basis. The first, traditionally employed by techniques such as wavelet cycle spinning, separately seeks sparsity w.r.t. each translate of the orthonormal basis, solving multiple partial optimizations and obtaining a collection of sparse approximations of the noise-free image, which are aggregated together to obtain a final estimate. The second approach, recently employed by convolutional sparse representations, instead seeks sparsity over the entire dictionary via a global optimization. It is tempting to view the former approach as providing a suboptimal solution of the latter. In this letter, we analyze whether global sparsity is a desirable property, and under what conditions the global optimization provides a better solution to the denoising problem. In particular, our experimental analysis shows that the two approaches attain comparable performance in case of natural images and global optimization outperforms the simpler aggregation of partial estimates only when the image admits an extremely sparse representation. We explain this phenomenon by separately studying the bias and variance of these solutions, and by noting that the variance of the global solution increases very rapidly as the original signal becomes less and less sparse.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Politecnico di Milano, Los Alamos National Laboratory
Contributors: Carrera, D., Boracchi, G., Foi, A., Wohlberg, B.
Number of pages: 5
Pages: 1468-1472
Publication date: 1 Oct 2017
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 24
Issue number: 10
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2017): CiteScore 3.81 SJR 0.732 SNIP 1.77
Original language: English
ASJC Scopus subject areas: Signal Processing, Applied Mathematics, Electrical and Electronic Engineering
Keywords: Convolutional sparse coding, denoising, overcomplete representations, sparse representations

Bibliographical note

EXT="Carrera, Diego"
EXT="Boracchi, Giacomo"

Source: Scopus
Source ID: 85029186566

Research output: Contribution to journalArticleScientificpeer-review

Convergence analysis of error-reduction algorithm for solving of the extended one-dimensional discrete phase retrieval problem

The iterative methods are well-known approaches to solve the one-dimensional phase retrieval problem. Amongst them, the error-reduction algorithm is often used since it can easily implement support constraints. Unfortunately this method often stagnates. Recently we have formulated the extended form of the one-dimensional discrete phase retrieval problem and we have assumed that the stagnation can be avoided by oversampling. Simulations have indicated that the conjecture is true. In this work we prove the convergence of the error-reduction algorithm in the proposed extended one-dimensional discrete phase retrieval framework.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Technical University of Cluj-NapocaUniversitatea Tehnica din Cluj-Napoca
Contributors: Rusu, C., Astola, J.
Publication date: 12 Sep 2017

Host publication information

Title of host publication: ISSCS 2017 - International Symposium on Signals, Circuits and Systems
Publisher: IEEE
ISBN (Electronic): 9781538606742
ASJC Scopus subject areas: Computer Networks and Communications, Electrical and Electronic Engineering, Signal Processing

Bibliographical note

EXT="Rusu, Corneliu"

Source: Scopus
Source ID: 85032302395

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Lossless compression of high resolution disparity map images

High resolution disparity images are stored in floating point raw files, where the number of bits per pixel is typically 32, although the number of used bits when converted to a fixed point representation is lower, e.g., between 24 and 26 in the dataset used in our experiments. In order to compress images with such high dynamic range, the bitplanes of the original image are combined into integer images with at most 16 bits, for which readily existing compressors are available. We introduce first a context predictive compressor (CPC) which can operate on integer images having more than 16 bits. The proposed overall compression scheme uses a first revertible linear transformation of the image as a first decorrelation process, and then splits the transformed image into integer images with smaller dynamic range, which are finally encoded. We experiment with schemes of split-into-2 and split-into-3, with combinations of several existing compressors for the integer image components and show that the newly introduced CPC operating over the least significant bitplanes combined with CERV operating over the most significant bitplanes achieves always the best compression, with final lossless compressed results of between 8 and 12 bits per pixel.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Signal Interpretation and Compression-SIC
Contributors: Astola, P., Tabus, I.
Publication date: 12 Sep 2017

Host publication information

Title of host publication: ISSCS 2017 - International Symposium on Signals, Circuits and Systems
Publisher: IEEE
ISBN (Electronic): 9781538606742
ASJC Scopus subject areas: Computer Networks and Communications, Electrical and Electronic Engineering, Signal Processing
Source: Scopus
Source ID: 85032263668

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Binary Non-Negative Matrix Deconvolution for Audio Dictionary Learning

In this study, we propose an unsupervised method for dictionary learning in audio signals. The new method, called binary nonnegative matrix deconvolution (BNMD), is developed and used to discover patterns from magnitude-scale spectrograms. The BNMD models an audio spectrogram as a sum of delayed patterns having binary gains (activations). Only small subsets of patterns can be active for a given spectrogram excerpt. The proposed method was applied to speaker identification and separation tasks. The experimental results show that dictionaries obtained by the BNMD bring much higher speaker identification accuracies averaged over a range of SNRs from -6 dB to 9 dB (91.3%) than the NMD-based dictionaries (37.8-75.4%). The BNMD also gives a benefit over dictionaries obtained using vector quantization (87.8%). For bigger dictionaries the difference between the BNMD and the vector quantization (VQ) is getting smaller. For the speech separation task the BNMD dictionary gave a slight improvement over the VQ.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Audio research group - ARG, Poznan University of Technology, University of Oldenburg, Yousician Ltd.
Contributors: Drgas, S., Virtanen, T., Lücke, J., Hurmalainen, A.
Number of pages: 13
Pages: 1644-1656
Publication date: 1 Aug 2017
Peer-reviewed: Yes

Publication information

Journal: IEEE/ACM Transactions on Audio Speech and Language Processing
Volume: 25
Issue number: 8
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2017): CiteScore 4.4 SJR 0.841 SNIP 2.138
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Instrumentation, Acoustics and Ultrasonics, Linguistics and Language, Speech and Hearing, Electrical and Electronic Engineering
Keywords: Sparse coding, speaker recognition, speech separation
Electronic versions: 

Bibliographical note

EXT="Hurmalainen, Antti"

Source: Scopus
Source ID: 85028409183

Research output: Contribution to journalArticleScientificpeer-review

Analysis of multipath propagation for 5G system at higher frequencies in microcellular environment

The main target of this paper is to perform the multidimensional analysis of multipath propagation at higher frequencies i.e. 15 GHz and 28 GHz, using 'sAGA' a 3D ray tracing tool. A real world outdoor Line of Sight (LOS) microcellular environment from the Yokusuka city of Japan is considered for the analysis. The simulation data acquired from the 3D ray tracing tool includes the received signal strength, power angular spectrum and the power delay profile. The different propagation mechanisms were closely analyzed. The simulation results show the difference of propagation at two frequencies i.e. 15 GHz and 28 GHz and draw a special attention on the impact of diffuse scattering at 28 GHz. In a simple outdoor microcellular environment with a valid LOS link between the transmitter and a receiver, a path loss difference of around 5.7 dB was found between 15 GHz and 28 GHz frequency of operation. However, the propagation loss at higher frequency can be compensated by using the antenna with narrow beamwidth and larger gain.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Laboratory of Radio Network Planning, Research group: Wireless Communications and Positioning
Contributors: Sheikh, M. U., Lempiäinen, J.
Number of pages: 5
Pages: 1660-1664
Publication date: 19 Jul 2017

Host publication information

Title of host publication: 2017 13th International Wireless Communications and Mobile Computing Conference, IWCMC 2017
Publisher: IEEE
ISBN (Electronic): 9781509043729
ASJC Scopus subject areas: Computer Networks and Communications, Hardware and Architecture, Signal Processing
Keywords: 3D ray tracing, 5G, Microcellular, Multipath propagation, System performance
Source: Scopus
Source ID: 85027853365

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Angular wall loss model and Extended Building Penetration model for outdoor to indoor propagation

The main target of this research work is to study the provision of indoor service (coverage) using outdoor base stations at higher frequencies i.e. 10 GHz in the context of a single building scenario. In an outdoor to indoor propagation, an angular wall loss model is used in the General Building Penetration (GBP) model for estimating the additional loss at the intercept point of the building exterior wall. A novel angular wall loss model based on a separate incidence angle in azimuth and elevation plane is proposed in this paper. In the second part of this study, an Extended Building Penetration (EBP) model is proposed, and the performance of EBP model is compared with the GBP model. In EBP model, the additional fifth path known as the 'Direct path' is proposed to be included in the GBP model. Based on the evaluation results, the impact of the direct path is found significant for the indoor users having the same or closed by height as that of the height of the transmitter.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Laboratory of Radio Network Planning, Research group: Wireless Communications and Positioning, Ericsson Research
Contributors: Sheikh, M. U., Hiltunen, K., Lempiäinen, J.
Number of pages: 6
Pages: 1291-1296
Publication date: 19 Jul 2017

Host publication information

Title of host publication: 2017 13th International Wireless Communications and Mobile Computing Conference, IWCMC 2017
Publisher: IEEE
ISBN (Electronic): 9781509043729
ASJC Scopus subject areas: Computer Networks and Communications, Hardware and Architecture, Signal Processing
Keywords: Angular loss, Building penetration loss, Outdoor to Indoor, Propagation, Wall loss model
Source: Scopus
Source ID: 85027886696

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

A k-nearest neighbor multilabel ranking algorithm with application to content-based image retrieval

Multilabel ranking is an important machine learning task with many applications, such as content-based image retrieval (CBIR). However, when the number of labels is large, traditional algorithms are either infeasible or show poor performance. In this paper, we propose a simple yet effective multilabel ranking algorithm that is based on k-nearest neighbor paradigm. The proposed algorithm ranks labels according to the probabilities of the label association using the neighboring samples around a query sample. Different from traditional approaches, we take only positive samples into consideration and determine the model parameters by directly optimizing ranking loss measures. We evaluated the proposed algorithm using four popular multilabel datasets. The proposed algorithm achieves equivalent or better performance than other instance-based learning algorithms. When applied to a CBIR system with a dataset of 1 million samples and over 190 thousand labels, which is much larger than any other multilabel datasets used earlier, the proposed algorithm clearly outperforms the competing algorithms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG, Qatar University
Contributors: Zhang, H., Kiranyaz, S., Gabbouj, M.
Number of pages: 5
Pages: 2587-2591
Publication date: 16 Jun 2017

Host publication information

Title of host publication: 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781509041176
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Content-Based Image Retrieval, k-Nearest Neighbor, Multilabel Learning
Source: Scopus
Source ID: 85023746875

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

An accumulative fusion architecture for discriminating people and vehicles using acoustic and seismic signals

In this paper, we develop new multiclass classification algorithms for detecting people and vehicles by fusing data from a multimodal, unattended ground sensor node. The specific types of sensors that we apply in this work are acoustic and seismic sensors. We investigate two alternative approaches to multiclass classification in this context - the first is based on applying Dempster-Shafer Theory to perform score-level fusion, and the second involves the accumulation of local similarity evidences derived from a feature-level fusion model that combines both modalities. We experiment with the proposed algorithms using different datasets obtained from acoustic and seismic sensors in various outdoor environments, and evaluate the performance of the two algorithms in terms of receiver operating characteristic and classification accuracy. Our results demonstrate overall superiority of the proposed new feature-level fusion approach for multiclass discrimination among people, vehicles and noise.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, U.S. Army Research Laboratory, Department of Electrical and Computer Engineering, University of Maryland
Contributors: Lee, K., Riggan, B. S., Bhattacharyya, S. S.
Number of pages: 5
Pages: 2976-2980
Publication date: 16 Jun 2017

Host publication information

Title of host publication: 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781509041176
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: multiclass classification, Sensor fusion, target detection, tracking
Source: Scopus
Source ID: 85023750747

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Inband full-duplex radio access system with self-backhauling: Transmit power minimization under QOS requirements

In this paper, a self-backhauling radio access system is studied and analyzed. In particular, we consider a scenario where a full-duplex access node is serving mobile users simultaneously in uplink and downlink, while also maintaining a wireless backhaul connection. The full-duplex capability of the access node, together with large antenna arrays, allows it to do all of this using the same center frequency. The minimum transmit powers for such a system are solved in a closed form under the condition that certain Quality of Service (QoS) requirements, defined in terms of minimum uplink and downlink data rates, are fulfilled. It is demonstrated with numerical results that, by using the derived expressions for the optimal transmit powers, the probability of fulfilling the QoS requirements is greatly increased, while simultaneously the overall transmit power usage of the system is significantly reduced when compared to a benchmark scheme.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Aalto University
Contributors: Korpi, D., Riihonen, T., Valkama, M.
Number of pages: 5
Pages: 6558-6562
Publication date: 16 Jun 2017

Host publication information

Title of host publication: 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781509041176
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: Full-duplex, massive MIMO, self-backhauling
Source: Scopus
Source ID: 85023776611

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Reduced-complexity digital predistortion for massive MIMO

A novel reduced-complexity digital predistortion (DPD) solution is presented in this paper. The proposed DPD can suppress the unwanted distortions due to power amplifier (PA) nonlinearity and I/Q modulator impairments in direct conversion transmitters using reduced-bandwidth filtered basis functions. Moreover, the DPD parameter estimation is based on very simple decorrelation based closed-loop processing and reduced-bandwidth observation, thus further reducing the overall complexity. The proposed DPD can be used in large array or massive MIMO systems with large number of radio transceivers and PAs, where reducing the complexity of the DPD processing is very critical.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning
Contributors: Abdelaziz, M., Anttila, L., Valkama, M.
Number of pages: 5
Pages: 6478-6482
Publication date: 16 Jun 2017

Host publication information

Title of host publication: 2017 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2017 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781509041176
ASJC Scopus subject areas: Software, Signal Processing, Electrical and Electronic Engineering
Keywords: 5G, Digital Predistortion, Direct Conversion Transmitters, I/Q imbalance, Massive MIMO, Power Amplifiers
Source: Scopus
Source ID: 85023738523

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Guest Editorial: Implementation Issues in System-on-Chip

General information

Publication status: Published
MoE publication type: B1 Article in a scientific magazine
Organisations: Electronics and Communications Engineering, Research group: System-on-Chip for GNSS, Wireless Communications and Cyber-Physical Embedded Computing, Tallinn University of Technology
Contributors: Ellervee, P., Nurmi, J.
Number of pages: 2
Pages: 269-270
Publication date: 1 Jun 2017
Peer-reviewed: No

Publication information

Journal: Journal of Signal Processing Systems
Volume: 87
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Electronic versions: 
URLs: 
Source: Scopus
Source ID: 85017177298

Research output: Contribution to journalEditorialScientific

Introduction to the Special Section on Sound Scene and Event Analysis

The papers in this special section are devoted to the growing field of acoustic scene classification and acoustic event recognition. Machine listening systems still have difficulties to reach the ability of human listeners in the analysis of realistic acoustic scenes. If sustained research efforts have been made for decades in speech recognition, speaker identification and to a lesser extent in music information retrieval, the analysis of other types of sounds, such as environmental sounds, is the subject of growing interest from the community and is targeting an ever increasing set of audio categories. This problem appears to be particularly challenging due to the large variety of potential sound sources in the scene, which may in addition have highly different acoustic characteristics, especially in bioacoustics. Furthermore, in realistic environments, multiple sources are often present simultaneously, and in reverberant conditions.

General information

Publication status: Published
MoE publication type: B1 Article in a scientific magazine
Organisations: Signal Processing, University Paris-Saclay, New York University, National Institute of Informatics, University of Toulon
Contributors: Richard, G., Virtanen, T., Bello, J. P., Ono, N., Glotin, H.
Number of pages: 3
Pages: 1169-1171
Publication date: 1 Jun 2017
Peer-reviewed: No

Publication information

Journal: IEEE/ACM Transactions on Audio Speech and Language Processing
Volume: 25
Issue number: 6
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2017): CiteScore 4.4 SJR 0.841 SNIP 2.138
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Instrumentation, Acoustics and Ultrasonics, Linguistics and Language, Speech and Hearing, Electrical and Electronic Engineering
Source: Scopus
Source ID: 85028340233

Research output: Contribution to journalArticleScientific

Power Mitigation by Performance Equalization in a Heterogeneous Reconfigurable Multicore Architecture

This paper presents an integrated self-aware computing model mitigating the power dissipation of a heterogeneous reconfigurable multicore architecture by dynamically scaling the operating frequency of each core. The power mitigation is achieved by equalizing the performance of all the cores for an uninterrupted exchange of data. The multicore platform consists of heterogeneous Coarse-Grained Reconfigurable Arrays (CGRAs) of application-specific sizes and a Reduced Instruction-Set Computing (RISC) core. The CGRAs and the RISC core are integrated with each other over a Network-on-Chip (NoC) of six nodes arranged in a topology of two rows and three columns. The RISC core constantly monitors and controls the performance of each CGRA accelerator by adjusting the operating frequencies unless the performance of all the CGRAs is optimally balanced over the platform. The CGRA cores on the platform are processing some of the most computationally-intensive signal processing algorithms while the RISC core establishes packet based synchronization between the cores for computation and communication. All the cores can access each other’s computational and memory resources while processing the kernels simultaneously and independently of each other. Besides general-purpose processing and overall platform supervision, the RISC processor manages performance equalization among all the cores which mitigates the overall dynamic power dissipation by 20.7 % for a proof-of-concept test.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: System-on-Chip for GNSS, Wireless Communications and Cyber-Physical Embedded Computing
Contributors: Hussain, W., Hoffmann, H., Ahonen, T., Nurmi, J.
Number of pages: 11
Pages: 287–297
Publication date: Jun 2017
Peer-reviewed: Yes
Early online date: 5 May 2016

Publication information

Journal: Journal of Signal Processing Systems
Volume: 87
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Modelling and Simulation, Hardware and Architecture, Information Systems, Signal Processing, Theoretical Computer Science
Keywords: CGRA, Dark silicon, Heterogeneous, Multicore, Power dissipation, Reconfigurable
Source: Scopus
Source ID: 84965022070

Research output: Contribution to journalArticleScientificpeer-review

The variation of air and surface temperatures in London within a 1km grid using vehicle-transect and ASTER data

Urbanisation can modify the local climate, increasing the temperature of cities compared to rural areas. This phenomenon is known as the Urban Heat Island (UHI), and this paper introduces a methodology to investigate the spatial variability of air and surface temperatures across London. In particular, this study aims to investigate if a widely used spatial resolution (1 km) is appropriate for heat-related health risk studies. Data from vehicle-transect and ASTER thermal images were overlaid on a reference grid of 1 km, used by UHI simulation models. The results showed higher variability of air temperature within some specific modelled grid cells in the city centre, while surface temperatures presented higher variability in the London borders. This investigation suggests that LST has larger variation levels and more grid cells with sub-grid variation above 1°C compared to air temperature measurements.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: University College London, London School of Hygiene and Tropical Medicine
Contributors: Dos Santos, R. S., Taylor, J., Davies, M., Mavrogianni, A., Milner, J.
Publication date: 10 May 2017

Host publication information

Title of host publication: 2017 Joint Urban Remote Sensing Event, JURSE 2017
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7924613
ISBN (Electronic): 9781509058082

