bibtype	C - Conference Paper (international conference)
ARLID	0473144
utime	20240111140937.5
mtime	20170317235959.9
SCOPUS	85013449803
WOS	000418581400017
DOI	10.1007/978-3-319-53547-0
title (primary) (eng)	Adaptive Blind Separation of Instantaneous Linear Mixtures of Independent Sources
specification	page_count 10 s. media_type P
serial	ARLID cav_un_epca*0472593 ISBN 978-3-319-53546-3 ISSN 0302-9743 title Latent Variable Analysis and Signal Separation, 13th International Conference, LVA/ICA 2017 page_num 172-181 publisher place Cham name Springer year 2017 editor name1 Tichavský name2 Petr editor name1 Babaie-Zadeh name2 Massoud editor name1 Michel name2 Olivier J.J. editor name1 Thirion-Moreau name2 Nadege
keyword	blind source separation
keyword	independent component analysis
keyword	autoregressive processes
author (primary)	ARLID cav_un_auth*0319418 full_dept (cz) Stochastická informatika full_dept (eng) Department of Stochastic Informatics department (cz) SI department (eng) SI full_dept Department of Stochastic Informatics share 40 name1 Šembera name2 Ondřej institution UTIA-B fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
author	ARLID cav_un_auth*0101212 full_dept (cz) Stochastická informatika full_dept Department of Stochastic Informatics department (cz) SI department SI full_dept Department of Stochastic Informatics share 40 name1 Tichavský name2 Petr institution UTIA-B fullinstit Ústav teorie informace a automatizace AV ČR, v. v. i.
author	ARLID cav_un_auth*0230113 share 20 name1 Koldovský name2 Z. country CZ
source	url http://library.utia.cas.cz/separaty/2017/SI/tichavsky-0473144.pdf source_size 627 kB
cas_special	project ARLID cav_un_auth*0345929 project_id GA17-00902S agency GA ČR abstract (eng) In many applications, there is a need to blindly separate independent sources from their linear instantaneous mixtures while the mixing matrix or source properties are slowly or abruptly changing in time. The easiest way to separate the data is to consider off-line estimation of the model parameters repeatedly in time shifting window. Another popular method is the stochastic natural gradient algorithm, which relies on non-Gaussianity of the separated signals and is adaptive by its nature. In this paper, we propose an adaptive version of two blind source separation algorithms which exploit non-stationarity of the original signals. The results indicate that the proposed algorithms slightly outperform the natural gradient in the trade-off between the algorithm’s ability to quickly adapt to changes in the mixing matrix and the variance of the estimate when the mixing is stationary. action ARLID cav_un_auth*0344250 name Latent Variable Analysis and Signal Separation dates 20170221 mrcbC20-s 20170223 place Grenoble country FR RIV BB FORD0 10000 FORD1 10100 FORD2 10103 reportyear 2018 num_of_auth 3 presentation_type PO inst_support RVO:67985556 permalink http://hdl.handle.net/11104/0271358 confidential S article_num 17 mrcbC86 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods mrcbC86 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods mrcbC86 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods mrcbT16-s 0.328 mrcbT16-4 Q2 mrcbT16-E Q2 arlyear 2017 mrcbU14 85013449803 SCOPUS mrcbU24 PUBMED mrcbU34 000418581400017 WOS mrcbU56 627 kB mrcbU63 cav_un_epca*0472593 Latent Variable Analysis and Signal Separation, 13th International Conference, LVA/ICA 2017 978-3-319-53546-3 0302-9743 1611-3349 172 181 Cham Springer 2017 Lecture Notes in Computer Science 10169 mrcbU67 340 Tichavský Petr mrcbU67 340 Babaie-Zadeh Massoud mrcbU67 340 Michel Olivier J.J. mrcbU67 340 Thirion-Moreau Nadege