Title :
On the stability of source separation algorithms
Author :
Cardoso, Jean-Francois
Author_Institution :
CNRS, Paris, France
fDate :
31 Aug-2 Sep 1998
Abstract :
This paper presents a unified view, new results and interpretations about the stability conditions of algorithms for independent component analysis and blind separation of signals
Keywords :
learning (artificial intelligence); maximum likelihood detection; maximum likelihood estimation; neural nets; numerical stability; probability; statistical analysis; blind source separation; independent component analysis; learning rules; maximum likelihood estimation; probability; signal separation; stability; Blind source separation; Convergence; Independent component analysis; Matrix decomposition; Maximum likelihood estimation; Neural networks; Probability density function; Signal processing algorithms; Source separation; Stability analysis;
Conference_Titel :
Neural Networks for Signal Processing VIII, 1998. Proceedings of the 1998 IEEE Signal Processing Society Workshop
Conference_Location :
Cambridge
Print_ISBN :
0-7803-5060-X
DOI :
10.1109/NNSP.1998.710625