Publication series

Name: 2017 Joint Urban Remote Sensing Event, JURSE 2017
ASJC Scopus subject areas: Signal Processing, Urban Studies, Management, Monitoring, Policy and Law, Instrumentation
Source: Scopus
Source ID: 85020236643

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

The Effect of Light Field Reconstruction and Angular Resolution Reduction on the Quality of Experience

The quality of visual contents displayed on 3D autostereoscopic displays-such as light field displays-essentially depend on factors that are not present in case of 3D stereoscopic or 2D ones, like angular resolution. A higher number of views in a given field of view enables a smoother, continuous motion parallax, but evidently requires more resources to transmit and display. However, in several cases a sufficiently high number of views might not even be available, thus light field reconstruction is required to increase the density of intermediate views. In this paper we introduce the results of a research aiming to measure the perceptual difference between light field reconstruction and different angular resolutions via a series of subjective image quality assessments. The analysis also calls attention to transmission requirements of content for light field displays.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: 3D MEDIA, Robotic Vision Team, Kingston University, Holografika
Contributors: Kara, P. A., Kovacs, P. T., Vagharshakyan, S., Martini, M. G., Barsi, A., Balogh, T., Chuchvara, A., Chehaibi, A.
Number of pages: 6
Pages: 781-786
Publication date: 21 Apr 2017

Host publication information

Title of host publication: 2016 12th International Conference on Signal-Image Technology & Internet-Based Systems (SITIS)
Publisher: IEEE
ISBN (Electronic): 9781509056989
ASJC Scopus subject areas: Computer Vision and Pattern Recognition, Radiology Nuclear Medicine and imaging, Computer Networks and Communications, Signal Processing
Keywords: Angular Resolution, Image Quality, Light Field Display, Light Field Reconstruction, Perceived Quality, Quality of Experience, View Synthesis
Source: Scopus
Source ID: 85019236013

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Compact modeling and management of reconfiguration in digital channelizer implementation

In this work, a novel digital channelizer design is developed through the use of a compact, system-level modeling approach. The model efficiently captures key properties of a digital channelizer system and its time-varying operation. The model applies powerful Markov Decision Process (MDP) techniques in new ways for design optimization of reconfigurable channelization processing. The result is a promising methodology for design and implementation of digital channelizers that adapt dynamically to changing use cases and stochastic environments while optimizing simultaneously for multiple conflicting performance goals. The method is used to employ an MDP to generate a runtime reconfiguration policy for a time-varying environment. Through extensive simulations, the robustness of the adaptation is demonstrated in comparison with the prior state of the art.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, University of Maryland, Georgia Institute of Technology
Contributors: Sapio, A. E., Wolf, M., Bhattacharyya, S. S.
Number of pages: 5
Pages: 595-599
Publication date: 19 Apr 2017

Host publication information

Title of host publication: 2016 IEEE Global Conference on Signal and Information Processing, GlobalSIP 2016 - Proceedings
Publisher: IEEE
ISBN (Electronic): 9781509045457
ASJC Scopus subject areas: Signal Processing, Computer Networks and Communications
Keywords: Channelizer, Markov decision processes, Reconfigurable architectures, Software defined radio
Source: Scopus
Source ID: 85019194314

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

HEVC-compliant viewport-adaptive streaming of stereoscopic panoramic video

Virtual reality (VR) provides unprecedented immersive experience using high-resolution spherical stereoscopic panoramic video. Such an experience is achieved by using head-mounted display (HMD) which has very strict latency bounds in order to respond promptly to user movements. Conventional streaming of VR video requires large bandwidth because the entire captured panorama is transmitted. However, only a limited field-of-view (FOV) is displayed by an HMD, resulting in wastage of bandwidth. To alleviate the problem, this paper proposes a High Efficiency Video Coding (HEVC) compliant approach for efficient coding and streaming of stereoscopic VR content. The proposed method is based on partitioning video pictures into tiles, where only the required tiles corresponding to the primary viewport are transmitted in high resolution, while the remaining parts are transmitted in low resolution. Furthermore, this method enables coding stereoscopic video contents using a conventional HEVC codec, while still achieving significant compression gain by means of adopting inter-view prediction only in intra random access point (IRAP) pictures. Using this method, the predicted view can be decoded independently of the main view, hence allowing simultaneous decoding instances. Experimental results demonstrate that the proposed approach is able to substantially improve compression efficiency and streaming bitrate performance.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Multimedia Research Group - MRG, Nokia
Contributors: Zare, A., Sreedhar, K. K., Vadakital, V. K. M., Aminlou, A., Hannuksela, M. M., Gabbouj, M.
Publication date: 19 Apr 2017

Host publication information

Title of host publication: 2016 Picture Coding Symposium, PCS 2016
Publisher: IEEE
ISBN (Electronic): 9781509059669
ASJC Scopus subject areas: Media Technology, Signal Processing

Bibliographical note

EXT="Vadakital, Vinod Kumar Malamal"

Source: Scopus
Source ID: 85019449939

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Sparse phase imaging based on complex domain nonlocal BM3D techniques

The paper is addressed to 2D phase and amplitude estimation of complex-valued signals – that is, in particular, to estimation of modulo-2π interferometric phase images from periodic and noisy observations. These degradation mechanisms make phase image estimation a challenging problem. A sparse nonlocal data-adaptive imaging formalized in complex domain is used for phase and amplitude image reconstruction. Following the procedure of patch-based technique, the image is partitioned into small overlapping square patches. Block Matching Three Dimensional (BM3D) technique is developed for forming complex domain sparse spectral representations of complex-valued data. High Order Singular Value Decomposition (HOSVD) applied to BM3D groups enables the design of the orthonormal complex domain 3D transforms which are data adaptive and different for each BM3Ds group. An iterative version of the complex domain BM3D is designed from variational formulation of the problem. The convergence of this algorithm is shown. The effectiveness of the new sparse coding based algorithms is illustrated in simulation experiments where they demonstrate the state-of-the-art performance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Research group: Computational Imaging-CI
Contributors: Katkovnik, V., Egiazarian, K.
Number of pages: 14
Pages: 72-85
Publication date: 1 Apr 2017
Peer-reviewed: Yes

Publication information

Journal: Digital Signal Processing
Volume: 63
ISSN (Print): 1051-2004
Ratings: 
  • Scopus rating (2017): CiteScore 2.73 SJR 0.55 SNIP 1.397
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Complex domain image denoising, Complex domain sparsity, HOSVD, Non-local block matching, Phase imaging
Source: Scopus
Source ID: 85009165629

Research output: Contribution to journalArticleScientificpeer-review

Advanced architectures for self-interference cancellation in full-duplex radios: Algorithms and measurements

In this paper, we describe an advanced real-time cancellation architecture for efficient digital-domain suppression of self-interference in inband full-duplex devices. The digital canceller takes into account the nonlinear distortion produced by the transmitter power amplifier, and is thereby a robust solution for low-cost implementations. The developed real-time digital canceller implementation is then evaluated with actual RF measurements, where it is complemented with a real-time adaptive RF canceller. The obtained results show that the RF canceller and the developed digital canceller implementation can together cancel the residual self-interference below the receiver noise floor in real-time for a 20 MHz cancellation bandwidth.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning
Contributors: Korpi, D., Aghababaeetafreshi, M., Piilila, M., Anttila, L., Valkama, M.
Number of pages: 5
Pages: 1553-1557
Publication date: 1 Mar 2017

Host publication information

Title of host publication: 2016 50th Asilomar Conference on Signals, Systems and Computers
Publisher: IEEE
ISBN (Electronic): 9781538639542
ASJC Scopus subject areas: Signal Processing, Computer Networks and Communications

Bibliographical note

INT=elt,"Piilila, Mauno"

Source: Scopus
Source ID: 85016330499

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Neighborhood Matching for Image Retrieval

In the last few years, large-scale image retrieval has attracted a lot of attention from the multimedia community. Usual approaches addressing this task first generate an initial ranking of the reference images using fast approximations that do not take into consideration the spatial arrangement of local features in the image (e.g., the bag-of-words paradigm). The top positions of the rankings are then re-estimated with verification methods that deal with more complex information, such as the geometric layout of the image. This verification step allows pruning of many false positives at the expense of an increase in the computational complexity, which may prevent its application to large-scale retrieval problems. This paper describes a geometric method known as neighborhood matching (NM), which revisits the keypoint matching process by considering a neighborhood around each keypoint and improves the efficiency of a geometric verification step in the image search system. Multiple strategies are proposed and compared to incorporate NM into a large-scale image retrieval framework. A detailed analysis and comparison of these strategies and baseline methods have been investigated. The experiments show that the proposed method not only improves the computational efficiency, but also increases the retrieval performance and outperforms state-of-the-art methods in standard datasets, such as the Oxford 5 k and 105 k datasets, for which the spatial verification step has a significant impact on the system performance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, Universidad Carlos III de Madrid, Purdue University
Contributors: González-Díaz, I., Birinci, M., Díaz-De-María, F., Delp, E. J.
Number of pages: 15
Pages: 544-558
Publication date: 1 Mar 2017
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Multimedia
Volume: 19
Issue number: 3
ISSN (Print): 1520-9210
Ratings: 
  • Scopus rating (2017): CiteScore 5.24 SJR 0.954 SNIP 2.283
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Computer Science Applications, Electrical and Electronic Engineering
Keywords: Geometric verification, image retrieval, neighborhood matching (NM), robust estimation
Source: Scopus
Source ID: 85013466531

Research output: Contribution to journalArticleScientificpeer-review

Low power design methodology for signal processing systems using lightweight dataflow techniques

Dataflow modeling techniques facilitate many aspects of design exploration and optimization for signal processing systems, such as efficient scheduling, memory management, and task synchronization. The lightweight dataflow (LWDF) programming methodology provides an abstract programming model that supports dataflow-based design and implementation of signal processing hardware and software components and systems. Previous work on LWDF techniques has emphasized their application to DSP software implementation. In this paper, we present new extensions of the LWDF methodology for effective integration with hardware description languages (HDLs), and we apply these extensions to develop efficient methods for low power DSP hardware implementation. Through a case study of a deep neural network application for vehicle classification, we demonstrate our proposed LWDF-based hardware design methodology, and its effectiveness in low power implementation of complex signal processing systems.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Signal Processing, Research group: Vision, Research area: Computer engineering, University of Maryland, Dept. of Electrical and Electronic Engineering, PolComIng - Information Engineering Unit, Department of Electrical and Computer Engineering
Contributors: Li, L., Fanni, T., Viitanen, T., Xie, R., Palumbo, F., Raffo, L., Huttunen, H., Takala, J., Bhattacharyya, S. S.
Number of pages: 8
Pages: 82-89
Publication date: 13 Feb 2017

Host publication information

Title of host publication: DASIP 2016 - Proceedings of the 2016 Conference on Design and Architectures for Signal and Image Processing
Publisher: IEEE COMPUTER SOCIETY PRESS
ISBN (Electronic): 9791092279153
ASJC Scopus subject areas: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering

Bibliographical note

INT=tie,"Xie, Renjie"

Source: Scopus
Source ID: 85014446650

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Color-distribution similarity by information theoretic divergence for color images

The divergence similarity between two color images is presented based on the Jensen-Shannon divergence to measure the color-distribution similarity. Subjective assessment experiments were developed to obtain mean opinion scores (MOS) of test images. It was found that the divergence similarity and MOS values showed statistically significant correlations.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing, Research group: Vision, University of Niigata, KLab, Japan, K-JIST, Dongguk University, Seoul
Contributors: Murayama, M., Oguro, D., Kikuchi, H., Huttunen, H., Ho, Y. S., Shin, J.
Publication date: 17 Jan 2017

Host publication information

Title of host publication: 2016 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference, APSIPA 2016
Publisher: IEEE
ISBN (Electronic): 9789881476821
ASJC Scopus subject areas: Artificial Intelligence, Computer Science Applications, Information Systems, Signal Processing

Bibliographical note

JUFOID=72850

Source: Scopus
Source ID: 85013813769

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

State Estimation for a Class of Piecewise Affine State-Space Models

We propose a filter for piecewise affine state-space models. In each filtering recursion, the true filtering posterior distribution is a mixture of truncated normal distributions. The proposed filter approximates the mixture with a single normal distribution via moment matching. The proposed algorithm is compared with the extended Kalman filter (EKF) in a numerical simulation, where the proposed method obtains, on average, better root mean square error than the EKF.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Automation Science and Engineering, Research area: Dynamic Systems, Research group: Positioning, Universidade Federal do Rio Grande do Sul, Porto Alegre, Linköping University
Contributors: Rui, R., Ardeshiri, T., Nurminen, H., Bazanella, A., Gustafsson, F.
Number of pages: 5
Pages: 61-65
Publication date: Jan 2017
Peer-reviewed: Yes
Early online date: 1 Dec 2016

Publication information

Journal: IEEE Signal Processing Letters
Volume: 24
Issue number: 1
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2017): CiteScore 3.81 SJR 0.732 SNIP 1.77
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Keywords: piecewise affine, state-space models, nonlinear filtering, Kalman filtering

Research output: Contribution to journalArticleScientificpeer-review

A Hybrid Task Graph Scheduler for High Performance Image Processing Workflows

Designing applications for scalability is key to improving their performance in hybrid and cluster computing. Scheduling code to utilize parallelism is difficult, particularly when dealing with data dependencies, memory management, data motion, and processor occupancy. The Hybrid Task Graph Scheduler (HTGS) improves programmer productivity when implementing hybrid workflows for multi-core and multi-GPU systems. The Hybrid Task Graph Scheduler (HTGS) is an abstract execution model, framework, and API that increases programmer productivity when implementing hybrid workflows for such systems. HTGS manages dependencies between tasks, represents CPU and GPU memories independently, overlaps computations with disk I/O and memory transfers, keeps multiple GPUs occupied, and uses all available compute resources. Through these abstractions, data motion and memory are explicit; this makes data locality decisions more accessible. To demonstrate the HTGS application program interface (API), we present implementations of two example algorithms: (1) a matrix multiplication that shows how easily task graphs can be used; and (2) a hybrid implementation of microscopy image stitching that reduces code size by ≈ 43% compared to a manually coded hybrid workflow implementation and showcases the minimal overhead of task graphs in HTGS. Both of the HTGS-based implementations show good performance. In image stitching the HTGS implementation achieves similar performance to the hybrid workflow implementation. Matrix multiplication with HTGS achieves 1.3x and 1.8x speedup over the multi-threaded OpenBLAS library for 16k × 16k and 32k × 32k size matrices, respectively.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, Research area: Computer engineering, University of Maryland Baltimore County, National Institute of Standards and Technology, Department of Electrical and Computer Engineering, University of Maryland
Contributors: Blattner, T., Keyrouz, W., Bhattacharyya, S. S., Halem, M., Brady, M.
Number of pages: 11
Pages: 457–467
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 89
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: Dataflow, Heterogeneous architectures, Hybrid workflows, Image processing, Matrix multiplication, Task graph
Source: Scopus
Source ID: 85025108758

Research output: Contribution to journalArticleScientificpeer-review

Data Flow Algorithms for Processors with Vector Extensions: Handling Actors With Internal State

Full use of the parallel computation capabilities of present and expected CPUs and GPUs requires use of vector extensions. Yet many actors in data flow systems for digital signal processing have internal state (or, equivalently, an edge that loops from the actor back to itself) that impose serial dependencies between actor invocations that make vectorizing across actor invocations impossible. Ideally, issues of inter-thread coordination required by serial data dependencies should be handled by code written by parallel programming experts that is separate from code specifying signal processing operations. The purpose of this paper is to present one approach for so doing in the case of actors that maintain state. We propose a methodology for using the parallel scan (also known as prefix sum) pattern to create algorithms for multiple simultaneous invocations of such an actor that results in vectorizable code. Two examples of applying this methodology are given: (1) infinite impulse response filters and (2) finite state machines. The correctness and performance of the resulting IIR filters and one class of FSMs are studied.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Pervasive Computing, Research area: Computer engineering, Signal Processing Research Community (SPRC), Keysight Technologies, University of Maryland
Contributors: Barford, L., Bhattacharyya, S. S., Liu, Y.
Pages: 21-31
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 87
Issue number: 1
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Modelling and Simulation, Hardware and Architecture, Information Systems, Signal Processing, Theoretical Computer Science
Keywords: Data flow computing, Digital signal processing, Graphics processing units, Parallel algorithms, Vector processors
Source: Scopus
Source ID: 84946115179

Research output: Contribution to journalArticleScientificpeer-review

Data Rate Assessment on L2–L3 CPU Bus and Bus between CPU and RAM in Modern CPUs

In this paper, modern CPU architecture with several different cache levels is described and current CPU performance limitations such as frequency increase bounds are discussed. As changes to the currently existing architecture are usually proposed as a way of increasing CPU performance, data rates of the internal and external CPU interfaces must be known. This information would help to assess the applicability of proposed solutions and to optimize them. This paper is aimed at obtaining real values of traffic on an L2–L3 cache interface inside a CPU and a CPU–RAM bus load, as well as showing the dependences of the total traffic on the studied interfaces on the number of active cores, CPU frequency, and test type. A measurement methodology using an Intel Performance Counter Monitor is provided and the equations used to obtain data rates from the internal CPU counters are explained. Both real-life and synthetic tests are described. The dependence of total traffic on the number of active cores and the dependence of total traffic on CPU frequency are provided as plots. The dependence of total traffic on test type is provided as a bar plot for multiple CPU frequencies.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Electronics and Communications Engineering, Yaroslavl State University
Contributors: Komar, M. S.
Number of pages: 8
Pages: 701-708
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Automatic Control and Computer Sciences
Volume: 51
Issue number: 7
ISSN (Print): 0146-4116
Ratings: 
  • Scopus rating (2017): CiteScore 0.53 SJR 0.218 SNIP 0.597
Original language: English
ASJC Scopus subject areas: Software, Control and Systems Engineering, Signal Processing
Keywords: data rate assessment, multicore CPUs, Network-on-Chip, NoC, System-on-Chip, Wireless Network- on-Chip, WNoC

Bibliographical note

INT=elt,"Komar, M. S."

