Performance Bounds for Complex-Valued Independent Vector Analysis

0531483 - ÚTIA 2021 RIV US eng J - Článek v odborném periodiku
Kautský, V. - Tichavský, Petr - Koldovský, Z. - Adali, T.
Performance Bounds for Complex-Valued Independent Vector Analysis.
IEEE Transactions on Signal Processing. Roč. 68, č. 1 (2020), s. 4258-4267. ISSN 1053-587X. E-ISSN 1941-0476
Grant CEP: GA ČR GA17-00902S
Grant ostatní: GA ČR(CZ) GA20-17720S
Institucionální podpora: RVO:67985556
Klíčová slova: Blind source separation * independent component/vector analysis * Cramér-Rao lower bound,
Obor OECD: Electrical and electronic engineering
Impakt faktor: 4.931, rok: 2020
Způsob publikování: Omezený přístup
http://library.utia.cas.cz/separaty/2020/SI/tichavsky-0531483.pdf https://ieeexplore.ieee.org/document/9141450

Independent Vector Analysis (IVA) is a method for joint Blind Source Separation of multiple datasets with wide area of applications including audio source separation, biomedical data analysis, etc. In this paper, identification conditions and Cramér-Rao Lower Bound (CRLB) on the achievable accuracy are derived for the complex-valued case involving circular and non-circular signals and correlated and uncorrelated datasets.The identification conditions describe when independent sources can be separated from their linear mixture in the statistically consistent manner. The CRLB shows how non-Gaussianty, non-circularity of sources and statistical dependence between datasets influence the attainable separation accuracy. Examples presented in the experimental part confirm the validity of the CRLB. Also, they show certain gap between the attainable accuracy and performance of state-of-the-art algorithms,especially, in case of highlynon-circular signals. Hence, there is a room for possible improvements.

Trvalý link: http://hdl.handle.net/11104/0310654