0473144 20240111140937.520170317235959.9 85013449803 000418581400017 10.1007/978-3-319-53547-0 10 s. P cav_un_epca*0472593978-3-319-53546-30302-9743172-181ChamSpringer2017TichavskýPetrBabaie-ZadehMassoudMichelOlivier J.J.Thirion-MoreauNadege blind source separation independent component analysis autoregressive processes cav_un_auth*0319418 Stochastická informatika Department of Stochastic Informatics SI SI Department of Stochastic Informatics 40 Šembera Ondřej UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0101212 Stochastická informatika Department of Stochastic Informatics SI SI Department of Stochastic Informatics 40 Tichavský Petr UTIA-B Ústav teorie informace a automatizace AV ČR, v. v. i. cav_un_auth*0230113 20 Koldovský Z. CZ http://library.utia.cas.cz/separaty/2017/SI/tichavsky-0473144.pdf 627 kB cav_un_auth*0345929 GA17-00902S GA ČR 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. cav_un_auth*0344250 Latent Variable Analysis and Signal Separation 20170221 20170223 Grenoble FR BB 10000 10100 10103 2018 3 PO RVO:67985556 http://hdl.handle.net/11104/0271358 S 17 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods 3+4 Proceedings Paper Acoustics|Computer Science Theory Methods 0.328 Q2 Q2 2017 85013449803 SCOPUS PUBMED 000418581400017 WOS 627 kB 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 340 Tichavský Petr 340 Babaie-Zadeh Massoud 340 Michel Olivier J.J. 340 Thirion-Moreau Nadege