Source: Scopus
Source ID: 85042229432

Research output: Contribution to journalArticleScientificpeer-review

Design Flow for GPU and Multicore Execution of Dynamic Dataflow Programs

Dataflow programming has received increasing attention in the age of multicore and heterogeneous computing. Modular and concurrent dataflow program descriptions enable highly automated approaches for design space exploration, optimization and deployment of applications. A great advance in dataflow programming has been the recent introduction of the RVC-CAL language. Having been standardized by the ISO, the RVC-CAL dataflow language provides a solid basis for the development of tools, design methodologies and design flows. This paper proposes a novel design flow for mapping RVC-CAL dataflow programs to parallel and heterogeneous execution platforms. Through the proposed design flow the programmer can describe an application in the RVC-CAL language and map it to multi- and many-core platforms, as well as GPUs, for efficient execution. The functionality and efficiency of the proposed approach is demonstrated by a parallel implementation of a video processing application and a run-time reconfigurable filter for telecommunications. Experiments are performed on GPU and multicore platforms with up to 16 cores, and the results show that for high-performance applications the proposed design flow provides up to 4 × higher throughput than the state-of-the-art approach in multicore execution of RVC-CAL programs.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, Research area: Computer engineering, Center for Machine Vision and Signal Analysis, Univ of Oulu
Contributors: Boutellier, J., Nyländen, T.
Number of pages: 10
Pages: 469–478
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 89
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: Dataflow computing, Design automation, Parallel processing, Signal processing
Source: Scopus
Source ID: 85021239311

Research output: Contribution to journalArticleScientificpeer-review

Implementation of a Multirate Resampler for Multi-carrier Systems on GPUs

Efficient sample rate conversion is of widespread importance in modern communication and signal processing systems. Although many efficient kinds of polyphase filterbank structures exist for this purpose, they are mainly geared toward serial, custom, dedicated hardware implementation for a single task. There is, therefore, a need for more flexible sample rate conversion systems that are resource-efficient, and provide high performance. To address these challenges, we present in this paper an all-software-based, fully parallel, multirate resampling method based on graphics processing units (GPUs). The proposed approach is well-suited for wireless communication systems that have simultaneous requirements on high throughput and low latency. Utilizing the multidimensional architecture of GPUs, our design allows efficient parallel processing across multiple channels and frequency bands at baseband. The resulting architecture provides flexible sample rate conversion that is designed to address modern communication requirements, including real-time processing of multiple carriers simultaneously.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Pervasive Computing, University of Maryland
Contributors: Kim, S. C., Bhattacharyya, S. S.
Number of pages: 11
Pages: 445–455
Publication date: 2017
Peer-reviewed: Yes
Early online date: 30 Mar 2017

Publication information

Journal: Journal of Signal Processing Systems
Volume: 89
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: Carrier aggregation, GPU-based radio, Multirate signal processing, Polyphase decimator, Polyphase interpolator, Polyphase resampler
Source: Scopus
Source ID: 85016560476

Research output: Contribution to journalArticleScientificpeer-review

Minimum description length sparse modeling and region merging for lossless plenoptic image compression

This paper proposes a complete lossless compression method for exploiting the redundancy of rectified light-field data. The light-field data consists of an array of rectified subaperture images, called for short views, which are segmented into regions according to an optimized partition of the central view. Each region of a view is predictively encoded using a specifically designed sparse predictor, exploiting the smoothness of each color component in the current view, and the cross-similarities with the other color components and already encoded neighbor views. The views are encoded sequentially, using a spiral scanning order, each view being predicted based on several similar neighbor views. The essential challenge for each predictor becomes choosing the most relevant regressors, from a large number of possible regressors belonging to the neighbor views. The proposed solution here is to couple sparse predictor design and minimum description length (MDL) principle, where the data description length is measured by an implementable code length, optimized for a class of probability models. The paper introduces a region merging segmentation under MDL criterion for partitioning the views into regions having their own specific sparse predictors. In experiments, several fast sparse design methods are considered. The proposed scheme is evaluated over a database of plenoptic images, achieving better lossless compression ratios than straightforward usage of standard image and video compression methods for the spiral sequence of views.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Signal Processing, University of California San Diego
Contributors: Helin, P., Astola, P., Rao, B., Tabus, I.
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: IEEE Journal on Selected Topics in Signal Processing
Volume: 11
Issue number: 7
ISSN (Print): 1932-4553
Ratings: 
  • Scopus rating (2017): CiteScore 6.96 SJR 1.331 SNIP 2.989
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Cameras, Data structures, Encoding, Image coding, Image color analysis, Image segmentation, light-field coding, lossless compression, minimum description length segmentation, plenoptics, sparse prediction, Spirals
Source: Scopus
Source ID: 85028989947

Research output: Contribution to journalArticleScientificpeer-review

On the existence of the solution for one-dimensional discrete phase retrieval problem

We consider the discrete form of the one-dimensional phase retrieval (1-D DPhR) problem from the point of view of input magnitude data. The direct method can provide a solution to the 1-D DPhR problem if certain conditions are satisfied by the input magnitude data, namely the corresponding trigonometric polynomial must be nonnegative. To test positivity of a trigonometric polynomial a novel DFT-based criterion is proposed. We use this DFT criterion for different sets of input magnitude data to evaluate whether the direct method applied to the 1-D DPhR problem leads to a solution in all explored cases.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Technical University of Cluj-NapocaUniversitatea Tehnica din Cluj-Napoca, Tampere International Center for Signal Processing
Contributors: Rusu, C., Astola, J.
Number of pages: 8
Pages: 195–202
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Signal, Image and Video Processing
Volume: 11
Issue number: 2
ISSN (Print): 1863-1703
Ratings: 
  • Scopus rating (2017): CiteScore 1.9 SJR 0.485 SNIP 1.228
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Discrete Fourier transform, Phase retrieval, Positive trigonometric polynomials, Signal reconstruction

Bibliographical note

EXT="Rusu, Corneliu"

Source: Scopus
Source ID: 84976319640

Research output: Contribution to journalArticleScientificpeer-review

Parallel Digital Predistortion Design on Mobile GPU and Embedded Multicore CPU for Mobile Transmitters

Digital predistortion (DPD) is a widely adopted baseband processing technique in current radio transmitters. While DPD can effectively suppress unwanted spurious spectrum emissions stemming from imperfections of analog RF and baseband electronics, it also introduces extra processing complexity and poses challenges on efficient and flexible implementations, especially for mobile cellular transmitters, considering their limited computing power compared to basestations. In this paper, we present high data rate implementations of broadband DPD on modern embedded processors, such as mobile GPU and multicore CPU, by taking advantage of emerging parallel computing techniques for exploiting their computing resources. We further verify the suppression effect of DPD experimentally on real radio hardware platforms. Performance evaluation results of our DPD design demonstrate the high efficacy of modern general purpose mobile processors on accelerating DPD processing for a mobile transmitter.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Rice University, Univ of Oulu
Contributors: Li, K., Ghazi, A., Tarver, C., Boutellier, J., Abdelaziz, M., Anttila, L., Juntti, M., Valkama, M., Cavallaro, J. R.
Number of pages: 14
Pages: 417–430
Publication date: 2017
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 89
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2017): CiteScore 0.96 SJR 0.216 SNIP 0.603
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Theoretical Computer Science, Signal Processing, Information Systems, Modelling and Simulation, Hardware and Architecture
Keywords: CUDA, Digital predistortion, Mobile SoC, NEON SIMD, Software-defined radio
Source: Scopus
Source ID: 85013872658

Research output: Contribution to journalArticleScientificpeer-review

Models of architecture: Reproducible efficiency evaluation for signal processing systems

The current trend in high performance and embedded signal processing consists of designing increasingly complex heterogeneous hardware architectures with non-uniform communication resources. In order to take hardware and software design decisions, early evaluations of the system non-functional properties are needed. These evaluations of system efficiency require high-level information on both the algorithms and the architecture. In this paper, we define the notion of Model of Architecture (MoA) and study the combination of a Model of Computation (MoC) and an MoA to provide a design space exploration environment for the study of the algorithmic and architectural choices. A cost is computed from the mapping of an application, represented by a model conforming a MoC onto an architecture represented by a model conforming an MoA. The cost is composed of a processing-related part and a communicationrelated part. It is an abstract scalar value to be minimized and can represent any non-functional requirement of a system such as memory, energy, throughput or latency.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Pervasive Computing, Research area: Computer engineering, Institut Pascal, UBL, Scuola Superiore sant'Anna, Salzburg University of Applied Sciences, University of Maryland
Contributors: Pelcat, M., Desnos, K., Maggiani, L., Liu, Y., Heulot, J., Nezan, J. F., Bhattacharyya, S. S.
Number of pages: 6
Pages: 121-126
Publication date: 9 Dec 2016

Host publication information

Title of host publication: IEEE International Workshop on Signal Processing Systems, SiPS 2016
Publisher: IEEE
Article number: 7780083
ISBN (Electronic): 9781509033614

Publication series

Name: IEEE International Workshop on Signal Processing Systems
ISSN (Electronic): 2374-7390
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics, Hardware and Architecture
Source: Scopus
Source ID: 85013141986

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Adaptive tracking of people and vehicles using mobile platforms

Tracking algorithms have important applications in detection of humans and vehicles for border security and other areas. For large-scale deployment of such algorithms, it is critical to provide methods for their cost- and energy-efficient realization. To this end, commodity mobile devices have significant potential for use as prototyping and testing platforms due to their low cost, widespread availability, and integration of advanced communications, sensing, and processing features. Prototypes developed on mobile platforms can be tested, fine-tuned, and demonstrated in the field and then provide reference implementations for application-specific disposable sensor node implementations that are targeted for deployment. In this paper, we develop a novel, adaptive tracking system that is optimized for energy-efficient, real-time operation on off-the-shelf mobile platforms. Our tracking system applies principles of dynamic data-driven application systems (DDDAS) to periodically monitor system operating characteristics and apply these measurements to dynamically adapt the specific classifier configurations that the system employs. Our resulting adaptive approach enables powerful optimization of trade-offs among energy consumption, real-time performance, and tracking accuracy based on time-varying changes in operational characteristics. Through experiments employing an Android-based tablet platform, we demonstrate the efficiency of our proposed tracking system design for multimode detection of human and vehicle targets.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Pervasive Computing, Research area: Computer engineering, University of Maryland, U.S. Army Research Laboratory, Technische Universitat Munchen
Contributors: Ben Salem, H., Damarla, T., Sudusinghe, K., Stechele, W., Bhattacharyya, S. S.
Publication date: 1 Dec 2016
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Advances in Signal Processing
Volume: 2016
Issue number: 1
Article number: 65
ISSN (Print): 1687-6172
Ratings: 
  • Scopus rating (2016): CiteScore 1.21 SJR 0.345 SNIP 1.05
Original language: English
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Keywords: Acoustic sensors, Dataflow graphs, DDDAS, Mobile platforms, Signal processing systems, Target tracking
Source: Scopus
Source ID: 84971500315

Research output: Contribution to journalArticleScientificpeer-review

Blind Separation of Audio Mixtures Through Nonnegative Tensor Factorization of Modulation Spectrograms

This paper presents an algorithm for unsupervised single-channel source separation of audio mixtures. The approach specifically addresses the challenging case of separation where no training data are available. By representing mixtures in the modulation spectrogram (MS) domain, we exploit underlying similarities in patterns present across frequency. A three-dimensional tensor factorization is able to take advantage of these redundant patterns, and is used to separate a mixture into an approximated sum of components by minimizing a divergence cost. Furthermore, we show that the basic tensor factorization can be extended with convolution in time being used to improve separation results and provide update rules to learn components in such a manner. Following factorization, sources are reconstructed in the audio domain from estimated components using a novel approach based on reconstruction masks that are learned using MS activations, and then applied to a mixture spectrogram. We demonstrate that the proposed method produces superior separation performance to a spectrally based nonnegative matrix factorization approach, in terms of source-to-distortion ratio. We also compare separation with the perceptually motivated interference-related perceptual score metric and identify cases with higher performance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Audio research group
Contributors: Barker, T., Virtanen, T.
Number of pages: 13
Pages: 2377-2389
Publication date: 1 Dec 2016
Peer-reviewed: Yes

Publication information

Journal: Ieee-Acm transactions on audio speech and language processing
Volume: 24
Issue number: 12
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2016): CiteScore 3.5 SJR 0.711 SNIP 2.949
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Instrumentation, Acoustics and Ultrasonics, Linguistics and Language, Speech and Hearing, Electrical and Electronic Engineering
Keywords: Factorization, nonnegative matrix factorization (NMF), source separation, speech enhancement
Source: Scopus
Source ID: 84991442139

Research output: Contribution to journalArticleScientificpeer-review

Generalized Hampel Filters

The standard median filter based on a symmetric moving window has only one tuning parameter: the window width. Despite this limitation, this filter has proven extremely useful and has motivated a number of extensions: weighted median filters, recursive median filters, and various cascade structures. The Hampel filter is a member of the class of decsion filters that replaces the central value in the data window with the median if it lies far enough from the median to be deemed an outlier. This filter depends on both the window width and an additional tuning parameter t, reducing to the median filter when t=0, so it may be regarded as another median filter extension. This paper adopts this view, defining and exploring the class of generalized Hampel filters obtained by applying the median filter extensions listed above: weighted Hampel filters, recursive Hampel filters, and their cascades. An important concept introduced here is that of an implosion sequence, a signal for which generalized Hampel filter performance is independent of the threshold parameter t. These sequences are important because the added flexibility of the generalized Hampel filters offers no practical advantage for implosion sequences. Partial characterization results are presented for these sequences, as are useful relationships between root sequences for generalized Hampel filters and their median-based counterparts. To illustrate the performance of this filter class, two examples are considered: one is simulation-based, providing a basis for quantitative evaluation of signal recovery performance as a function of t, while the other is a sequence of monthly Italian industrial production index values that exhibits glaring outliers.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Video, DataRobot, Aalto University
Contributors: Pearson, R. K., Neuvo, Y., Astola, J., Gabbouj, M.
Publication date: 1 Dec 2016
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Advances in Signal Processing
Volume: 2016
Issue number: 1
Article number: 87
ISSN (Print): 1687-6172
Ratings: 
  • Scopus rating (2016): CiteScore 1.21 SJR 0.345 SNIP 1.05
Original language: English
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Electronic versions: 
Source: Scopus
Source ID: 84981165367

Research output: Contribution to journalArticleScientificpeer-review

Ensembles of dense and dense sampling descriptors for the HEp-2 cells classification problem

The classification of Human Epithelial (HEp-2) cells images, acquired through Indirect Immunofluorescence (IIF) microscopy, is an effective method to identify staining patterns in patient sera. Indeed it can be used for diagnostic purposes, in order to reveal autoimmune diseases. However, the automated classification of IIF HEp-2 cell patterns represents a challenging task, due to the large intra-class and the small inter-class variability. Consequently, recent HEp-2 cell classification contests have greatly spurred the development of new IIF image classification systems.Here we propose an approach for the automatic classification of IIF HEp-2 cell images by fusion of several texture descriptors by ensemble of support vector machines combined by sum rule. Its effectiveness is evaluated using the HEp-2 cells dataset used for the "Performance Evaluation of Indirect Immunofluorescence Image Analysis Systems" contest, hosted by the International Conference on Pattern Recognition in 2014: the accuracy on the testing set is 79.85%.The same dataset was used to test an ensemble of ternary-encoded local phase quantization descriptors, built by perturbation approaches: the accuracy on the training set is 84.16%. Finally, this ensemble was validated on 14 additional datasets, obtaining the best performance on 11 datasets.Our MATLAB code is available at https://www.dei.unipd.it/node/2357.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Computational Biophysics and Imaging Group, BioMediTech, Integrated Technologies for Tissue Engineering Research (ITTE), Universita degli Studi di Padova, Italy, University of Bologna
Contributors: Nanni, L., Lumini, A., dos Santos, F. L. C., Paci, M., Hyttinen, J.
Pages: 28-35
Publication date: 15 Oct 2016
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition Letters
Volume: 82
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2016): CiteScore 2.9 SJR 0.729 SNIP 1.646
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Bag-of-features, Ensemble, HEp-2 cell classification, Machine learning, Support vector machine, Texture descriptors
URLs: 
Source: Scopus
Source ID: 84961195136

Research output: Contribution to journalArticleScientificpeer-review

Improved image quality in fast inpainting with omnidirectional filling

Image inpainting is an active research field of image processing. Previous inpainting methods often require a long computational time to give sufficient results, especially due to the extensive search process of exemplar-based methods. This work improves a previous fast inpainting method based on local similarity, which achieves runtimes measured in tens of milliseconds per image, but often results in unacceptable artifacts. We improve the resulting image quality by allowing pixels to be filled at any angle, determining the angle based only on the vicinity of the target region, and cross-fading between source pixels from opposite sides of the target region. The proposed method is shown to eliminate the two drawbacks, while retaining a fast runtime.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Pervasive Computing, Research area: Computer engineering, School of Science, Tohoku University
Contributors: Hirvola, O., Viitanen, T., Sintunata, V., Aoki, T.
Number of pages: 5
Pages: 31-35
Publication date: 22 Sep 2016

Host publication information

Title of host publication: International Conference on Image, Vision and Computing (ICIVC)
Publisher: IEEE
ISBN (Print): 978-1-5090-3756-8
ISBN (Electronic): 978-1-5090-3755-1
ASJC Scopus subject areas: Signal Processing

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

A software tool for studying the size and shape of human cardiomyocytes

Background and objectives Due to development of imaging systems the amount of digital images obtained in the biological field has been growing in recent years. These images contain information that is not directly measurable, e.g. the area covered by a single cell. In most of the current imaging programs the regions of interest (ROI), e.g. individual cells, need to be manually outlined. Automation of processing and analyzing the images would ease researchers’ workload and provide results that are more reliable. In this work our goal was to write software that automatically segments human cardiomyocytes from images, calculates their areas and variations in the direction of the largest and smallest spread. Results We developed software that eased the workload of biomedical laboratory personnel such that they do not have to do manual image segmentation or learn to use software that requires programming skills. The software made a correct segmentation in most of the cases and outperformed the intensity oriented baseline method written in ImageJ in 95% of comparisons. The baseline method estimated cell- and background areas by averaging dark background and bright foreground areas. Conclusions Our software can be used in the calculation of cell areas and extents in the case where immunolabeled cells are imaged with a fluorescent microscope. In the future the functionality of the program could be extended with machine learning methods that use the user actions as teaching material in the cases where automatic segmentation fails.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Tampere University Hospital, BioMediTech
Contributors: Rasku, J., Ojala, M., Pölönen, R. P., Joutsijoki, H., Gizatdinova, Y., Laurikkala, J., Kartasalo, K., Aalto-Setälä, K., Juhola, M.
Number of pages: 6
Pages: 134-139
Publication date: 1 Sep 2016
Peer-reviewed: Yes

Publication information

Journal: Biomedical Signal Processing and Control
Volume: 30
ISSN (Print): 1746-8094
Ratings: 
  • Scopus rating (2016): CiteScore 2.82 SJR 0.674 SNIP 1.69
Original language: English
ASJC Scopus subject areas: Health Informatics, Signal Processing
Keywords: Cardiomyocyte, Segmentation, Threshold
Source: Scopus
Source ID: 84978280164

Research output: Contribution to journalArticleScientificpeer-review

Image interpolation based on non-local geometric similarities and directional gradients

