Title :
Sources indentification: a solution based on the cumulants
Author :
Lacoume, J.J. ; Ruiz, P.
Author_Institution :
CNRS, St. Martin d´´Heres, France
Abstract :
It is noted that the problem of source separation has no solution without a priori information when only the spectral matrix is used. The authors have developed an algorithm using the fourth-order cumulants that solves the problem of the two-source separation without the necessity of a priori information. They show that, in the case of two sources and two sensors, the model is described by two parameters. They then present the identification of these two parameters with the cumulants up to order 4. The potentialities of this source-identification algorithm are illustrated using simulated data
Keywords :
identification; information theory; matrix algebra; parameter estimation; cumulants; parameter identification; source separation; sources identification; spectral matrix; Acoustic sensors; Array signal processing; Filtering; Sensor arrays; Signal processing; Source separation; Statistics; Stochastic processes; Underwater acoustics; Underwater vehicles;
Conference_Titel :
Spectrum Estimation and Modeling, 1988., Fourth Annual ASSP Workshop on
Conference_Location :
Minneapolis, MN
DOI :
10.1109/SPECT.1988.206191