Number of the records: 1
Orthogonally-Constrained Extraction of Independent Non-Gaussian Component from Non-Gaussian Background Without ICA
- 1.
SYSNO ASEP 0492879 Document Type C - Proceedings Paper (int. conf.) R&D Document Type Conference Paper Title Orthogonally-Constrained Extraction of Independent Non-Gaussian Component from Non-Gaussian Background Without ICA Author(s) Koldovský, Z. (CZ)
Tichavský, Petr (UTIA-B) RID, ORCID
Ono, N. (JP)Number of authors 3 Source Title Latent Variable Analysis and Signal Separation. - Cham : Springer, 2018 / Deville Yannick ; Gannot Sharon ; Mason Russell ; Plumbley Mark D. ; Ward Dominic - ISSN 0302-9743 - ISBN 978-3-319-93763-2 Pages s. 161-170 Number of pages 10 s. Publication form Online - E Action Latent Variable Analysis and Signal Separation Event date 02.07.2018 - 05.07.2018 VEvent location Guilford Country GB - United Kingdom Event type WRD Language eng - English Country GB - United Kingdom Keywords Independent Component Analysis ; Blind source separation ; blind source extraction Subject RIV BB - Applied Statistics, Operational Research OECD category Statistics and probability R&D Projects GA17-00902S GA ČR - Czech Science Foundation (CSF) Institutional support UTIA-B - RVO:67985556 EID SCOPUS 85048543885 DOI 10.1007/978-3-319-93764-9_16 Annotation We propose a new algorithm for Independent Component Extraction that extracts one non-Gaussian component and is capable to exploit the non-Gaussianity of background signals without decomposing them into independent components. The algorithm is suitable for situations when the signal to be extracted is determined through initialization, it shows an extra stable convergence when the target component is dominant. In simulations, the proposed method is compared with Natural Gradient and One-unit FastICA, and it yields improved results in terms of the Signal-to-Interference ratio and the number of successful extractions. Workplace Institute of Information Theory and Automation Contact Markéta Votavová, votavova@utia.cas.cz, Tel.: 266 052 201. Year of Publishing 2019
Number of the records: 1