Image interpolation offers an efficient way to compose a high-resolution (HR) image from the observed low-resolution (LR) image. Advanced interpolation techniques design the interpolation weighting coefficients by solving a minimum mean-square-error (MMSE) problem in which the local geometric similarity is often considered. However, using local geometric similarities cannot usually make the MMSE-based interpolation as reliable as expected. To solve this problem, we propose a robust interpolation scheme by using the nonlocal geometric similarities to construct the HR image. In our proposed method, the MMSE-based interpolation weighting coefficients are generated by solving a regularized least squares problem that is built upon a number of dual-reference patches drawn from the given LR image and regularized by the directional gradients of these patches. Experimental results demonstrate that our proposed method offers a remarkable quality improvement as compared to some state-of-the-art methods, both objectively and subjectively.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Video, University of Electronic Science and Technology of China
Contributors: Zhu, S., Zeng, B., Zeng, L., Gabbouj, M.
Number of pages: 13
Pages: 1707-1719
Publication date: 1 Sep 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Multimedia
Volume: 18
Issue number: 9
ISSN (Print): 1520-9210
Ratings: 
  • Scopus rating (2016): CiteScore 5.4 SJR 1.298 SNIP 2.47
Original language: English
ASJC Scopus subject areas: Signal Processing, Media Technology, Computer Science Applications, Electrical and Electronic Engineering
Keywords: Directional gradient, geometric similarity, image interpolation, minimum mean square error (MMSE)
Source: Scopus
Source ID: 84983409176

Research output: Contribution to journalArticleScientificpeer-review

Nyström-based approximate kernel subspace learning

In this paper, we describe a method for the determination of a subspace of the feature space in kernel methods, which is suited to large-scale learning problems. Linear model learning in the obtained space corresponds to a nonlinear model learning process in the input space. Since the obtained feature space is determined only by exploiting properties of the training data, this approach can be used for generic nonlinear pattern recognition. That is, nonlinear data mapping can be considered to be a pre-processing step exploiting nonlinear relationships between the training data. Linear techniques can be subsequently applied in the new feature space and, thus, they can model nonlinear properties of the problem at hand. In order to appropriately address the inherent problem of kernel learning methods related to their time and memory complexities, we follow an approximate learning approach. We show that the method can lead to considerable operation speed gains and achieve very good performance. Experimental results verify our analysis.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Video, Research Community on Data-to-Decision (D2D)
Contributors: Iosifidis, A., Gabbouj, M.
Number of pages: 8
Pages: 190-197
Publication date: Sep 2016
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition
ISSN (Print): 0031-3203
Ratings: 
  • Scopus rating (2016): CiteScore 5.36 SJR 1.501 SNIP 2.953
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Kernel methods, Nonlinear pattern recognition, Nonlinear projection trick, Nyström approximation
Source: Scopus
Source ID: 85013223573

Research output: Contribution to journalArticleScientificpeer-review

Automatic objective thresholding to detect neuronal action potentials

In this paper, we introduce a fully objective method to set thresholds (THs) for neuronal action potential spike detection from extracellular field potential signals. Although several more sophisticated methods exist, thresholding is still the most used spike detection method. In general, it is employed by setting a TH as per convention or operator decision, and without considering either the undetected or spurious spikes. Here, we demonstrate with both simulations and real microelectrode measurement data that our method can fully automatically and objectively yield THs comparable to those set by an expert operator. A Matlab function implementation of the method is described, and provided freely in Matlab Central File Exchange.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, Research group: Computational Biophysics and Imaging Group, Department of Pervasive Computing, BioMediTech
Contributors: Tanskanen, J. M., Kapucu, F. E., Välkki, I., Hyttinen, J. A.
Number of pages: 5
Pages: 662-666
Publication date: 29 Aug 2016

Host publication information

Title of host publication: Proceedings of 2016 24th European Signal Processing Conference (EUSIPCO)
ISBN (Print): 978-1-5090-1891-8
ISBN (Electronic): 978-0-9928-6265-7
ASJC Scopus subject areas: Signal Processing, Neuroscience(all)
Keywords: Neuronal action potential, Thresholding, Spike detection, Microelectrode array, Electric field potential
Electronic versions: 

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Compressive Detection of Random Subspace Signals

The problem of compressive detection of random subspace signals is studied. We consider signals modeled as s = Hx where H is an N × K matrix with K ≤ N and x ~ N(0K,12IK). We say that signal s lies in or leans toward a subspace if the largest eigenvalue of HHT is strictly greater than its smallest eigenvalue. We first design a measurement matrix Φ = [ΦsT, ΦoT]T comprising of two sub-matrices Φs and Φo where Φs projects the signal to the strongest left-singular vectors, i.e., the left-singular vectors corresponding to the largest singular values, of subspace matrix H and Φo projects it to the weakest left-singular vectors. We then propose two detectors that work based on the difference in energies of the samples measured by the two sub-matrices Φs and Φo and provide theoretical proofs for their optimality. Simplified versions of the proposed detectors for the case when the variance of noise is known are also provided. Furthermore, we study the performance of the detector when measurements are imprecise and show how imprecision can be compensated by employing more measurement devices. The problem is then re-formulated for the generalized case when the signal lies in the union of a finite number of linear subspaces instead of a single linear subspace. Finally, we study the performance of the proposed methods by simulation examples.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Electrical Engineering Department, University of California, Los Angeles (UCLA)
Contributors: Razavi, A., Valkama, M., Cabric, D.
Number of pages: 14
Pages: 4166-4179
Publication date: 15 Aug 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 64
Issue number: 16
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2016): CiteScore 5.54 SJR 1.385 SNIP 2.533
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Compressive detection, F -distribution, hypothesis testing, random subspace signals, unknown noise variance
Source: Scopus
Source ID: 84980329115

Research output: Contribution to journalArticleScientificpeer-review

Variance Stabilization for Noisy+Estimate Combination in Iterative Poisson Denoising

We denoise Poisson images with an iterative algorithm that progressively improves the effectiveness of variance-stabilizing transformations (VST) for Gaussian denoising filters. At each iteration, a combination of the Poisson observations with the denoised estimate from the previous iteration is treated as scaled Poisson data and filtered through a VST scheme. Due to the slight mismatch between a true scaled Poisson distribution and this combination, a special exact unbiased inverse is designed. We present an implementation of this approach based on the BM3D Gaussian denoising filter. With a computational cost at worst twice that of the noniterative scheme, the proposed algorithm provides significantly better quality, particularly at low signal-to-noise ratio, outperforming much costlier state-of-the-art alternatives.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Signal and Image Restoration-RST
Contributors: Azzari, L., Foi, A.
Number of pages: 5
Pages: 1086-1090
Publication date: 1 Aug 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 23
Issue number: 8
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2016): CiteScore 3.54 SJR 0.798 SNIP 1.758
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Keywords: Anscombe transformation, image denoising, iterative filtering, photon-limited imaging, Poisson noise
Electronic versions: 
Additional files: 
Source: Scopus
Source ID: 84979891796

Research output: Contribution to journalArticleScientificpeer-review

Optimization of Flexible Filter Banks Based on Fast Convolution

Multirate filter banks can be implemented efficiently using fast-convolution (FC) processing. The main advantage of the FC filter banks (FC-FB) compared with the conventional polyphase implementations is their increased flexibility, that is, the number of channels, their bandwidths, and the center frequencies can be independently selected. In this paper, an approach to optimize the FC-FBs is proposed. First, a subband representation of the FC-FB is derived. Then, the optimization problems are formulated with the aid of the subband model. Finally, these problems are conveniently solved with the aid of a general nonlinear optimization algorithm. Several examples are included to demonstrate the proposed overall design scheme as well as to illustrate the efficiency and the flexibility of the resulting FC-FB.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning
Contributors: Yli-Kaakinen, J., Renfors, M.
Pages: 101-111
Publication date: Aug 2016
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 85
Issue number: 1
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2016): CiteScore 0.78 SJR 0.212 SNIP 0.65
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Modelling and Simulation, Hardware and Architecture, Information Systems, Signal Processing, Theoretical Computer Science
Keywords: Digital filters, Filter banks, Multirate signal processing, Optimization, Sampling rate conversion
Source: Scopus
Source ID: 84929682954

Research output: Contribution to journalArticleScientificpeer-review

Rotation Invariant Texture Description Using Symmetric Dense Microblock Difference

This letter is devoted to the problem of rotation invariant texture classification. Novel rotation invariant feature, symmetric dense microblock difference (SDMD), is proposed which captures the information at different orientations and scales. N-fold symmetry is introduced in the feature design configuration, while retaining the random structure that provides discriminative power. The symmetry is utilized to achieve a rotation invariance. The SDMD is extracted using an image pyramid and encoded by the Fisher vector approach resulting in a descriptor which captures variations at different resolutions without increasing the dimensionality. The proposed image representation is combined with the linear SVM classifier. Extensive experiments are conducted on four texture data sets [Brodatz, UMD, UIUC, and Flickr material data set (FMD)] using standard protocols. The results demonstrate that our approach outperforms the state of the art in texture classification. The MATLAB code is made available.1 1Matlab Code: http://www.cs.tut.fi/~mehta/symdmd.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Computational Imaging-CI
Contributors: Mehta, R., Egiazarian, K.
Number of pages: 5
Pages: 833-837
Publication date: 1 Jun 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 23
Issue number: 6
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2016): CiteScore 3.54 SJR 0.798 SNIP 1.758
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Keywords: image representation, local features, Rotation invariant features, texture classification
Source: Scopus
Source ID: 84970004897

Research output: Contribution to journalArticleScientificpeer-review

Visual Voice Activity Detection in the Wild

The visual voice activity detection (V-VAD) problem in unconstrained environments is investigated in this paper. A novel method for V-VAD in the wild, exploiting local shape and motion information appearing at spatiotemporal locations of interest for facial video segment description and the bag of words model for facial video segment representation, is proposed. Facial video segment classification is subsequently performed using the state-of-The-Art classification algorithms. Experimental results on one publicly available V-VAD dataset denote the effectiveness of the proposed method, since it achieves better generalization performance in unseen users, when compared to the recently proposed state-of-The-Art methods. Additional results on a new unconstrained dataset provide evidence that the proposed method can be effective even in such cases in which any other existing method fails.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Aristotle University of Thessaloniki
Contributors: Patrona, F., Iosifidis, A., Tefas, A., Nikolaidis, N., Pitas, I.
Number of pages: 11
Pages: 967-977
Publication date: 1 Jun 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Multimedia
Volume: 18
Issue number: 6
ISSN (Print): 1520-9210
Ratings: 
  • Scopus rating (2016): CiteScore 5.4 SJR 1.298 SNIP 2.47
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Media Technology, Computer Science Applications
Keywords: Action Recognition, Bag of Words model, Voice Activity Detection in the wild
Source: Scopus
Source ID: 84971281783

Research output: Contribution to journalArticleScientificpeer-review

Blind sub-Nyquist GNSS signal detection

A satellite navigation receiver traditionally searches for positioning signals using an acquisition procedure. In situations, in which the required information is only a binary decision whether at least one positioning signal is present or absent, the procedure represents an unnecessarily complex solution. This paper presents a different approach for the binary detection problem with significantly reduced computational complexity. The approach is based on a novel decision metric which is utilized to design two binary detectors. The first detector operates under the theoretical assumption of additive white Gaussian noise and is evaluated by means of Receiver Operating Characteristics. The second one considers also additional interferences and is suitable to operate in a real environment. Its performance is verified using a signal captured by a receiver front-end.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, Department of Mathematics, Research group: MAT Intelligent Information Systems Laboratory, Research group: System-on-Chip for GNSS, Wireless Communications and Cyber-Physical Embedded Computing, Advanced Technology Europe, Nottingham Geospatial Institute, University of Nottingham, Chalmers University of Technology
Contributors: Daniel, O., Raasakka, J., Peltola, P., Fröhle, M., Rivero-Rodriguez, A., Wymeersch, H., Nurmi, J.
Number of pages: 5
Pages: 6575-6579
Publication date: 18 May 2016

Host publication information

Title of host publication: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher: IEEE
ISBN (Print): 9781479999880
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: Global Navigation Satellite Systems, indoor/outdoor detection, signal detection

Bibliographical note

EXT="Raasakka, Jussi"
EXT="Peltola, Pekka"

Source: Scopus
Source ID: 84973379949

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

On Renyi's entropy estimation with one-dimensional Gaussian kernels

Rényi's entropies play a significant role in many signal processing applications. Plug-in kernel density estimation methods have been employed to estimate such entropies with good results. However, they become computationally intractable in higher dimensions, because of the requirement to store intermediate probability density values for a large number of data points. We propose a method to reduce the number of the samples in a plug-in kernel density estimation method for Rényi's entropies of real exponents and to improve the result of the standard plug-in kernel density method. To this end, we derive a univariate estimator, using an Hermite expansion of sums of Gaussian kernels and a hierarchical clustering of the samples. On simulated data from a univariate Gaussian distribution, our method performs better than a k-nearest neigbour algorithm and other kernel density estimation methods.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing
Contributors: Sarbu, S.
Number of pages: 5
Pages: 4408-4412
Publication date: 18 May 2016

Host publication information

Title of host publication: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher: IEEE
ISBN (Print): 9781479999880
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: Gaussian kernels, Hermite expansion, hierarchical clustering, Rényi's entropy estimation
Source: Scopus
Source ID: 84973308366

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Supervised subspace learning based on deep randomized networks

In this paper, we propose a supervised subspace learning method that exploits the rich representation power of deep feedforward networks. In order to derive a fast, yet efficient, learning scheme we employ deep randomized neural networks that have been recently shown to provide good compromise between training speed and performance. For optimally determining the learnt subspace, we formulate a regression problem where we employ target vectors designed to encode both the labeling information available for the training data and geometric properties of the training data, when represented in the feature space determined by the network's last hidden layer outputs. We experimentally show that the proposed approach is able to outperform deep randomized neural networks trained by using the standard network target vectors.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Video
Contributors: Iosifidis, A., Gabbouj, M.
Number of pages: 5
Pages: 2584-2588
Publication date: 18 May 2016

Host publication information

Title of host publication: 2016 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
ISBN (Print): 9781479999880
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: Deep Neural Networks, Network targets calculation, Supervised Subspace Learning
Source: Scopus
Source ID: 84973351642

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

A Wideband Front-End Receiver Implementation on GPUs

Modern communication systems have frequency bands that are shared with multiple channels. A typical front-end transceiver is responsible for processing multiple channels simultaneously within the band. Although it is simpler to design a dedicated sub-transceiver for each channel, the overall cost of implementation is prohibitive. It is more desirable to design a single system that can process many channels at the same time. However, it is difficult to realize such a transceiver in wireless communication systems, since it needs to accommodate different standards, data rates, sampling rates, etc. In this paper, we address the challenges of flexible, cost-effective, multi-channel implementation of wideband receiver systems. In particular, we develop a novel implementation by applying graphics processing unit (GPU) technology to a wideband receiver that processes multiple channels at the same time, including channelization and arbitrary resampling. Our proposed new receiver architecture is flexible and reconfigurable via software, while providing high throughput and low latency.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Pervasive Computing, Research area: Computer engineering, University of Maryland, Department of Electrical and Computer Engineering
Contributors: Kim, S. C., Bhattacharyya, S. S.
Number of pages: 11
Pages: 2602-2612
Publication date: 15 May 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 64
Issue number: 10
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2016): CiteScore 5.54 SJR 1.385 SNIP 2.533
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Arbitrary resampler, carrier aggregation, GPU front-end receiver, GPU-based radio, Polyphase channelizer
Source: Scopus
Source ID: 84968750704

Research output: Contribution to journalArticleScientificpeer-review

Coordinating proactive social devices in a mobile cloud: Lessons learned and a way forward

Recent device shipment trends strongly indicate that the number of Web-enabled devices other than PCs and smart phones are growing rapidly. Marking the end of the dominant era of these two traditional device categories, people will soon commonly use various types of Internet-connected devices in their daily lives, where no single device will dominate. Since today's devices are mostly standalone and only stay in sync in limited ways, new approaches are needed for mastering the complexity arising from the world of many types of devices, created by different manufacturers and implementing competing standards. Today, the most common denominator for dealing with the differences is using clouds. Unfortunately, however, while the cloud is well suited for numerous activities, there are also serious limitations, especially when considering systems that consist of numerous, battery-powered computing devices that have limited connectivity. In this paper, we provide an insight to our research where totally cloud-based orchestration of cooperating devices is partitioned into more local actions, where constant communication with the cloud backend can be at least partially omitted.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Pervasive Computing, Research area: Software engineering
Contributors: Mäkitalo, N., Aaltonen, T., Mikkonen, T.
Number of pages: 10
Pages: 179-188
Publication date: 14 May 2016

Host publication information

Title of host publication: MOBILESoft '16 Proceedings of the International Conference on Mobile Software Engineering and Systems
Publisher: ACM
ISBN (Electronic): 9781450341783
ASJC Scopus subject areas: Control and Systems Engineering, Software, Computer Science Applications, Signal Processing
Keywords: Cloud computing, Cyber-physical systems, Internet of things, Mobile cloud, Multi-device ownership, Multi-device programming
Source: Scopus
Source ID: 84983554842

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Statistical analysis of interference for nanoscale electromechanical wireless communication at VHF-band

Nanoscale electromechanical wireless communication with on-off keying in the very high frequency (VHF) band (30-300 MHz) is studied for a receiver using a carbon nanotube (CNT). Previous studies on this topic have only considered continuous wave (CW) on-off keying which suffers from spectral widening due to sharp changes in the signal. Effects of the inter-symbol interference (ISI), the co-channel interference, and the adjacent channel interference on the received signal statistics have not been analyzed. The rise- and fall-times associated with the filtering of the incoming signal by the mechanical frequency response of the receiver's CNT have also been ignored. In this paper, Fourier-series based modeling and statistical analysis of decision variables are performed. The results and modeling in this study enable performance evaluation of CNT based receivers with an arbitrary number of interfering signals with arbitrary pulse shapes, and fully incorporates the transient signal components. Received signal statistics under interference are derived using the developed model. Numerical results are presented for Hanning pulse and trapezoidal pulse (which includes rectangular pulses corresponding to CW as a special case). The required guard intervals between pulses to mitigate ISI, required frequency separation between channels, and required spatial separation of co-channel networks (frequency reuse distance) are shown. These results show that large frequency reuse distance is required, limiting efficient spectrum utilization. However, the ISI and adjacent channel interference can be controlled more easily with a proper selection of parameters.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Emerging Technologies for Nano-Bio-Info-Cogno, University of Oulu, Georgia Institute of Technology, School of Electrical and Computer Engineering, Nano Communication Centre, Aristotle University of Thessaloniki, School of Electrical and Computer Engineering
Contributors: Lehtomäki, J. J., Bicen, A. O., Akyildiz, I. F.
Number of pages: 11
Pages: 2040-2050
Publication date: 15 Apr 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 64
Issue number: 8
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2016): CiteScore 5.54 SJR 1.385 SNIP 2.533
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Carbon nanotube, frequency division multiple access, nanodevice, nanonetwork, on-off keying, VHF-band, wireless communication

Bibliographical note

INT=elt,"Bicen, A. Ozan"

Source: Scopus
Source ID: 84964402214

Research output: Contribution to journalArticleScientificpeer-review

On prediction of DCT-based denoising efficiency under spatially correlated noise conditions

In this paper, results of image denoising efficiency prediction for filter based on discrete cosine transform (DCT) for the case of spatially correlated additive Gaussian Noise (SCGN) are given. The considered noise model is analyzed for different degrees of spatial correlation that produce varying non-homogeneous spectrum of the noise. PSNR metric is exploited to assess denoising efficiency. It is shown in this paper, that a prediction of denoising efficiency has high accuracy for data distorted by noise with different degrees of spatial correlation, and require low computational resources.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Computational Imaging-CI, Kharkiv National Aerospace University
Contributors: Rubel, O., Lukin, V., Egiazarian, K.
Number of pages: 5
Pages: 750-754
Publication date: 12 Apr 2016

Host publication information

Title of host publication: 2016 13th International Conference on Modern Problems of Radio Engineering, Telecommunications and Computer Science (TCSET)
Publisher: IEEE
ISBN (Print): 9786176078067
ASJC Scopus subject areas: Computer Networks and Communications, Electrical and Electronic Engineering, Instrumentation, Radiation, Computer Science (miscellaneous), Signal Processing, Modelling and Simulation
Keywords: Denoising, Efficiency Prediction, Fitting, Spatially Correlated Noise
Source: Scopus
Source ID: 84969277857

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

BL-LDA: Bringing bigram to supervised topic model

With the increasing amount of data being published on the Web, it is difficult to analyze their content within a short time. Topic modeling techniques can summarize textual data that contains several topics. Both the label (such as category or tag) and word co-occurrence play a significant role in understanding textual data. However, many conventional topic modeling techniques are limited to the bag-of-words assumption. In this paper, we develop a probabilistic model called Bigram Labeled Latent Dirichlet Allocation (BL-LDA), to address the limitation of the bag-of-words assumption. The proposed BL-LDA incorporates the bigram into the Labeled LDA (L-LDA) technique. Extensive experiments on Yelp data show that the proposed scheme is better than the L-LDA in terms of accuracy.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Research Community on Data-to-Decision (D2D), Korea University
Contributors: Park, Y., Alam, M. H., Ryu, W. J., Lee, S.
Number of pages: 6
Pages: 83-88
Publication date: 2 Mar 2016

Host publication information

Title of host publication: Proceedings - 2015 International Conference on Computational Science and Computational Intelligence, CSCI 2015
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7424068
ISBN (Electronic): 9781467397957
ASJC Scopus subject areas: Computational Theory and Mathematics, Artificial Intelligence, Computer Networks and Communications, Hardware and Architecture, Signal Processing
Keywords: Data Analysis, Data Mining, Text Classification, Topic Modeling
Source: Scopus
Source ID: 84964476038

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Which wavelength is the best for arterial pulse waveform extraction using laser speckle imaging?

