Blind wideband source separation

Author

Nájar, Montse ; Lagunas, Miguel A. ; Bonet, Ignasi

Author_Institution

Dept. of Signal Theory & Commun., Univ. Politecnica de Catalunya, Barcelona, Spain

Volume

iv

fYear

1994

fDate

19-22 Apr 1994

Abstract

This paper deals with the general problem of separating two independent wideband sources when they are mixed by unknown filters. In order to solve this problem, a backward framework is proposed which is composed of two different stages. The first one consists of two linear predictors devoted to improve the source separation, whitening the input signals. Their coefficients are calculated applying the LMS algorithm, minimizing the mean squared errors between the predicted signals and the output of the separation network. The second stage is formed by decoupling filters that have to be blindly estimated imposing an independence criterion to the outputs

Keywords

adaptive estimation; adaptive signal processing; filtering theory; least mean squares methods; prediction theory; LMS algorithm; adaptive signal estimation; backward framework; blind wideband source separation; coefficients; decoupling filters; filters; input signals whitening; linear predictors; mean squared errors; predicted signals; separation network; Apertures; Array signal processing; Filtering theory; Filters; Least squares approximation; Sensor arrays; Source separation; Transfer functions; White noise; Wideband;

fLanguage

English

Publisher

ieee

Conference_Titel

Acoustics, Speech, and Signal Processing, 1994. ICASSP-94., 1994 IEEE International Conference on

Conference_Location

Adelaide, SA

ISSN

1520-6149

Print_ISBN

0-7803-1775-0

Type

conf

DOI

10.1109/ICASSP.1994.389875

Filename

389875