Source separation based on second order statistics-an algebraic approach

Author

Lindgren, Ulf ; Van der Veen, Alle-Jan

Author_Institution

Dept. of Appl. Electron., Chalmers Univ. of Technol., Goteborg, Sweden

fYear

1996

fDate

24-26 Jun 1996

Firstpage

324

Lastpage

327

Abstract

Two unknown non-white stochastic sources (e.g. speech signals) are dynamically mixed by an unknown multipath channel and subsequently measured by two sensors. The objective is to construct an inverse filter that separates the two signals, based only on their independence. It is known that, under certain conditions, second-order statistics provide sufficient information to identify the filter. In contrast to the usual cost function optimization techniques, we propose an algorithm that computes the filter coefficients algebraically, using linear algebra techniques such as the singular value decomposition

Keywords

filtering theory; inverse problems; linear algebra; multipath channels; optimisation; singular value decomposition; stochastic processes; algebraic approach; algorithm; array signal processing; cost function optimization; filter coefficients; inverse filter; linear algebra; multipath channel; nonwhite stochastic sources; second order statistics; second-order statistics; sensors; signal separation; singular value decomposition; source separation; speech signals; Cost function; Information filtering; Information filters; Linear algebra; Multipath channels; Nonlinear filters; Source separation; Speech; Statistics; Stochastic processes;

fLanguage

English

Publisher

ieee

Conference_Titel

Statistical Signal and Array Processing, 1996. Proceedings., 8th IEEE Signal Processing Workshop on (Cat. No.96TB10004

Conference_Location

Corfu

Print_ISBN

0-8186-7576-4

Type

conf

DOI

10.1109/SSAP.1996.534882

Filename

534882