Kernel-Based Nonlinear Independent Component Analysis for Underdetermined Blind Source Separation


NAIST \| Library	Advanced Search	Japanese \| English

Please use this identifier to cite or link to this item: http://hdl.handle.net/10061/8075

Title:	Kernel-Based Nonlinear Independent Component Analysis for Underdetermined Blind Source Separation
Authors:	Shigeki Miyabe Biing-Hwang (Fred) Juang Hiroshi Saruwatari Kiyohiro Shikano
Keywords:	Blind source separation independent component analysis reproducing kernel Hilbert space underdetermined problem beamforming
Issue Date:	Apr-2009
Publisher:	IEEE
Start page:	1641
End page:	1644
Abstract:	In this paper we propose a new unsupervised training method for nonlinear spatial filter using a new independent component analysis based on kernel infomax. The nonlinearity of the spatial filter used in this paper is equivalent to the integration of beamforming and spectral subtraction, and the whole structure is optimized by independent component analysis in the reproducing kernel Hilbert space. The optimized filter is shown to be capable of achieving better quality output than the conventional method based on time-frequency binary masking.
Description:	ICASSP2009: IEEE International Conference on Acoustics, Speech, and Signal Processing, April 19-24, 2009, Taipei, Taiwan.
URI:	http://hdl.handle.net/10061/8075
ISBN:	9781424423545
ISSN:	1520-6149
Rights:	Copyright 2009 IEEE
Text Version:	Publisher
Publisher DOI:	10.1109/ICASSP.2009.4959915
Appears in Collections:	情報科学研究科 / Graduate School of Information Science

Files in This Item:

File	Size	Format
ICASSP_2009_1461.pdf	605.49 kB	Adobe PDF	View/Open