Presentation 2005-11-17
Blind Souce Separation under a noisy environment Impruvement of subspace method and noise control in real environment
Ryouta SHIMIZU, Toshihiro FURUKAWA, Nari TANABE,
PDF Download Page PDF download Page Link
Abstract(in Japanese) (See Japanese page)
Abstract(in English) Not only in the communication system and the signal processing, but also in blind souce separation in a real environment, the performance of the noise suppression is one of the main research themes. Recently, to this theme, the technique based on the subspace method attracts attention. This method uses both signal subspace and noise subspace which are generated by eigendecomposition with auto correlation matrix of received signal, and the reduction of noise is attempted by projecting the received signal to the signal subspace. Though this technique can reduce the noise in the noise space, it cannot reduce the noise in the signal space. In this paper, we propose the method for solving this problem, and try to use both a past method and the proposed method adaptively according to SNR(Signal to Noise Ratio). As a result, it was confirmed that our methods are effective in a real environment.
Keyword(in Japanese) (See Japanese page)
Keyword(in English) blind souce separation / subspace method / signal subspace / noise subspace / noise control / SNR
Paper # IN2005-102
Date of Issue

Conference Information
Committee IN
Conference Date 2005/11/10(1days)
Place (in Japanese) (See Japanese page)
Place (in English)
Topics (in Japanese) (See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To Information Networks (IN)
Language JPN
Title (in Japanese) (See Japanese page)
Sub Title (in Japanese) (See Japanese page)
Title (in English) Blind Souce Separation under a noisy environment Impruvement of subspace method and noise control in real environment
Sub Title (in English)
Keyword(1) blind souce separation
Keyword(2) subspace method
Keyword(3) signal subspace
Keyword(4) noise subspace
Keyword(5) noise control
Keyword(6) SNR
1st Author's Name Ryouta SHIMIZU
1st Author's Affiliation Tokyo University of Science()
2nd Author's Name Toshihiro FURUKAWA
2nd Author's Affiliation Tokyo University of Science
3rd Author's Name Nari TANABE
3rd Author's Affiliation Tokyo Institute of Technology
Date 2005-11-17
Paper # IN2005-102
Volume (vol) vol.105
Number (no) 408
Page pp.pp.-
#Pages 6
Date of Issue