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.

Decline in respiratory regulation demonstrates the primary forewarning for the onset of physiological aberrations. In clinical environment, the obtrusive nature and cost of instrumentation have retarded the integration of continuous respiration monitoring for standard practice. Photoplethysmography (PPG) presents a non-invasive, optical method of assessing blood flow dynamics in peripheral vasculature. Incidentally, respiration couples as a surrogate constituent in PPG signal, justifying respiratory rate (RR) estimation. The physiological processes of respiration emerge as distinctive oscillations that are fluctuations in various parameters extracted from PPG signal. We propose a novel algorithm designed to account for intermittent diminishment of the respiration induced variabilities (RIV) by a fusion-based enhancement of wavelet synchrosqueezed spectra. We have combined the information on intrinsic mode functions (IMF) of five RIVs to enhance mutually occurring, instantaneous frequencies of the spectra. The respiration rate estimate is obtained by tracking the spectral ridges with a particle filter. We have evaluated the method with a dataset recorded from 29 young adult subjects (mean: 24.17 y, SD: 4.19 y) containing diverse, voluntary, and periodically metronome-assisted respiratory patterns. Bayesian inference on fusion-enhanced Respiration Induced Frequency Variability (RIFV) indicated MAE and RMSE of 1.764 and 3.996 BPM, respectively. The fusion approach was deemed to improve MAE and RMSE of RIFV by 0.185 BPM (95% HDI: 0.0285-0.3488, effect size: 0.548) and 0.250 BPM (95% HDI: 0.0733-0.431, effect size: 0.653), respectively, with further pronounced improvements to other RIVs. We conclude that the fusion of variability signals proves important to IMF localization in the spectral estimation of RR.

Audio source separation is usually achieved by estimating the short-time Fourier transform (STFT) magnitude of each source, and then applying a spectrogram inversion algorithm to retrieve time-domain signals. In particular, the multiple input spectrogram inversion (MISI) algorithm has been exploited successfully in several recent works. However, this algorithm suffers from two drawbacks, which we address in this letter. First, it has originally been introduced in a heuristic fashion: we propose here a rigorous optimization framework in which MISI is derived, thus proving the convergence of this algorithm. Besides, while MISI operates offline, we propose here an online version of MISI called oMISI, which is suitable for low-latency source separation, an important requirement for e.g., hearing aids applications. oMISI also allows one to use alternative phase initialization schemes exploiting the temporal structure of audio signals. Experiments conducted on a speech separation task show that oMISI performs as well as its offline counterpart, thus demonstrating its potential for real-time source separation.

Facial pacing systems aim to reanimate paralyzed facial muscles with electrical stimulation. To aid the development of such systems, the frontalis muscle responsible for eyebrow raising was transcutaneously stimulated in 12 healthy participants using four waveforms: square wave, square wavelet, sine wave, and sinusoidal wavelet. The aim was to investigate the effects of the waveform on muscle activation magnitude, perceived discomfort, and the relationship between the stimulus signal amplitude and the magnitude of evoked movement. The magnitude of movement was measured offline using video recordings and compared to the magnitude of maximum voluntary movement (MVM) of eyebrows. Results showed that stimulations evoked forehead movement at a magnitude comparable to the MVM in 67% of the participants and close to comparable (80% of the MVM) in 92%. All the waveforms were equally successful in evoking movements. Perceived discomfort did not differ between the waveforms in relation to the movement magnitude, but some individual preferences did exist. Further, regression analysis showed a statistically significant linear relation between stimulation amplitudes and the evoked movement in 98% of the cases. As the waveforms performed equally well in evoking muscle activity, the waveform in pacing systems could be selected by emphasizing technical aspects such as the possibility to suppress stimulation artifacts from simultaneous electromyography measurement.

