Title :
Improved contrast dedicated to blind separation in communications
Author :
Comon, Pierre ; Moreau, Eric
Author_Institution :
Thomson Marconi Sonar, Sophia-Antipolis, France
Abstract :
Contrast-based separation of sources have a number of advantages. Among others, they are optimal (in a precise sense) in the presence of noise of unknown statistics. Here a new contrast is proposed that allows not only to obtain the optimal solution analytically, but also yields better performances in terms of variance of the estimated mixing matrix. This contrast needs the source kurtosis to have the same sign, and is thus appropriate to multichannel blind equalization in communications
Keywords :
Gaussian noise; adaptive equalisers; adaptive signal processing; array signal processing; higher order statistics; matrix algebra; blind separation; communications; contrast-based source separation; cumulants; mixing matrix variance; multichannel blind equalization; noise; optimal solution; source kurtosis; Analysis of variance; Blind equalizers; Gaussian noise; Narrowband; Performance analysis; Random processes; Reactive power; Sonar; Statistics; Yield estimation;
Conference_Titel :
Acoustics, Speech, and Signal Processing, 1997. ICASSP-97., 1997 IEEE International Conference on
Conference_Location :
Munich
Print_ISBN :
0-8186-7919-0
DOI :
10.1109/ICASSP.1997.604607
