International Workshop on Hands-Free Speech Communication (HSC2001) April 9-11, 2001 Kyoto, Japan

Fast-Convergence Algorithm for ICA-Based Blind Source Separation Utilizing Array Signal Processing

Toshiya Kawamura, Hiroshi Saruwatari, and Kiyohiro Shikano

Graduate School of Information Science, Nara Institute of Science and Technology, Japan

We propose a new algorithm for blind source separation (BSS), in which independent component analysis (ICA) and beamforming are combined to resolve the low-convergence problem through optimization in ICA. The proposed method consists of the following two parts: frequency-domain ICA with direction-of-arrival (DOA) estimation, and null beamforming based on the estimated DOA. The alternation of learning between ICA and beamforning can realize fast- and high-convergence optimization. The results of the signal separation expenments reveal that the signal separation performance of the proposed algorithm is superior to that of the conventional ICA-based BSS method.

Full Paper

Bibliographic reference.  Kawamura, Toshiya / Saruwatari, Hiroshi / Shikano, Kiyohiro (2001): "Fast-convergence algorithm for ICA-based blind source separation utilizing array signal processing", In HSC2001, 71-74.