On the indeterminacies of convolutive blind signal separation based on second order statistics

Author

Schobben, Daniël W E ; Sommen, Piet C W

Author_Institution

Eindhoven Univ. of Technol., Netherlands

Volume

1

fYear

1999

fDate

1999

Firstpage

215

Abstract

Several blind signal separation algorithms have been developed which are based on second order statistics. Little has been published however on whether second order statistics are sufficient to obtain a unique solution. Especially for applications that involve convolutive mixing and unmixing of signals that are correlated in time, there is a lack of knowledge on why and in what cases second order statistics suffice. This paper investigates the indeterminacies that are introduced when second order statistics are used and presents a theorem for the unmixing system to be uniquely found using second order statistics

Keywords

convolution; correlation methods; filtering theory; matrix algebra; statistical analysis; blind signal separation algorithms; convolutive blind signal separation; convolutive signals mixing; convolutive signals unmixing; indeterminacies; mixing filters; second order statistics; source correlation matrix rank; time correlated signals; unmixing filters; unmixing system; Australia; Blind source separation; Higher order statistics; Maximum likelihood estimation; Mutual information; Sensor systems; Signal processing algorithms; Source separation; Uniform resource locators;

fLanguage

English

Publisher

ieee

Conference_Titel

Signal Processing and Its Applications, 1999. ISSPA '99. Proceedings of the Fifth International Symposium on

Conference_Location

Brisbane, Qld.

Print_ISBN

1-86435-451-8

Type

conf

DOI

10.1109/ISSPA.1999.818151

Filename

818151