Title :
Approximate maximum likelihood source separation using the natural gradient
Author :
Choi, Seungjin ; Cichocki, Andrzej ; Zhang, Liqing ; Amari, Shunichi
Author_Institution :
Dept. of Comput. Sci. & Eng., POSTECH, South Korea
Abstract :
This paper addresses a maximum likelihood approach to source separation in the case of overdetermined mixtures corrupted by additive white Gaussian noise. We present an objective function that is an approximate likelihood function based on the Laplace approximation. Then we derive a natural gradient adaptation algorithm which maximizes the corresponding approximate likelihood function. Useful behavior of the proposed method is verified by numerical experiments
Keywords :
AWGN; approximation theory; gradient methods; maximum likelihood detection; optimisation; Laplace approximation; additive white Gaussian noise; approximate maximum likelihood source separation; maximization; natural gradient adaptation algorithm; objective function; overdetermined mixtures; Additive white noise; Artificial intelligence; Computer science; Context; Information systems; Maximum likelihood estimation; Signal processing; Signal processing algorithms; Source separation; Statistical analysis;
Conference_Titel :
Wireless Communications, 2001. (SPAWC '01). 2001 IEEE Third Workshop on Signal Processing Advances in
Conference_Location :
Taiwan
Print_ISBN :
0-7803-6720-0
DOI :
10.1109/SPAWC.2001.923891