A multi-wavelengths analysis for pulse waveform extraction using laser speckle is conducted. The proposed system consists of three coherent light sources (532 nm, 635 nm, 850 nm). A bench-test composed of a moving skin-like phantom (silicone membrane) is used to compare the results obtained from different wavelengths. The system is able to identify a skin-like phantom vibration frequency, within physiological values, with a minimum error of 0.5 mHz for the 635 nm and 850 nm wavelengths and a minimum error of 1.3 mHz for the 532 nm light wavelength using a FFT-based algorithm. The phantom velocity profile is estimated with an error ranging from 27% to 9% using a bidimensional correlation coefficient-based algorithm. An in vivo trial is also conducted, using the 532 nm and 635 nm laser sources. The 850 nm light source has not been able to extract the pulse waveform. The heart rate is identified with a minimum error of 0.48 beats per minute for the 532 nm light source and a minimal error of 1.15 beats per minute for the 635 nm light source. Our work reveals that a laser speckle-based system with a 532 nm wavelength is able to give arterial pulse waveform with better results than those given with a 635 nm laser.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Computational Biophysics and Imaging Group, LARIS-Laboratoire Angevin de Recherche en Ingénierie des Systèmes, Center for Mathematics Probability and Statistics
Contributors: Vaz, P., Pereira, T., Figueiras, E., Correia, C., Humeau-Heurtier, A., Cardoso, J.
Number of pages: 8
Pages: 188-195
Publication date: Mar 2016
Peer-reviewed: Yes

Publication information

Journal: Biomedical Signal Processing and Control
Volume: 25
ISSN (Print): 1746-8094
Ratings: 
  • Scopus rating (2016): CiteScore 2.82 SJR 0.674 SNIP 1.69
Original language: English
ASJC Scopus subject areas: Health Informatics, Signal Processing
Keywords: Arterial pulse waveform, Correlation, Fast Fourier transform, Laser speckle, Multi-spectral
Source: Scopus
Source ID: 84950341719

Research output: Contribution to journalArticleScientificpeer-review

Digital full-band linearization of wideband direct-conversion receiver for radar and communications applications

This paper proposes a fully digital post-processing solution for cancelling nonlinear distortion and mirror-frequency interference in wideband direct-conversion receivers (DCRs). Favorable cost, integrability, and power efficiency have made DCRs a popular choice in communication systems. It is also an emerging trend in radar systems since digital post-processing enables sufficient performance. The proposed method cancels the most essential distortion adaptively during normal receiver operation without any prior information. Improved cancellation performance compared to the state-of-the-art is achieved considering inband and neighboring band distortion induced by the strong received signals. This is verified and demonstrated with extensive simulations and true RF hardware measurements.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Airbus Group, Airbus Defence and Space GmbH
Contributors: Allen, M., Marttila, J., Valkama, M., Singh, S., Epp, M., Schlecker, W.
Number of pages: 8
Pages: 1361-1368
Publication date: 26 Feb 2016

Host publication information

Title of host publication: 2015 49th Asilomar Conference on Signals, Systems and Computers
Publisher: IEEE COMPUTER SOCIETY PRESS
ISBN (Print): 9781467385763
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing
Keywords: Direct-conversion receiver, interference cancellation, nonlinear distortion, nonlinearity modeling, radar
Electronic versions: 

Bibliographical note

EXT="Singh, Simran"

Source: Scopus
Source ID: 84969781278

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Random Value Impulse Noise Removal Based on Most Similar Neighbors

A novel filter based on four most similar neighbors (MSN) is proposed in this paper which considers all the pixels of the sliding window except the central pixel after taking the first order absolute differences from the central pixel. The proposed filter is composed of two steps: noise detection followed by filtering. In noise detection, first order absolute differences are calculated and sorted in ascending order. Clusters of equal sizes are formed based on most similar pixels and then fuzzy rules are applied to detect the noise present in the current pixel. Threshold parameters are set adaptively. In filtering phase, median based fuzzy filter is used to restore the corrupted pixels. Experimental results show that the proposed filter outperforms several state-of-the-art filers for random value impulse noise removal in an image.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, International Islamic University Islamabad
Contributors: Habib, M., Rasheed, S., Hussain, A., Ali, M.
Number of pages: 5
Pages: 329-333
Publication date: 26 Feb 2016

Host publication information

Title of host publication: 2015 13th International Conference on Frontiers of Information Technology (FIT)
Publisher: IEEE
ISBN (Print): 9781467396660
ASJC Scopus subject areas: Health Informatics, Computer Science Applications, Signal Processing
Keywords: fuzzy logic, Image processing, impulse noise, noise removal

Bibliographical note

INT=elt,"Ali, Mubashir"

Source: Scopus
Source ID: 84964689604

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Innovative e-Tourism Services on Top of Geo2Tag LBS Platform

Location based and geo-context aware services form the new fast growing domain of commercially successful ICT solutions. These services play the key role in IoT scenarios and development of smart spaces and proactive solutions. One of the most attractive application areas is e-Tourism. More people can afford travelling and over the last few decades we see continues growth of the tourist activity. At the same time we see huge increase of demand both in quantity and quality of tourist services. Many experts foresee that this growth cannot any longer be fulfilled by applying traditional approaches. Similarly to the change in tickets and hotel booking, it is expected that soon we will witness major transformation in the whole industry towards e-Tourism driven market, where roles of traditional service providers, e.g., tourist agents, guides, will disappear or seriously changed. Internet of Things (IoT) is an integral part of the Future Internet ecosystem that has major impact on development of e-Tourism services. IoT provides an infrastructure to uniquely identify and link physical objects with virtual representations. As a result any physical object can have virtual reflection in the service space. This gives an opportunity to replace actions on physical objects by operations on their virtual reflections, which is faster, cheaper and more comfortable for the user. In this paper we summarize our research in the field, share ideas of innovative e-Tourism services and present Geo2Tag LBS platform that allows easy and fast development of such services.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, Research group: Emerging Technologies for Nano-Bio-Info-Cogno, FRUCT Oy, Laboratory of Information Science and Semantic Technologies, ITMO University
Contributors: Balandina, E., Balandin, S., Koucheryavy, Y., Mouromtsev, D.
Number of pages: 8
Pages: 752-759
Publication date: 5 Feb 2016

Host publication information

Title of host publication: Proceedings - 11th International Conference on Signal-Image Technology and Internet-Based Systems, SITIS 2015
Publisher: IEEE
ISBN (Print): 9781467397216
ASJC Scopus subject areas: Signal Processing, Computer Networks and Communications, Information Systems
Keywords: e-Tourism, Geo2Tag, ICT for Cultural Heritage, Internet of Things, IoT, Tourist services, Use Cases

Bibliographical note

EXT="Balandin, Sergey"

Source: Scopus
Source ID: 84966375788

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Guest editorial special issue on the internet of nano things

The six papers in this special section focus on the Internet of nanotechnology things. While researchers are currently investigating these challenges to develop fully functional nano communication systems, a question remains as to whether they can represent an extended communication network that is part of the broader Internet. These papers address new solutions for the Internet of Nano Things. The Internet of Things paradigm has transformed the way we operate our personal and professional lives, it is driving our economy and will continue to enable many new opportunities in broad research areas. As this pervasive and ubiquitous interconnection of our everyday life appliances continues into the future, new types of devices enabled by nano and biotechnology promise to push engineering to previously unexplored application domains, where the exchange of information and access from/to the broader Internet for their monitoring and control are even more essential. The research on nanoscale communication and networks aims to develop systems for interconnecting these novel devices at the nanoscale, i.e., the Internet of Nano Things.

General information

Publication status: Published
MoE publication type: B1 Article in a scientific magazine
Organisations: Department of Electronics and Communications Engineering, Research group: Emerging Technologies for Nano-Bio-Info-Cogno, Nano Communication Centre, State University of New York, University of Nebraska - Lincoln
Contributors: Balasubramaniam, S., Jornet, J. M., Pierobon, M., Koucheryavy, Y.
Number of pages: 3
Pages: 1-3
Publication date: 1 Feb 2016
Peer-reviewed: No

Publication information

Journal: IEEE Internet of Things Journal
Volume: 3
Issue number: 1
ISSN (Print): 2327-4662
Ratings: 
  • Scopus rating (2016): CiteScore 9.9 SJR 1.447 SNIP 5.896
Original language: English
ASJC Scopus subject areas: Computer Networks and Communications, Computer Science Applications, Hardware and Architecture, Information Systems, Signal Processing, Information Systems and Management
Source: Scopus
Source ID: 84959329564

Research output: Contribution to journalArticleScientific

Importance of maintenance data quality in extended warranty simulation

As manufacturing industries are transforming towards service orientation, predicting the costs of product-service systems is becoming essential. Simulation is one possibility for evaluating the costs and risks involved in product-service systems, such as extended warranty agreements. We conducted a case study with a globally operating manufacturer of industrial goods who also provides services for the equipment. We created equipment performance simulation (EPSi) models and a tool, EPSitor, for using the models in predicting extended warranty costs. However, reliable simulation results require good quality maintenance and operation data from existing installations. We discovered that it is difficult to collect the data needed for simulations and there were many challenges with data quality. Quality problems were mainly observed in manually collected data. Insufficient data quality leads to a wider margin of error in the simulation models, which increases business risk. Identifying these challenges is the first step in transforming the data collection routines to support equipment performance simulations. The key to long-term business benefits of simulation is to acknowledge the importance of data quality and to establish efficient data collection routines. Future research should find ways to motivate maintenance technicians to collect good quality data. This would contribute to more accurate cost analysis and thus to better profitability of extended warranty contracts.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Mechanical Engineering and Industrial Systems, Research group: Käyttövarmuuden suunnittelu ja kunnossapito, Research area: Life-cycle Management, Aalto University
Contributors: Mahlamäki, K., Niemi, A., Jokinen, J., Borgman, J.
Number of pages: 8
Pages: 3-10
Publication date: 1 Jan 2016
Peer-reviewed: Yes

Publication information

Journal: International Journal of COMADEM
Volume: 19
Issue number: 1
ISSN (Print): 1363-7681
Ratings: 
  • Scopus rating (2016): CiteScore 0.11 SJR 0.128 SNIP 0.163
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Industrial and Manufacturing Engineering, Safety, Risk, Reliability and Quality, Bioengineering, Signal Processing, Strategy and Management
Keywords: Asset management, Data quality, Hitman factors

Bibliographical note

INT=mei,”Jokinen, Juuso”

Source: Scopus
Source ID: 84960940492

Research output: Contribution to journalArticleScientificpeer-review

Accurate depth estimation from a sequence of monocular images supported by proprioceptive sensors

This paper describes an extended Kalman filter based algorithm for fusion of monocular vision measurements, inertial rate sensor measurements, and camera motion. The motion of the camera between successive images generates a baseline for range computations by triangulation. The recursive estimation algorithm is based on extended Kalman filtering. The depth estimation accuracy is strongly affected by mutual observer and feature point geometry, measurement accuracy of observer motion parameters and line of sight to a feature point. The simulation study investigates how the estimation accuracy is affected by the following parameters: linear and angular velocity measurement errors, camera noise, and observer path. These results draw requirements to the instrumentation and observation scenarios. It was found that under favorable conditions the error in distance estimation does not exceed 2% of the distance to a feature point.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Automation Science and Engineering, Research area: Dynamic Systems, Research area: Measurement Technology and Process Control, Research group: Positioning
Contributors: Davidson, P., Raunio, J. P., Piché, R.
Number of pages: 9
Pages: 249-257
Publication date: 2016

Host publication information

Title of host publication: 23rd Saint Petersburg International Conference on Integrated Navigation Systems, ICINS 2016 - Proceedings
Publisher: State Research Center of the Russian Federation
ISBN (Electronic): 9785919950370
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Electrical and Electronic Engineering, Information Systems
Keywords: Computer vision, Gyroscope, IMU, Odometer, Structure from motion
Source: Scopus
Source ID: 84979573597

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Computer vision aided navigation systems

The paper considers the possible use of computer vision systems for INS aiding. Two methods of navigation data obtaining from the image sequence are analyzed. The first method uses the features of architectural elements in indoor and urban conditions for generation of object attitude parameters. The second method is based on extraction of general features in the image and is more widely applied. Besides the orientation parameters, the second method estimates the object displacement, and thus can be used as visual odometry technique. The described algorithms can be used to develop small-sized MEMS navigation systems efficiently operating in urban conditions.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Automation Science and Engineering, Research group: Positioning, ITMO University
Contributors: Davidson, P., Merkulova, I.
Number of pages: 3
Pages: 560-562
Publication date: 2016

Host publication information

Title of host publication: 23rd Saint Petersburg International Conference on Integrated Navigation Systems, ICINS 2016 - Proceedings
Publisher: State Research Center of the Russian Federation
ISBN (Electronic): 9785919950370
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Electrical and Electronic Engineering, Information Systems
Keywords: Camera, Computer vision, Data fusion, Image processing, Inertial system, Orientation
Source: Scopus
Source ID: 84979499890

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Dominant Rotated Local Binary Patterns (DRLBP) for texture classification

In this paper, we present a novel rotation-invariant and computationally efficient texture descriptor called Dominant Rotated Local Binary Pattern (DRLBP). A rotation invariance is achieved by computing the descriptor with respect to a reference in a local neighborhood. A reference is fast to compute maintaining the computational simplicity of the Local Binary Patterns (LBP). The proposed approach not only retains the complete structural information extracted by LBP, but it also captures the complimentary information by utilizing the magnitude information, thereby achieving more discriminative power. For feature selection, we learn a dictionary of the most frequently occurring patterns from the training images, and discard redundant and non-informative features. To evaluate the performance we conduct experiments on three standard texture datasets: Outex12, Outex 10 and KTH-TIPS. The performance is compared with the state-of-the-art rotation invariant texture descriptors and results show that the proposed method is superior to other approaches.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Computational Imaging-CI
Contributors: Mehta, R., Egiazarian, K.
Number of pages: 7
Pages: 16-22
Publication date: 2016
Peer-reviewed: Yes
Early online date: 30 Nov 2015

Publication information

Journal: Pattern Recognition Letters
Volume: 71
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2016): CiteScore 2.9 SJR 0.729 SNIP 1.646
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Feature Selection, KTH-TIPS, Local Binary Pattern (LBP), Outex, Rotation Invariance, Texture Classification
Source: Scopus
Source ID: 84951106920

Research output: Contribution to journalArticleScientificpeer-review

Efficiency of texture image filtering and its prediction

Textures are typical elements of natural scene images widely used in pattern recognition and image classification. Noise, often being present in acquired images, deteriorates texture features (characteristics), and it is desirable both to suppress it and to preserve a texture. This task is quite difficult even for the most advanced filters, and the resulting denoising efficiency can be quite low. Due to this, it is desirable to predict a denoising efficiency before filtering to decide whether it is worth filtering a given image or not. In this paper, we analyze several quantitative criteria (metrics) that can characterize filtering efficiency. Prediction strategy is described and its accuracy is studied. Several modern filtering techniques are analyzed and compared. Based on this, practical recommendations are given.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Kharkiv National Aerospace University, Universite de Rennes, Instituto Politecnico Nacional
Contributors: Rubel, O., Lukin, V., Abramov, S., Vozel, B., Egiazarian, K., Pogrebnyak, O.
Number of pages: 8
Pages: 1543–1550
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: Signal, Image and Video Processing
Volume: 10
Issue number: 8
ISSN (Print): 1863-1703
Ratings: 
  • Scopus rating (2016): CiteScore 1.35 SJR 0.373 SNIP 1.118
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Filtering efficiency, Image enhancement, Noise suppression, Visual quality
Source: Scopus
Source ID: 84983806396

