Tampere University of Technology

TUTCRIS Research Portal

Optimized fast convolution based filtered-OFDM processing for 5G

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


Original languageEnglish
Title of host publicationProceedings of the European Conference on Networks and Communications (EuCNC): Physical Layer and Fundamentals (PHY)
Number of pages6
ISBN (Electronic)978-1-5386-3873-6
Publication statusPublished - 2017
Publication typeA4 Article in a conference publication
EventEuropean Conference on Networks and Communications - Oulu City Theatre, Oulu, Finland
Duration: 12 Jun 201715 Jun 2017
Conference number: 26


ConferenceEuropean Conference on Networks and Communications
Abbreviated titleEuCNC
Internet address


This paper investigates the application of flexible fast-convolution (FC) filtering scheme for multiplexing orthogonal frequency-division multiplexing (OFDM) physical resource blocks (PRBs) in a spectrally well-localized manner. This scheme is able to suppress interference leakage between adjacent PRBs, thus supporting independent waveform parametrization and numerologies for different PRBs, as well as asynchronous multiuser operation. These are considered as important features in the 5G waveform development. This contribution focuses on optimizing FC based OFDM transmultiplexers such that the in-band interference is minimized subject to the given out-of-band emission constraint. The performance of the optimized designs is demonstrated using resource block groups (RBGs) of different sizes and with various design parameters. The proposed scheme has great flexibility in tuning the filtering bandwidths dynamically according the resource allocation to different users with different requirements regarding the OFDM waveform numerology. Also the computational complexity is competitive with existing time-domain filtering approaches and becomes superior when the number of filtering bands is increased.

Publication forum classification

Downloads statistics

No data available