Decorrelation: sufficient for convolutive blind source separation?

J. Xi, University of Wollongong
T. Mei, Dalian University of Technology, China
J. F. Chicharo, University of Wollongong
F. Yin, Dalian University of Technology, China

Article comments

This paper originally appeared as: Xi, J, Mei, T, Chicharo, JF & Yin, F, Decorrelation: sufficient for convolutive blind source separation?, Proceedings of 2004 International Symposium on Intelligent Multimedia, Video and Speech Processing, 20-22 October 2004, 475-478. Copyright IEEE 2004.

Abstract

This paper considers blind separation of signal sources in a convolutive mixing environment. It tries to show that decorrelation is sufficient for separation of convolutively mixed sources. Two algorithms are also proposed and tested by computer simulations.