Blind separation of convolved sources based on information maximization

Author

Torkkola, Kari

Author_Institution

Phoenix Corp. Res. Lab., Motorola Inc., Tempe, AZ, USA

fYear

1996

fDate

4-6 Sep 1996

Firstpage

423

Lastpage

432

Abstract

Blind separation of independent sources from their convolutive mixtures is a problem in many real world multi-sensor applications. In this paper we present a solution to this problem based on the information maximization principle, which was proposed by Bell and Sejnowski (1995) for the case of blind separation of instantaneous mixtures. We present a feedback network architecture capable of coping with convolutive mixtures, and we derive the adaptation equations for the adaptive filters in the network by maximizing the information transferred through the network. Examples using speech signals are presented to illustrate the algorithm

Keywords

adaptive filters; convolution; information theory; recurrent neural nets; signal reconstruction; blind separation; convolutive mixtures; convolved sources; feedback network architecture; information maximization; instantaneous mixtures; multi-sensor applications; speech signals; Acoustic noise; Adaptive filters; Equations; Higher order statistics; Independent component analysis; Laboratories; Signal processing algorithms; Source separation; Speech; Vectors;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks for Signal Processing [1996] VI. Proceedings of the 1996 IEEE Signal Processing Society Workshop

Conference_Location

Kyoto

ISSN

1089-3555

Print_ISBN

0-7803-3550-3

Type

conf

DOI

10.1109/NNSP.1996.548372

Filename

548372