The authors consider the problem of compressive sensed video recovery via iterative thresholding algorithm. Traditionally, it is assumed that some fixed sparsifying transform is applied at each iteration of the algorithm. In order to improve the recovery performance, at each iteration the thresholding could be applied for different transforms in order to obtain several estimates for each pixel. Then the resulting pixel value is computed based on obtained estimates using simple averaging. However, calculation of the estimates leads to significant increase in reconstruction complexity. Therefore, the authors propose a heuristic approach, where at each iteration only one transform is randomly selected from some set of transforms. First, they present simple examples, when block-based 2D discrete cosine transform is used as the sparsifying transform, and show that the random selection of the block size at each iteration significantly outperforms the case when fixed block size is used. Second, building on these simple examples, they apply the proposed approach when video block-matching and 3D filtering (VBM3D) is used for the thresholding and show that the random transform selection within VBM3D allows to improve the recovery performance as compared with the recovery based on VBM3D with fixed transform.

This paper studies vehicle attribute recognition by appearance. In the literature, image-based target recognition has been extensively investigated in many use cases, such as facial recognition, but less so in the field of vehicle attribute recognition. We survey a number of algorithms that identify vehicle properties ranging from coarse-grained level (vehicle type) to fine-grained level (vehicle make and model). Moreover, we discuss two alternative approaches for these tasks, including straightforward classification and a more flexible metric learning method. Furthermore, we design a simulated real-world scenario for vehicle attribute recognition and present an experimental comparison of the two approaches.

Efficient mitigation of power amplifier (PA) nonlinear distortion in multi-user hybrid precoding based broadband mmWave systems is an open research problem. In this article, we carry out detailed signal and distortion modeling in broadband multi-user hybrid MIMO systems, with a bank of nonlinear PAs in each subarray, while also take the inevitable crosstalk between the antenna/PA branches into account. Building on the derived models, we adopt and describe an efficient closed-loop (CL) digital predistortion (DPD) solution that utilizes only a single-input DPD unit per transmit chain or subarray, despite crosstalk, providing thus substantial complexity-benefit compared to the state-of-the art multi-dimensional DPD solutions. We show that under spatially correlated multipath propagation, each single-input DPD unit can provide linearization towards every intended user, or more generally, towards all spatial directions where coherent propagation is taking place, and that the adopted CL DPD system is robust against crosstalk. Extensive numerical results building on practical measurement-based mmWave PA models are provided, demonstrating and verifying the excellent linearization performance of the overall DPD system in different evaluation scenarios.

In this paper, we propose a novel method for projecting data from multiple modalities to a new subspace optimized for one-class classification. The proposed method iteratively transforms the data from the original feature space of each modality to a new common feature space along with finding a joint compact description of data coming from all the modalities. For data in each modality, we define a separate transformation to map the data from the corresponding feature space to the new optimized subspace by exploiting the available information from the class of interest only. We also propose different regularization strategies for the proposed method and provide both linear and non-linear formulations. The proposed Multimodal Subspace Support Vector Data Description outperforms all the competing methods using data from a single modality or fusing data from all modalities in four out of five datasets.

Shearlet Transform (ST) has been instrumental for the Densely-Sampled Light Field (DSLF) reconstruction, as it sparsifies the underlying Epipolar-Plane Images (EPIs). The sought sparsification is implemented through an iterative regularization, which tends to be slow because of the time spent on domain transformations for dozens of iterations. To overcome this limitation, this letter proposes a novel self-supervised DSLF reconstruction method, CycleST, which employs ST and cycle consistency. Specifically, CycleST is composed of an encoder-decoder network and a residual learning strategy that restore the shearlet coefficients of densely-sampled EPIs using EPI-reconstruction and cycle-consistency losses. CycleST is a self-supervised approach that can be trained solely on Sparsely-Sampled Light Fields (SSLFs) with small disparity ranges (⩽ 8 pixels). Experimental results of DSLF reconstruction on SSLFs with large disparity ranges (16-32 pixels) demonstrate the effectiveness and efficiency of the proposed CycleST method. Furthermore, CycleST achieves ∼ 9x speedup over ST, at least.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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.

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 1}Matlab Code: http://www.cs.tut.fi/~mehta/symdmd.

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.

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.

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.

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.