Research output: Contribution to journalArticleScientificpeer-review

IEEE 802.11ac MIMO Transceiver Baseband Processing on a VLIW Processor

Wireless standards are evolving rapidly due to the exponential growth in the number of portable devices along with the applications with high data rate requirements. Adaptable software based signal processing implementations for these devices can make the deployment of the constantly evolving standards faster and less expensive. The flagship technology from the IEEE WLAN family, the IEEE 802.11ac, aims at achieving very high throughputs in local area connectivity scenarios. This article presents a software based implementation for the Multiple Input and Multiple Output (MIMO) transmitter and receiver baseband processing conforming to the IEEE 802.11ac standard which can achieve transmission bit rates beyond 1Gbps. This work focuses on the Physical layer frequency domain processing. Various configurations, including 2×2 and 4×4 MIMO are considered for the implementation. To utilize the available data and instruction level parallelism, a DSP core with vector extensions is selected as the implementation platform. Then, the feasibility of the presented software-based solution is assessed by studying the number of clock cycles and power consumption of the different scenarios implemented on this core. Such Software Defined Radio based approaches can potentially offer more flexibility, high energy efficiency, reduced design efforts and thus shorter time-to-market cycles in comparison with the conventional fixed-function hardware methods.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Department of Pervasive Computing, Research area: Computer engineering
Contributors: Aghababaeetafreshi, M., Lehtonen, L. K., Levanen, T., Valkama, M., Takala, J.
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2016): CiteScore 0.78 SJR 0.212 SNIP 0.65
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Modelling and Simulation, Hardware and Architecture, Information Systems, Signal Processing, Theoretical Computer Science
Keywords: MIMO, OFDM, Parallel processing, Software defined radio, VLIW, WLAN

Bibliographical note

ORG=elt,0.5
ORG=tie,0.5

Source: Scopus
Source ID: 84942023616

Research output: Contribution to journalArticleScientificpeer-review

Instrumentation-Driven Validation of Dataflow Applications

Dataflow modeling offers a myriad of tools for designing and optimizing signal processing systems. A designer is able to take advantage of dataflow properties to effectively tune the system in connection with functionality and different performance metrics. However, a disparity in the specification of dataflow properties and the final implementation can lead to incorrect behavior that is difficult to detect. This motivates the problem of ensuring consistency between dataflow properties that are declared or otherwise assumed as part of dataflow-based application models, and the dataflow behavior that is exhibited by implementations that are derived from the models. In this paper, we address this problem by introducing a novel dataflow validation framework (DVF) that is able to identify disparities between an application’s formal dataflow representation and its implementation. DVF works by instrumenting the implementation of an application and monitoring the instrumentation data as the application executes. This monitoring process is streamlined so that DVF achieves validation without major overhead. We demonstrate the utility of our DVF through design and implementation case studies involving an automatic speech recognition application, a JPEG encoder, and an acoustic tracking application.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Pervasive Computing, Research area: Computer engineering, Signal Processing Research Community (SPRC), University of Maryland, Technische Universitat Munchen, Institute for Advanced Computer Studies
Contributors: Chukhman, I., Jiao, Y., Salem, H. B., Bhattacharyya, S. S.
Pages: 383–397
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: Journal of Signal Processing Systems
Volume: 84
Issue number: 3
ISSN (Print): 1939-8018
Ratings: 
  • Scopus rating (2016): CiteScore 0.78 SJR 0.212 SNIP 0.65
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Modelling and Simulation, Hardware and Architecture, Information Systems, Signal Processing, Theoretical Computer Science
Keywords: Dataflow graphs, Design validation, Models of computation, Signal processing systems
Source: Scopus
Source ID: 84946128443

Research output: Contribution to journalArticleScientificpeer-review

Online tests of Kalman filter consistency

The normalised innovation squared (NIS) test, which is used to assess whether a Kalman filter's noise assumptions are consistent with realised measurements, can be applied online with real data, and does not require future data, repeated experiments or knowledge of the true state. In this work, it is shown that the NIS test is equivalent to three other model criticism procedures, which are as follows: (i) it can be derived as a Bayesian p-test for the prior predictive distribution; (ii) as a nested-model parameter significance test; and (iii) from a recently-proposed filter residual test. A new NIS-like test corresponding to a posterior predictive Bayesian p-test is presented.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Automation Science and Engineering, Research group: Positioning
Contributors: Piché, R.
Pages: 115–124
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: International Journal of Adaptive Control and Signal Processing
Volume: 30
Issue number: 1
ISSN (Print): 0890-6327
Ratings: 
  • Scopus rating (2016): CiteScore 2.04 SJR 0.749 SNIP 1.026
Original language: English
ASJC Scopus subject areas: Control and Systems Engineering, Electrical and Electronic Engineering, Signal Processing
Keywords: Kalman filter, Model consistency, Normalised innovations squared, Predictive distribution
Electronic versions: 
Source: Scopus
Source ID: 84954027695

Research output: Contribution to journalArticleScientificpeer-review

OpenCL Programmable Exposed Datapath High Performance Low-Power Image Signal Processor

Sophisticated computational imaging algorithms require both high performance and good energy-efficiency when executed on mobile devices. Recent trend has been to exploit the abundant data-level parallelism found in general purpose programmable GPUs. However, for low-power mobile use cases, generic GPUs consume excessive amounts of power. This paper proposes a programmable computational imaging processor with 16-bit half-precision SIMD floating point vector processing capabilities combined with power efficiency of an exposed datapath. In comparison to traditional VLIW architectures with similar computational resources, the exposed datapath reduces the register file traffic and complexity. These and the specific optimizations enabled by the explicit programming model enable extremely good power-performance. When synthesized on a 28nm ASIC technology, the accelerator consumes 71mW of power while running a state-of-the-art denoising algorithm, and occupies only 0.2mm² of chip area. For the algorithm, energy usage per frame is 7mJ, which is 10x less than the best found GPU-based implementation.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Pervasive Computing, Research area: Computer engineering, Noiseless Imaging Oy Ltd
Contributors: Multanen, J., Kultala, H., Koskela, M., Viitanen, T., Jääskeläinen, P., Takala, J., Danielyan, A., Cruz, C.
Number of pages: 6
Publication date: 2016

Host publication information

Title of host publication: 2016 IEEE Nordic Circuits and Systems Conference (NORCAS)
Publisher: IEEE
ISBN (Electronic): 978-1-5090-1095-0
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing
Electronic versions: 

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Still image/video frame lossy compression providing a desired visual quality

The problem of how to automatically provide a desired (required) visual quality in lossy compression of still images and video frames is considered in this paper. The quality can be measured based on different conventional and visual quality metrics. In this paper, we mainly employ human visual system (HVS) based metrics PSNR-HVS-M and MSSIM since both of them take into account several important peculiarities of HVS. To provide a desired visual quality with high accuracy, iterative image compression procedures are proposed and analyzed. An experimental study is performed for a large number of grayscale test images. We demonstrate that there exist several coders for which the number of iterations can be essentially decreased using a reasonable selection of the starting value and the variation interval for the parameter controlling compression (PCC). PCC values attained at the end of the iterative procedure may heavily depend upon the coder used and the complexity of the image. Similarly, the compression ratio also considerably depends on the above factors. We show that for some modern coders that take HVS into consideration it is possible to give practical recommendations on setting a fixed PCC to provide a desired visual quality in a non-iterative manner. The case when original images are corrupted by visible noise is also briefly studied.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Computational Imaging-CI, National Aerospace University
Contributors: Zemliachenko, A., Lukin, V., Ponomarenko, N., Egiazarian, K., Astola, J.
Pages: 697-718
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: Multidimensional Systems and Signal Processing
Volume: 27
Issue number: 3
ISSN (Print): 0923-6082
Ratings: 
  • Scopus rating (2016): CiteScore 1.47 SJR 0.424 SNIP 1.123
Original language: English
ASJC Scopus subject areas: Computer Science Applications, Information Systems, Signal Processing, Software, Artificial Intelligence, Hardware and Architecture, Applied Mathematics
Keywords: Compression ratio, Lossy compression, Required quality, Visual quality metrics
Source: Scopus
Source ID: 84930357751

Research output: Contribution to journalArticleScientificpeer-review

Unbiased Injection of Signal-Dependent Noise in Variance-Stabilized Range

The design, optimization, and validation of many image processing or image-based analysis systems often requires testing of the system performance over a dataset of images corrupted by noise at different signal-to-noise ratio regimes. A noise-free ground-truth image may not be available, and different SNRs are simulated by injecting extra noise into an already noisy image. However, noise in real-world systems is typically signal-dependent, with variance determined by the noise-free image. Thus, also the noise to be injected shall depend on the unknown ground-truth image. To circumvent this issue, we consider the additive injection of noise in variance-stabilized range, where no previous knowledge of the ground-truth signal is necessary. Specifically, we design a special noise-injection operator that prevents the errors on expectation and variance that would otherwise arise when standard variance-stabilizing transformations are used for this task. Thus, the proposed operator is suitable for accurately injecting signal-dependent noise even to images acquired at very low counts.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Signal and Image Restoration-RST, Department of Applied Physics
Contributors: Borges, L., Vieira, M., Foi, A.
Pages: 1494-1498
Publication date: 2016
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 23
Issue number: 10
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2016): CiteScore 3.54 SJR 0.798 SNIP 1.758
Original language: English
ASJC Scopus subject areas: Signal Processing, Applied Mathematics, Electrical and Electronic Engineering
Keywords: Anscombe transformation, Noise injection, optimization, Poisson noise, variance stabilization
Electronic versions: 
Source: Scopus
Source ID: 85027556734

Research output: Contribution to journalArticleScientificpeer-review

BM3D image denoising using heterogeneous computing platforms

Noise reduction is often performed at an early stage of the image processing path. In order to keep the processing delays small in different computing platforms, it is important that the noise reduction is performed swiftly. In this paper, the block-matching and three-dimensional filtering (BM3D) denoising algorithm is implemented on heterogeneous computing platforms using OpenCL and CUDA frameworks. To our knowledge, these implementations are the first successful open source attempts to use GPU computation for BM3D denoising. The presented GPU implementations are up to 7.5 times faster than their respective CPU implementations. At the same time, the experiments illustrate general design challenges in using massively parallel processing platforms for the calculation of complex imaging algorithms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing Research Community (SPRC), Univ of Oulu, Center for Machine Vision Research
Contributors: Sarjanoja, S., Boutellier, J., Hannuksela, J.
Publication date: 28 Dec 2015

Host publication information

Title of host publication: DASIP 2015 - Proceedings of the 2015 Conference on Design and Architectures for Signal and Image Processing
Volume: 2015-December
Publisher: IEEE COMPUTER SOCIETY PRESS
Article number: 7367257
ISBN (Electronic): 9791092279108
ASJC Scopus subject areas: Computer Graphics and Computer-Aided Design, Computer Vision and Pattern Recognition, Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Keywords: Image denoising, Mobile computing, Parallel algorithms, Parallel processing
Source: Scopus
Source ID: 84959887479

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Using multi-step proposal distribution for improved MCMC convergence in Bayesian network structure learning

Bayesian networks have become popular for modeling probabilistic relationships between entities. As their structure can also be given a causal interpretation about the studied system, they can be used to learn, for example, regulatory relationships of genes or proteins in biological networks and pathways. Inference of the Bayesian network structure is complicated by the size of the model structure space, necessitating the use of optimization methods or sampling techniques, such Markov Chain Monte Carlo (MCMC) methods. However, convergence of MCMC chains is in many cases slow and can become even a harder issue as the dataset size grows. We show here how to improve convergence in the Bayesian network structure space by using an adjustable proposal distribution with the possibility to propose a wide range of steps in the structure space, and demonstrate improved network structure inference by analyzing phosphoprotein data from the human primary T cell signaling network.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Aalto University, Department of Computer Science and Information Systems
Contributors: Larjo, A., Lähdesmäki, H.
Publication date: 27 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Bioinformatics and Systems Biology
Volume: 2015
Issue number: 1
Article number: 6
ISSN (Print): 1687-4145
Ratings: 
  • Scopus rating (2015): CiteScore 0.68 SJR 0.314 SNIP 0.381
Original language: English
ASJC Scopus subject areas: Medicine(all), Computer Science(all), Signal Processing, Statistics and Probability, General
Keywords: Bayesian network, MCMC, Proposal distribution, Structure learning

Bibliographical note

EXT="Lähdesmäki, Harri"

Source: Scopus
Source ID: 84932633872

Research output: Contribution to journalArticleScientificpeer-review

Energy cooperation for throughput optimization based on save-then-transmit protocol in wireless communication system

Green communication and energy saving have been a critical issue in modern wireless communication systems. The concepts of energy harvesting and energy transfer are recently receiving much attention in academic research field. In this paper, we study energy cooperation problems based on save-then-transmit protocol and propose two energy cooperation schemes for different system models: two-node communication model and three-node relay communication model. In both models, all of the nodes transmitting information have no fixed energy supplies and gain energy only via wireless energy harvesting from nature. Besides, these nodes also follow a save-then-transmit protocol. Namely, for each timeslot, a fraction (referred to as save-ratio) of time is devoted exclusively to energy harvesting while the remaining fraction is used for data transmission. In order to maximize the system throughput, energy transfer mechanism is introduced in our schemes, i.e., some nodes are permitted to share their harvested energy with other nodes by means of wireless energy transfer. Simulation results demonstrate that our proposed schemes can outperform both the schemes with half-allocate save-ratio and the schemes without energy transfer in terms of throughput performance, and also characterize the dependencies of system throughput, transferred energy, and save-ratio on energy harvesting rate.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Wireless Communications and Positioning (WICO), Chongqing Key Lab of Mobile Communication Technology, Chongqing University of Posts and Telecommunications
Contributors: Dai, C., Li, F., Renfors, M.
Publication date: 26 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Wireless Communications and Networking
Volume: 2015
Issue number: 1
Article number: 119
ISSN (Print): 1687-1472
Ratings: 
  • Scopus rating (2015): CiteScore 0.96 SJR 0.362 SNIP 1.138
Original language: English
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Computer Science Applications
Keywords: Energy harvesting, Energy transfer, Save-then-transmit protocol, Throughput optimization
Source: Scopus
Source ID: 84930207810

Research output: Contribution to journalArticleScientificpeer-review

Generative part-based Gabor object detector

Discriminative part-based models have become the approach for visual object detection. The models learn from a large number of positive and negative examples with annotated class labels and location (bounding box). In contrast, we propose a part-based generative model that learns from a small number of positive examples. This is achieved by utilizing "privileged information", sparse class-specific landmarks with semantic meaning. Our method uses bio-inspired complex-valued Gabor features to describe local parts. Gabor features are transformed to part probabilities by unsupervised Gaussian Mixture Model (GMM). GMM estimation is robustified for a small amount of data by a randomization procedure inspired by random forests. The GMM framework is also used to construct a probabilistic spatial model of part configurations. Our detector is invariant to translation, rotation and scaling. On part level invariance is achieved by pose quantization which is more efficient than previously proposed feature transformations. In the spatial model, invariance is achieved by mapping parts to an "aligned object space". Using a small number of positive examples our generative method performs comparably to the state-of-the-art discriminative method.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Vision, Field robotics for efficient work sites (FIRE), Prostate cancer research center (PCRC), Lappeenranta University of Technology
Contributors: Riabchenko, E., Kämäräinen, J.
Number of pages: 8
Pages: 1-8
Publication date: 15 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition Letters
Volume: 68
Issue number: P1
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2015): CiteScore 2.87 SJR 0.95 SNIP 1.979
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Gabor feature, Gaussian mixture model, Generative learning, Object detection, Visual classification

Bibliographical note

EXT="Riabchenko, Ekaterina"

Source: Scopus
Source ID: 84941570575

Research output: Contribution to journalArticleScientificpeer-review

Whole-cell morphological properties of neurons constrain the nonrandom features of network connectivity

We addressed the principles of micro-level organization of neuronal circuits and explored how the neuronal morphology constrains this organization. Several studies have demonstrated the non-trivial properties of the network connectivity using in vitro recordings from multiple neurons [1, 2, 3], yet it is unclear to what extent this structure reflects reorganization caused by synaptic plasticity, and what is imposed by the morphological constraints. Two recent articles explored this issue using the simulated neural circuits and demonstrated the specific structural properties in those circuits [4, 5].We analyzed a model that emphasizes the role of single-cell morphology, a homogeneous population of neurons in a planar space without boundaries. Each neuron is composed of two displaced neurite fields defined on the limited support. A neurite field describes the likelihood of finding a neurite segment at a certain point in the plane. Using a proximity criterion (Peters' rule) the expected number of potential synapses is estimated between each pair of neurons. Alternatively, this number can be estimated from the realistic morphology of a simulated neuron, or from the morphologies reconstructed from in vitro/in vivo recordings. The number of potential synapses depends on the axon-dendrite distance, which leads to a definition of the expected radius. An axon-dendrite pair that is expected to form at least one synapse must be on a distance not larger than the effective radius. All considered statistical measures of network connectivity are expressed as the functions of the effective radius normalized with the neuron size. In this study, we considered the standard graph theoretic measures of network connectivity, the motif counts, clustering coefficient, path length, and small-world coefficient. It has been demonstrated that they have a significant impact on the population activity in simulated networks [6].Changing the normalized effective radius from small (<0.3) to big (>10) we vary the network properties between the two extremes. For the small values of the effective radius, the networks favor unidirectional connections and sparse local connectivity. The clustering coefficient and the path length are similar to those obtained in uniform random networks, i.e. in the networks independent of topology. For the large values of the effective radius, the local connectivity is dense with the majority of bidirectional connections. As the normalized effective radius increases, the clustering coefficient increases towards the values obtained for the networks with dominant local connectivity, while the path length remains close to the one of the uniform random networks. The normalized effective radius on the interval 1-2, provides the biggest variability of connectivity patterns and the optimized properties relevant for the information transfer.Conclusions: We present a theoretical framework that relates neuromorphology with the connectivity in neuronal circuits, and that can be solved analytically. The normalized effective radius was found to be the key morphological property that dominantly affects considered connectivity measures. By tuning it we can obtain the networks with the biggest variability of local connectivity patterns. At the same time, those networks acquire the key characteristics of the small-world networks, known to optimize the information transfer.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Faculty of Biomedical Sciences and Engineering
Contributors: Acimovic, J., Mäki-Marttunen, T. M., Linne, M.
Number of pages: 1
Pages: P:O7
Publication date: 4 Dec 2015

Host publication information

Title of host publication: 24th Annual Computational Neuroscience Meeting: CNS*2015
Volume: 16 (Suppl 1)
Place of publication: Prague
Publisher: BioMed Central
Editors: Cymbalyuk, G., Burkitt, A.
Article number: O7
ASJC Scopus subject areas: Neuroscience (miscellaneous), Signal Processing
Keywords: network connectivity, clustering coefficient, effective radius, neuronal circuit, connectivity pattern, morphology, dendrites, Neurite density field, neurites

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Binomial Gaussian mixture filter

