论文信息 - Comparison between SIMO-ICA with least squares criterion and SIMO-ICA with information-geometric learning

Comparison between SIMO-ICA with least squares criterion and SIMO-ICA with information-geometric learning

High-fidelity blind source separation (BSS) using SingleInput Multiple-Output (SIMO)-model-based Independent Component Analysis (SIMO-ICA) is now being studied by the authors. This paper describes a comparison of two types of SIMO-ICAs, SIMO-ICA-LS and SIMOICA-IG, with different constraints, and gives explicit discussion on the sensitivity of the parameters settings in the methods. In order to discuss the difference, the sourceseparation experiments using two SIMO-ICAs are carried out under the same real acoustic conditions. The experiment results reveal that SIMO-ICA-IG outperforms SIMO-ICA-LS, and the parameter setting in SIMO-ICAIG does not depend on the source signals’ properties compared with that of SIMO-ICA-LS.

K. Shikano | T. Takatani | T. Nishikawa | H. Saruwatari

[1] Pierre Comon,et al. Independent component analysis, A new concept? , 1994, Signal Process..

[2] Paris Smaragdis,et al. Blind separation of convolved mixtures in the frequency domain , 1998, Neurocomputing.

[3] Shun-ichi AMARIyy,et al. NATURAL GRADIENT LEARNING WITH A NONHOLONOMIC CONSTRAINT FOR BLIND DECONVOLUTION OF MULTIPLE CHANNELS , 1999 .

[4] Kiyohiro Shikano,et al. Blind source separation for speech based on fast-convergence algorithm with ICA and beamforming , 2001, INTERSPEECH.

[5] K. Shikano,et al. Sound Reproduction System Including Adaptive Compensation of Temperature Fluctuation Effect for Broad-Band Sound Control , 2002, IEICE Transactions on Fundamentals of Electronics Communications and Computer Sciences.

[6] Andrzej Cichocki,et al. Adaptive Blind Signal and Image Processing - Learning Algorithms and Applications , 2002 .

[7] Andrzej Cichocki,et al. Adaptive blind signal and image processing , 2002 .

[8] K. Matsuoka,et al. Minimal distortion principle for blind source separation , 2002, Proceedings of the 41st SICE Annual Conference. SICE 2002..

[9] Kiyohiro Shikano,et al. SIMO-Model-Based Independent Component Analysis for High-Fidelity Blind Separation of Acoustic Signals , 2003 .

[10] Kiyohiro Shikano,et al. High-Fidelity Blind Separation of Acoustic Signals using SIMO-Model-based ICA with Information-Geometric Learning , 2003 .