In this work, we present a novel method for approximating a normal distribution with a weighted sum of normal distributions. The approximation is used for splitting normally distributed components in a Gaussian mixture filter, such that components have smaller covariances and cause smaller linearization errors when nonlinear measurements are used for the state update. Our splitting method uses weights from the binomial distribution as component weights. The method preserves the mean and covariance of the original normal distribution, and in addition, the resulting probability density and cumulative distribution functions converge to the original normal distribution when the number of components is increased. Furthermore, an algorithm is presented to do the splitting such as to keep the linearization error below a given threshold with a minimum number of components. The accuracy of the estimate provided by the proposed method is evaluated in four simulated single-update cases and one time series tracking case. In these tests, it is found that the proposed method is more accurate than other Gaussian mixture filters found in the literature when the same number of components is used and that the proposed method is faster and more accurate than particle filters.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Automation Science and Engineering, Department of Mathematics, Research group: MAT Positioning, Research group: Positioning, Wireless Communications and Positioning (WICO)
Contributors: Raitoharju, M., Ali-Löytty, S., Piché, R.
Publication date: 2 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Advances in Signal Processing
Volume: 2015
Issue number: 1
Article number: 36
ISSN (Print): 1687-6172
Ratings: 
  • Scopus rating (2015): CiteScore 0.83 SJR 0.351 SNIP 0.934
Original language: English
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Keywords: Estimation, Gaussian mixture filter, Nonlinear filtering
Electronic versions: 

Bibliographical note

ORG=ase,0.75
ORG=mat,0.25

Source: Scopus
Source ID: 84934283964

Research output: Contribution to journalArticleScientificpeer-review

Programming graphics processing units in the RVC-CAL dataflow language

The interest towards programming of streaming applications using dataflow models of computation has been increasing steadily in the recent years. Among the numerous dataflow formalisms, the ISO-standardized RVC-CAL dataflow language has offered a solid basis for programming tool development and research. To this date RVC-CAL programming tools have enabled transforming dataflow programs into concurrent executables for multicore processors, as well as for generating synthesizable hardware descriptions. In this paper it is shown how the RVC-CAL dataflow language can be used for programming graphics processing units (GPUs) with high efficiency. Considering the processing architectures of recent mobile and desktop computing devices, this advance is of high importance, as most consumer devices contain a graphics processing unit nowadays. To evaluate the proposed solution, the paper presents a video processing application case study. At best, the solution is shown to provide a speedup of 42× over single-threaded CPU execution.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Signal Processing Research Community (SPRC), Dept. of Computer Science and Engineering, Univ of Oulu
Contributors: Boutellier, J., Nyländen, T.
Publication date: 2 Dec 2015

Host publication information

Title of host publication: Electronic Proceedings of the 2015 IEEE International Workshop on Signal Processing Systems, SiPS 2015
Volume: 2015-December
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7344994
ISBN (Electronic): 9781467396042
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics, Hardware and Architecture
Keywords: Dataflow computing, design automation, parallel processing
Source: Scopus
Source ID: 84958191208

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

A fast universal self-tuned sampler within Gibbs sampling

Bayesian inference often requires efficient numerical approximation algorithms, such as sequential Monte Carlo (SMC) and Markov chain Monte Carlo (MCMC) methods. The Gibbs sampler is a well-known MCMC technique, widely applied in many signal processing problems. Drawing samples from univariate full-conditional distributions efficiently is essential for the practical application of the Gibbs sampler. In this work, we present a simple, self-tuned and extremely efficient MCMC algorithm which produces virtually independent samples from these univariate target densities. The proposal density used is self-tuned and tailored to the specific target, but it is not adaptive. Instead, the proposal is adjusted during an initial optimization stage, following a simple and extremely effective procedure. Hence, we have named the newly proposed approach as FUSS (Fast Universal Self-tuned Sampler), as it can be used to sample from any bounded univariate distribution and also from any bounded multi-variate distribution, either directly or by embedding it within a Gibbs sampler. Numerical experiments, on several synthetic data sets (including a challenging parameter estimation problem in a chaotic system) and a high-dimensional financial signal processing problem, show its good performance in terms of speed and estimation accuracy.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Industrial Management, Research group: Financial Engineering, Research Community on Data-to-Decision (D2D), Universidad Politecnica de Madrid
Contributors: Martino, L., Yang, H., Luengo, D., Kanniainen, J., Corander, J.
Pages: 68-83
Publication date: 1 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Digital Signal Processing
Volume: 47
ISSN (Print): 1051-2004
Ratings: 
  • Scopus rating (2015): CiteScore 2.08 SJR 0.566 SNIP 1.363
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering
Keywords: Adaptive rejection Metropolis sampling, Bayesian inference, Gibbs sampling, Markov Chain Monte Carlo (MCMC), Metropolis within Gibbs
Source: Scopus
Source ID: 84948117324

Research output: Contribution to journalArticleScientificpeer-review

Kalman filter with a linear state model for PDR+WLAN positioning and its application to assisting a particle filter

Indoor positioning based on wireless local area network (WLAN) signals is often enhanced using pedestrian dead reckoning (PDR) based on an inertial measurement unit. The state evolution model in PDR is usually nonlinear. We present a new linear state evolution model for PDR. In simulated-data and real-data tests of tightly coupled WLAN-PDR positioning, the positioning accuracy with this linear model is better than with the traditional models when the initial heading is not known, which is a common situation. The proposed method is computationally light and is also suitable for smoothing. Furthermore, we present modifications to WLAN positioning based on Gaussian coverage areas and show how a Kalman filter using the proposed model can be used for integrity monitoring and (re)initialization of a particle filter.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Automation Science and Engineering, Research area: Dynamic Systems, Research group: Positioning, Wireless Communications and Positioning (WICO)
Contributors: Raitoharju, M., Nurminen, H., Piché, R.
Publication date: 1 Dec 2015
Peer-reviewed: Yes

Publication information

Journal: Eurasip Journal on Advances in Signal Processing
Volume: 2015
Issue number: 1
Article number: 33
ISSN (Print): 1687-6172
Ratings: 
  • Scopus rating (2015): CiteScore 0.83 SJR 0.351 SNIP 0.934
Original language: English
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering
Keywords: Computational modeling, Indoor positioning, Pedestrian dead reckoning, Wireless LAN
Electronic versions: 
Source: Scopus
Source ID: 84928397748

Research output: Contribution to journalArticleScientificpeer-review

A New Approach to Sign-Bit-Based Parameter Estimation in OFDM Receivers

This article combines algorithm development, thorough analysis and implementation of sign-bit (SB) estimation techniques for symbol timing, carrier frequency offset (CFO) and signal-to-noise ratio (SNR) in orthogonal frequency division multiplexing receivers. The SB estimation is compared in terms of performance and hardware complexity to an equivalent implementation with higher quantization. The techniques are demonstrated by simulation of a SB time/frequency and SB-SNR estimator for 3rd Generation Partnership Project long-term evolution (LTE) cell search in 65-nm technology operating at nominal voltage of 1.2 V. According to post-layout power simulations with toggling information, the architecture estimates the corresponding CFO and SNR for as little as $$479\,\upmu \hbox {W}$$479μW average power for LTE-R8/10, while occupying a silicon area as small as $$0.03\,\hbox {mm}^2$$0.03<sup>mm2</sup>. Even though SB estimation experiences some relative performance penalty when compared to 8-bit quantization, this paper demonstrates various advantages and the potential of employing these techniques in low-complexity terminals.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Wireless Communications and Positioning (WICO), Lund Univ, Lund University, Ericsson Research
Contributors: Diaz, I., Wilhelmsson, L. R., Sofotasios, P. C., Miao, Y., Tan, S., Edfors, O., Öwall, V.
Number of pages: 30
Pages: 3631-3660
Publication date: 23 Nov 2015
Peer-reviewed: Yes

Publication information

Journal: Circuits, Systems and Signal Processing
Volume: 34
Issue number: 11
ISSN (Print): 0278-081X
Ratings: 
  • Scopus rating (2015): CiteScore 1.54 SJR 0.506 SNIP 0.909
Original language: English
ASJC Scopus subject areas: Signal Processing, Applied Mathematics
Keywords: CFO, LTE, OFDM, Receiver design, SNR, Symbol timing, Time synchronization
Source: Scopus
Source ID: 84941992241

Research output: Contribution to journalArticleScientificpeer-review

Cosparse dictionary learning for the orthogonal case

Dictionary learning is usually approached by looking at the support of the sparse representations. Recent years have shown results in dictionary improvement by investigating the cosupport via the analysis-based cosparse model. In this paper we present a new cosparse learning algorithm for orthogonal dictionary blocks that provides significant dictionary recovery improvements and representation error shrinkage. Furthermore, we show the beneficial effects of using this algorithm inside existing methods based on building the dictionary as a structured union of orthonormal bases.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Signal Processing Research Community (SPRC), University Politehnica of Bucharest
Contributors: Irofti, P., Dumitrescu, B.
Number of pages: 5
Pages: 343-347
Publication date: 5 Nov 2015

Host publication information

Title of host publication: 2015 19th International Conference on System Theory, Control and Computing, ICSTCC 2015 - Joint Conference SINTES 19, SACCS 15, SIMSIS 19
Publisher: IEEE
ISBN (Print): 9781479984817
ASJC Scopus subject areas: Control and Systems Engineering, Artificial Intelligence, Information Systems, Signal Processing
Keywords: cosparse, dictionary design, orthogonal blocks, sparse representation
Source: Scopus
Source ID: 84957825116

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Coupled dictionaries for exemplar-based speech enhancement and automatic speech recognition

Exemplar-based speech enhancement systems work by decomposing the noisy speech as a weighted sum of speech and noise exemplars stored in a dictionary and use the resulting speech and noise estimates to obtain a time-varying filter in the full-resolution frequency domain to enhance the noisy speech. To obtain the decomposition, exemplars sampled in lower dimensional spaces are preferred over the full-resolution frequency domain for their reduced computational complexity and the ability to better generalize to unseen cases. But the resulting filter may be sub-optimal as the mapping of the obtained speech and noise estimates to the full-resolution frequency domain yields a low-rank approximation. This paper proposes an efficient way to directly compute the full-resolution frequency estimates of speech and noise using coupled dictionaries: an input dictionary containing atoms from the desired exemplar space to obtain the decomposition and a coupled output dictionary containing exemplars from the full-resolution frequency domain. We also introduce modulation spectrogram features for the exemplar-based tasks using this approach. The proposed system was evaluated for various choices of input exemplars and yielded improved speech enhancement performances on the AURORA-2 and AURORA-4 databases. We further show that the proposed approach also results in improved word error rates (WERs) for the speech recognition tasks using HMM-GMM and deep-neural network (DNN) based systems.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Audio research group, Research Community on Data-to-Decision (D2D), Signal Processing Research Community (SPRC), Speech Processing Research Group, Electrical Engineering Department (ESAT), KU Leuven
Contributors: Baby, D., Virtanen, T., Gemmeke, J. F., Van hamme, H.
Number of pages: 12
Pages: 1788-1799
Publication date: 1 Nov 2015
Peer-reviewed: Yes

Publication information

Journal: Ieee-Acm transactions on audio speech and language processing
Volume: 23
Issue number: 11
ISSN (Print): 2329-9290
Ratings: 
  • Scopus rating (2015): CiteScore 2.4 SJR 1.13 SNIP 1.58
Original language: English
ASJC Scopus subject areas: Signal Processing, Electrical and Electronic Engineering, Media Technology, Acoustics and Ultrasonics, Instrumentation, Linguistics and Language, Speech and Hearing
Keywords: Exemplar-based, Modulation envelope, Noise robust automatic speech recognition, Non-negative sparse coding
Source: Scopus
Source ID: 84942609488

Research output: Contribution to journalArticleScientificpeer-review

Entropy and Channel Capacity under Optimum Power and Rate Adaptation over Generalized Fading Conditions

Accurate fading characterization and channel capacity determination are of paramount importance in both conventional and emerging communication systems. The present work addresses the non-linearity of the propagation medium and its effects on the channel capacity. Such fading conditions are first characterized using information theoretic measures, namely, Shannon entropy, cross entropy and relative entropy. The corresponding effects on the channel capacity with and without power adaptation are then analyzed. Closed-form expressions are derived and validated through computer simulations. It is shown that the effects of nonlinearities are significantly larger than those of fading parameters such as the scattered-wave power ratio, and the correlation coefficient between the in-phase and quadrature components in each cluster of multipath components.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Research group: Wireless Communications and Positioning, Department of Electronics and Communications Engineering, Wireless Communications and Positioning (WICO), Department of Electrical Engineering and Computer Science, Aristotle University of Thessaloniki, Department of Electronic Engineering, Khalifa University, Centre for Communication Systems Research, University of Surrey, University of Leeds, International University of Rabat
Contributors: Sofotasios, P. C., Muhaidat, S., Valkama, M., Ghogho, M., Karagiannidis, G. K.
Number of pages: 5
Pages: 2162-2166
Publication date: 1 Nov 2015
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 22
Issue number: 11
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2015): CiteScore 3.13 SJR 0.822 SNIP 1.788
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Keywords: Adaptation policies, channel capacity, entropy
Source: Scopus
Source ID: 84939865476

Research output: Contribution to journalArticleScientificpeer-review

Robust Inference for State-Space Models with Skewed Measurement Noise

Filtering and smoothing algorithms for linear discrete-time state-space models with skewed and heavy-tailed measurement noise are presented. The algorithms use a variational Bayes approximation of the posterior distribution of models that have normal prior and skew-t-distributed measurement noise. The proposed filter and smoother are compared with conventional low-complexity alternatives in a simulated pseudorange positioning scenario. In the simulations the proposed methods achieve better accuracy than the alternative methods, the computational complexity of the filter being roughly 5 to 10 times that of the Kalman filter.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Automation Science and Engineering, Research group: Positioning, Wireless Communications and Positioning (WICO), Linköping University
Contributors: Nurminen, H., Ardeshiri, T., Piché, R., Gustafsson, F.
Number of pages: 5
Pages: 1898-1902
Publication date: 1 Nov 2015
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 22
Issue number: 11
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2015): CiteScore 3.13 SJR 0.822 SNIP 1.788
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Keywords: Kalman filter, robust filtering, RTS smoother, skew t, skewness, t-distribution, variational Bayes
Electronic versions: 
Source: Scopus
Source ID: 84933056939

Research output: Contribution to journalArticleScientificpeer-review

Sparse extreme learning machine classifier exploiting intrinsic graphs

This paper presents an analysis of the recently proposed sparse extreme learning machine (S-ELM) classifier and describes an optimization scheme that can be used to calculate the network output weights. This optimization scheme exploits intrinsic graph structures in order to describe geometric data relationships in the so-called ELM space. Kernel formulations of the approach operating in ELM spaces of arbitrary dimensions are also provided. It is shown that the application of the optimization scheme exploiting geometric data relationships in the original ELM space is equivalent to the application of the original S-ELM to a transformed ELM space. The experimental results show that the incorporation of geometric data relationships in S-ELM can lead to enhanced performance.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research Community on Data-to-Decision (D2D), Aristotle University of Thessaloniki, Department of Informatics, Aristotle University of Thessaloniki
Contributors: Iosifidis, A., Tefas, A., Pitas, I.
Number of pages: 5
Pages: 192-196
Publication date: 1 Nov 2015
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition Letters
Volume: 65
ISSN (Print): 0167-8655
Ratings: 
  • Scopus rating (2015): CiteScore 2.87 SJR 0.95 SNIP 1.979
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Intrinsic graphs, Single-hidden layer neural networks, Sparse extreme learning machine
Source: Scopus
Source ID: 84940388000

Research output: Contribution to journalArticleScientificpeer-review

Model selection for linear classifiers using Bayesian error estimation

Regularized linear models are important classification methods for high dimensional problems, where regularized linear classifiers are often preferred due to their ability to avoid overfitting. The degree of freedom of the model dis determined by a regularization parameter, which is typically selected using counting based approaches, such as K-fold cross-validation. For large data, this can be very time consuming, and, for small sample sizes, the accuracy of the model selection is limited by the large variance of CV error estimates. In this paper, we study the applicability of a recently proposed Bayesian error estimator for the selection of the best model along the regularization path. We also propose an extension of the estimator that allows model selection in multiclass cases and study its efficiency with L-1 regularized logistic regression and L-2 regularized linear support vector machine. The model selection by the new Bayesian error estimator is experimentally shown to improve the classification accuracy, especially in small sample-size situations, and is able to avoid the excess variability inherent to traditional cross-validation approaches. Moreover, the method has significantly smaller computational complexity than cross-validation. (C) 2015 Elsevier Ltd. All rights reserved.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: Vision, Research Community on Data-to-Decision (D2D), Universidad Carlos III de Madrid
Contributors: Huttunen, H., Tohka, J.
Number of pages: 10
Pages: 3739-3748
Publication date: Nov 2015
Peer-reviewed: Yes

Publication information

Journal: Pattern Recognition
Volume: 48
Issue number: 11
ISSN (Print): 0031-3203
Ratings: 
  • Scopus rating (2015): CiteScore 4.83 SJR 1.579 SNIP 2.987
Original language: English
ASJC Scopus subject areas: Software, Artificial Intelligence, Computer Vision and Pattern Recognition, Signal Processing
Keywords: Logistic regression, Support vector machine, Regularization, Bayesian error estimator, Linear classifier, MULTINOMIAL LOGISTIC-REGRESSION, SUPPORT VECTOR MACHINES, CLASSIFICATION, PERFORMANCE, BOUNDS

Bibliographical note

EXT="Tohka, Jussi"

Source: Scopus
Source ID: 84937812363

Research output: Contribution to journalArticleScientificpeer-review

The effect of region of interest size on textural parameters

Texture analysis provides quantitative information describing the properties of a digital image. The value of texture analysis has been tested in various medical applications, using mostly magnetic resonance images because of the amount of information the method is capable to provide. However, there exists no certain practice to define the region of interest (ROI) within the texture parameters are calculated. Many parameters seem to be dependent on the ROI size. We studied the effect of the ROI size with magnetic resonance head images from 64 healthy adults and artificial noise images. According to our results, ROI size has a significant effect on the computed value of several second-order texture features. We conclude that comparisons of different size ROIs will therefore lead to falsely optimistic classification between analyzed tissues.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Electronics and Communications Engineering, Research group: Quantative medical imaging, Integrated Technologies for Tissue Engineering Research (ITTE), Tampere University Hospital, Department of Radiology
Contributors: Sikiö, M., Holli-Helenius, K. K., Ryymin, P., Dastidar, P., Eskola, H., Harrison, L.
Number of pages: 5
Pages: 149-153
Publication date: 23 Oct 2015

Host publication information

Title of host publication: 2015 9th International Symposium on Image and Signal Processing and Analysis (ISPA)
Publisher: IEEE
ISBN (Electronic): 9781467380324
ASJC Scopus subject areas: Signal Processing, Computer Vision and Pattern Recognition
Keywords: magnetic resonance imaging, random pattern, region of interest, size, texture analysis

Bibliographical note

EXT="Dastidar, Prasun"
EXT="Sikiö, Minna"

Source: Scopus
Source ID: 84978524965

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Solutions to integrals involving the marcum Q-function and applications

Novel analytic solutions are derived for integrals that involve the generalized Marcum Q -function, exponential functions and arbitrary powers. Simple closed-form expressions are also derived for specific cases of the generic integrals. The offered expressions are both convenient and versatile, which is particularly useful in applications relating to natural sciences and engineering, including wireless communications and signal processing. To this end, they are employed in the derivation of the average probability of detection in energy detection of unknown signals over multipath fading channels as well as of the channel capacity with fixed rate and channel inversion in the case of correlated multipath fading and switched diversity.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Electronics and Communications Engineering, Research group: Wireless Communications and Positioning, Wireless Communications and Positioning (WICO), Department of Electrical and Computer Engineering, Khalifa University, Centrefor Communication Systems Research, Department of Electronic Engineering, University of Surrey, Department of Electrical and Computer Engineering, Aristotle University of Thessaloniki
Contributors: Sofotasios, P. C., Muhaidat, S., Karagiannidis, G. K., Sharif, B. S.
Number of pages: 5
Pages: 1752-1756
Publication date: 1 Oct 2015
Peer-reviewed: Yes

Publication information

Journal: IEEE Signal Processing Letters
Volume: 22
Issue number: 10
ISSN (Print): 1070-9908
Ratings: 
  • Scopus rating (2015): CiteScore 3.13 SJR 0.822 SNIP 1.788
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing, Applied Mathematics
Source: Scopus
Source ID: 84930225545

Research output: Contribution to journalArticleScientificpeer-review

MOBILE3DTV: Content delivery optimization over DVB-H system

Mobile TV has recently received a lot of attention worldwide with the advances in technologies such as Digital Multimedia Broadcasting (DMB), Digital Video Broadcasting - Handheld (DVB-H) and MediaFLO. On the other hand 3DTV is a new approach to watching TV, introducing the third dimension for a more realistic and interactive experience. With the merge of these two technologies it will be possible to have 3DTV products and services based on portable platforms with switchable 2D/3D autostereoscopic displays. The paper presents the European Mobile3DTV project approach toward achieving such a merge. The project specifically addresses the mobile 3DTV delivery over DVB-H system. It develops a technology demonstration system comprising suitable stereo-video content-creation techniques; efficient, scalable and flexible stereo-video encoders with error resilience and error-concealment capabilities, tailored for robust transmission over DVB-H; and also the corresponding stereo-video decoders and players working on a portable terminal device equipped with an autostereoscopic display.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: 3D MEDIA, Signal Processing Research Community (SPRC), Middle East Technical Univ.
Contributors: Akar, G. B., Gotchev, A.
Publication date: 15 Sep 2015

Host publication information

Title of host publication: SMPTE International Conference on Stereoscopic 3D for Media and Entertainment
Publisher: SMPTE
ISBN (Electronic): 9781614829508
ASJC Scopus subject areas: Signal Processing, Media Technology
Keywords: DVB-H, error resilient transmission, Mobile3DTV
Source: Scopus
Source ID: 84978435566

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

The extended one-dimensional discrete phase retrieval problem

It has been known for decades that the iterative methods are perhaps the most popular approaches to solve the phase retrieval problem. Unfortunately the iterative methods often stagnate. This happens also in the case of the 1-D Discrete Phase Retrieval (1-D DPhR) problem. Recently it has been shown that some requirements in the input magnitude data might be one of the reasons why the direct method cannot solve the 1-D DPhR Problem. In this work we present some difficulties that can be encountered when one has to implement the iterative method for finding a solution of 1-D DPhR problem. We shall formulate the extended form of 1-D DPhR problem. Simulations indicate the conjecture to be true.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Algebraic and Algorithmic Methods in Signal Processing AAMSP, Signal Processing Research Community (SPRC), FETTI, Technical University of Cluj-NapocaUniversitatea Tehnica din Cluj-Napoca
Contributors: Rusu, C., Astola, J.
Publication date: 14 Aug 2015

Host publication information

Title of host publication: 2015 International Symposium on Signals, Circuits and Systems (ISSCS)
Publisher: IEEE
ISBN (Print): 9781467374873
ASJC Scopus subject areas: Hardware and Architecture, Signal Processing, Electrical and Electronic Engineering

Bibliographical note

EXT="Rusu, Corneliu"

Source: Scopus
Source ID: 84955586971

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Collaborative filtering based on group coordinates for smoothing and directional sharpening

Groups of mutually similar image blocks are the key element in nonlocal image processing. In this work, the spatial coordinates of grouped blocks are leveraged in two distinct parts of the transform-domain collaborative filtering within the BM3D algorithm. First, we introduce an adaptive 1-D transform for 3-D collaborative filtering based on sampling 2-D smooth functions at the positions of grouped blocks. This adaptive transform is applied for improved decorrelation of the 2-D spectra of the grouped blocks. Second, we propose a directional sharpening procedure whose strength varies adaptively according to the relative orientation of the transform basis functions with respect to the group coordinates. Experiments confirm the efficacy of the proposed adaptations, for denoising as well as for sharpening of noisy images.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Signal and Image Restoration-RST, Research area: Signal and Information Processing, Signal Processing Research Community (SPRC)
Contributors: Azzari, L., Foi, A.
Number of pages: 5
Pages: 1573-1577
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: IEEE
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: adaptive transforms, BM3D, collaborative filtering, denoising, sharpening
Source: Scopus
Source ID: 84946074037

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Enhancing class discrimination in Kernel Discriminant Analysis

In this paper, we propose an optimization scheme aiming at optimal nonlinear data projection, in terms of Fisher ratio maximization. To this end, we formulate an iterative optimization scheme consisting of two processing steps: optimal data projection calculation and optimal class representation determination. Compared to the standard approach employing the class mean vectors for class representation, the proposed optimization scheme increases class discrimination in the reduced-dimensionality feature space. We evaluate the proposed method in standard classification problems, as well as on the classification of human actions and face, and show that it is able to achieve better generalization performance, when compared to the standard approach.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Research Community on Data-to-Decision (D2D), Aristotle University of Thessaloniki
Contributors: Iosifidis, A., Tefas, A., Pitas, I.
Number of pages: 5
Pages: 1926-1930
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
Article number: 7178306
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: Kernel Discriminant Analysis, Nonlinear data projection, Optimized Class Representation
Source: Scopus
Source ID: 84946088724

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Exemplar-based speech enhancement for deep neural network based automatic speech recognition

Deep neural network (DNN) based acoustic modelling has been successfully used for a variety of automatic speech recognition (ASR) tasks, thanks to its ability to learn higher-level information using multiple hidden layers. This paper investigates the recently proposed exemplar-based speech enhancement technique using coupled dictionaries as a pre-processing stage for DNN-based systems. In this setting, the noisy speech is decomposed as a weighted sum of atoms in an input dictionary containing exemplars sampled from a domain of choice, and the resulting weights are applied to a coupled output dictionary containing exemplars sampled in the short-time Fourier transform (STFT) domain to directly obtain the speech and noise estimates for speech enhancement. In this work, settings using input dictionary of exemplars sampled from the STFT, Mel-integrated magnitude STFT and modulation envelope spectra are evaluated. Experiments performed on the AURORA-4 database revealed that these pre-processing stages can improve the performance of the DNN-HMM-based ASR systems with both clean and multi-condition training.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Audio research group, Signal Processing Research Community (SPRC), KU Leuven
Contributors: Baby, D., Gemmeke, J. F., Virtanen, T., Van Hamme, H.
Number of pages: 5
Pages: 4485-4489
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: coupled dictionaries, deep neural networks, modulation envelope, non-negative matrix factorisation, speech enhancement
Source: Scopus
Source ID: 84946079930

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Exploiting subclass information in one-class support vector machine for video summarization

In this paper, we propose a method for video summarization based on human activity description. We formulate this problem as the one of automatic video segment selection based on a learning process that employs salient video segment paradigms. For this one-class classification problem, we introduce a novel variant of the One-Class Support Vector Machine (OC-SVM) classifier that exploits subclass information in the OC-SVM optimization problem, in order to jointly minimize the data dispersion within each subclass and determine the optimal decision function. We evaluate the proposed approach in three Hollywood movies, where the performance of the proposed SOC-SVM algorithm is compared with that of the OC-SVM. Experimental results denote that the proposed approach is able to outperform OC-SVM-based video segment selection.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Research Community on Data-to-Decision (D2D), Aristotle University of Thessaloniki
Contributors: Mygdalis, V., Iosifidis, A., Tefas, A., Pitas, I.
Number of pages: 5
Pages: 2259-2263
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Volume: 2015-August
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: One class classification, Subclass One-Class SVM, Supervised Video Summarization
Source: Scopus
Source ID: 84945586271

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Low-complexity robust DOA estimation

We propose a low complexity method for estimating direction of arrival (DOA) when the positions of the array sensors are affected by errors with known magnitude bound. This robust DOA method is based on solving an optimization problem whose solution is obtained in two stages. First, the problem is relaxed and the corresponding power estimation has an expression similar to that of standard beamforming. If the relaxed solution does not satisfy the magnitude bound, an approximation is made by projection. Unlike other robust DOA methods, no eigenvalue decomposition is necessary and the complexity is similar to that of MVDR. For low and medium SNR, the proposed method competes well with more complex methods and is clearly better than MVDR.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Algebraic and Algorithmic Methods in Signal Processing AAMSP, Signal Processing Research Community (SPRC), Dept. of Automatic Control and Computers, University Politehnica of Bucharest
Contributors: Dumitrescu, B., Rusu, C., Tabus, I., Astola, J.
Number of pages: 5
Pages: 2794-2798
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: direction of arrival, optimization, robustness, spectral power density
Source: Scopus
Source ID: 84946064969

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Low-Latency Sound-Source-Separation using Non-Negative Matrix Factorisation with Coupled Analysis and Synthesis Dictionaries

For real-time or close to real-time applications, sound source separation can be performed on-line, where new frames of incoming data for a mixture signal are processed as they arrive, at very low delay. We propose an approach which generates the separation filters for short synthesis frames to achieve low latency source separation, based on a compositional model mixture of the audio to be separated. Filter parameters are derived from a longer temporal context than the current processing frame through use of a longer analysis frame. A pair of dictionaries are used, one for analysis and one for reconstruction. With this approach we are able to increase separation performance at low latencies whilst retaining the low-latency provided by the use of short synthesis frames. The proposed data handling scheme and parameters can be adjusted to achieve real-time performance, given sufficient computational power. Low-latency output allows a human listener to use the results of such a separation scheme directly, without a perceptible delay. With the proposed method, separated source-to-distortion ratios (SDRs) can be improved by over 1 dB for latencies below 20 ms, without any affect on latency.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Audio research group, Signal Processing Research Community (SPRC)
Contributors: Barker, T., Virtanen, T., Pontoppidan, N. H.
Number of pages: 5
Pages: 241-245
Publication date: 4 Aug 2015

Host publication information

Title of host publication: 2015 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP)
Publisher: IEEE
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: low-latency, NMF, Non-negative matrix factorisation, real-time, source separation
Source: Scopus
Source ID: 84946043719

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Similarity induced group sparsity for non-negative matrix factorisation

Non-negative matrix factorisations are used in several branches of signal processing and data analysis for separation and classification. Sparsity constraints are commonly set on the model to promote discovery of a small number of dominant patterns. In group sparse models, atoms considered to belong to a consistent group are permitted to activate together, while activations across groups are suppressed, reducing the number of simultaneously active sources or other structures. Whereas most group sparse models require explicit division of atoms into separate groups without addressing their mutual relations, we propose a constraint that permits dynamic relationships between atoms or groups, based on any defined distance measure. The resulting solutions promote approximation with components considered similar to each other. Evaluation results are shown for speech enhancement and noise robust speech and speaker recognition.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Department of Signal Processing, Research group: Audio research group, Signal Processing Research Community (SPRC), Aalto University
Contributors: Hurmalainen, A., Saeidi, R., Virtanen, T.
Number of pages: 5
Pages: 4425-4429
Publication date: 4 Aug 2015

Host publication information

Title of host publication: ICASSP, IEEE International Conference on Acoustics, Speech and Signal Processing - Proceedings
Publisher: The Institute of Electrical and Electronics Engineers, Inc.
ISBN (Print): 9781467369978
ASJC Scopus subject areas: Signal Processing, Software, Electrical and Electronic Engineering
Keywords: group sparsity, non-negative matrix factorization, sparse representations, speaker recognition, speech recognition
Source: Scopus
Source ID: 84946089342

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Adaptive Randomized Coordinate Descent for Sparse Systems: Lasso and Greedy Algorithms

Coordinate descent (CD) is a simple optimization technique suited to low complexity requirements and also for solving large problems. In randomized version, CD was recently shown as very effective for solving least-squares (LS) and other optimization problems. We propose here an adaptive version of randomized coordinate descent (RCD) for finding sparse LS solutions, from which we derive two algorithms, one based on the lasso criterion, the other using a greedy technique. Both algorithms employ a novel way of adapting the probabilities for choosing the coordinates, based on a matching pursuit criterion. Another new feature is that, in the lasso algorithm, the penalty term values are built without knowing the noise level or using other prior information. The proposed algorithms use efficient computations and have a tunable trade-off between complexity and performance through the number of CD steps per time instant. Besides a general theoretical convergence analysis, we present simulations that show good practical behavior, comparable to or better than that of state of the art methods.

General information

Publication status: Published
MoE publication type: A1 Journal article-refereed
Organisations: Department of Signal Processing, Research group: SGN-SPAG, Signal Processing Research Community (SPRC)
Contributors: Onose, A., Dumitrescu, B.
Number of pages: 11
Pages: 4091-4101
Publication date: 1 Aug 2015
Peer-reviewed: Yes

Publication information

Journal: IEEE Transactions on Signal Processing
Volume: 63
Issue number: 15
ISSN (Print): 1053-587X
Ratings: 
  • Scopus rating (2015): CiteScore 4.65 SJR 1.581 SNIP 2.611
Original language: English
ASJC Scopus subject areas: Electrical and Electronic Engineering, Signal Processing
Keywords: Adaptive algorithms, coordinate descent, randomization, sparse filters
Source: Scopus
Source ID: 84934311735

Research output: Contribution to journalArticleScientificpeer-review

3D/inkjet-printed origami antennas for multi-direction RF harvesting

A system design is presented for radio frequency (RF) energy harvesting on wireless sensor network (WSN) nodes, where all electronics reside inside a 3D structure and the antennas lie on the surfaces of it. Additive manufacturing techniques are used for the packaging and antenna fabrication: A 3D-printed cross-shaped structure is built that folds to a cuboid in an 'origami' fashion and retains its shape at room temperature. Inkjet printing is used to directly fabricate antennas on the surfaces of the 3D-printed plastic, enabling a fully additive manufacturing of the structure. Multiple antennas on the cube's surfaces can be used for RF energy harvesting of signals arriving from totally orthogonal directions, with the use of an appropriate harvester. The system modules (cube, antenna, harvester) are described and characterized, offering a proof-of-concept for the combination of fabrication techniques to build systems for demanding RF applications.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Sensing Systems for Wireless Medicine (MediSense), Georgia Institute of Technology, School of Electrical and Computer Engineering, Centre Tecnologic de Telecommunicacions de Catalunya, Georgia Institute of Technology
Contributors: Kimionis, J., Georgiadis, A., Isakov, M., Qi, H. J., Tentzeris, M. M.
Publication date: 24 Jul 2015

Host publication information

Title of host publication: 2015 IEEE MTT-S International Microwave Symposium, IMS 2015
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7166878
ISBN (Electronic): 9781479982752
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Electrical and Electronic Engineering
Keywords: 3D printing, additive manufacturing, inkjet printing, multiple antennas, RF energy harvesting
Source: Scopus
Source ID: 84946068129

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Additively manufactured multilayer high performance RF passive components on cellulose substrates for internet-of-things electronic circuits

This paper demonstrates the feasibility to manufacture high performance radio-frequency passive components on cellulose substrates by exploiting two novel technologies: the vertically integrated inkjet printing and the copper laminate method. Both processes are substrate independent and thus suitable for fabricating circuits on paper as well; moreover, in a future perspective, they can be easily combined together in order to exploit their complementarity. Passive components such as capacitors and inductors, with Qs up to 22, never registered before on cellulose substrates, and Self-Resonant-Frequency (SRF) up to 4 GHz are described. The obtained values of the capacitance and inductance per unit area are 0.8 pF/mm2 and 43 nH/mm2, respectively.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Sensing Systems for Wireless Medicine (MediSense), University of Perugia, Georgia Institute of Technology, School of Electrical and Computer Engineering
Contributors: Mariotti, C., Cook, B. S., Alimenti, F., Roselli, L., Tentzeris, M. M.
Publication date: 24 Jul 2015

Host publication information

Title of host publication: 2015 IEEE MTT-S International Microwave Symposium, IMS 2015
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7166924
ISBN (Electronic): 9781479982752
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Electrical and Electronic Engineering
Keywords: Additive Manufacturing, Copper Laminate, Flexible Electronics, green electronics, Inkjet Printing, Internet of Things, Passives
Source: Scopus
Source ID: 84946068619

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

A flexible hybrid printed RF energy harvester utilizing catalyst-based copper printing technologies for far-field RF energy harvesting applications

In this paper, the design of a novel flexible RF energy harvester utilizing hybrid printed electronics technology is presented for the first time. The proposed RF energy harvester operates at UHF RFID band (868 MHz ∼ 915 MHz) for far-field RF energy harvesting applications. A concept of hybrid printed electronics which takes advantage of both flexibility of low-cost printed electronics and high performances of ICs is introduced. The passive components of the RF energy harvester, such as the circuit layout and the antenna, are printed on a flexible low-cost polymer substrate utilizing a catalyst-based inkjet printing process for the fabrication of copper metallization layers. The surface-mount devices (SMDs) are soldered on the printed circuit board. The proposed approach demonstrates the feasibility of implementing low-cost flexible printed electronics for the Internet of Things (IoT) and stand-alone ('zero-power') wireless sensor platforms.

General information

Publication status: Published
MoE publication type: A4 Article in a conference publication
Organisations: Sensing Systems for Wireless Medicine (MediSense), Georgia Institute of Technology, Centre Tecnològic Telecomunicacions Catalunya (CTTC)
Contributors: Kim, S., Bito, J., Jeong, S., Georgiadis, A., Tentzeris, M. M.
Publication date: 24 Jul 2015

Host publication information

Title of host publication: 2015 IEEE MTT-S International Microwave Symposium, IMS 2015
Publisher: Institute of Electrical and Electronics Engineers Inc.
Article number: 7166723
ISBN (Electronic): 9781479982752
ASJC Scopus subject areas: Computer Networks and Communications, Signal Processing, Electrical and Electronic Engineering
Keywords: Catalyst-based inkjet printing, charge pump, hybrid printed electronics, printed copper, RF energy harvester, RF-DC converter, wireless power transfer
Source: Scopus
Source ID: 84946035832

Research output: Chapter in Book/Report/Conference proceedingConference contributionScientificpeer-review

Ambient energy harvesting from a two-way talk radio for flexible wearable devices utilizing inkjet printing masking

In this paper, the design, on a flexible LCP substrate, and fabrication process of a wearable circui