Title :
A Classical Adaptive Filtering Blind Signal Separation
Author :
ElSabrouty, M. ; Bouchard, Martin ; Aboulnasr, Tyseer
Author_Institution :
Fac. of Inf. & Eng. Technol., German Univ., Cairo
Abstract :
A classical adaptive filtering view of the problem of instantaneous blind signal separation is presented. This classical form enables an easy understanding of the natural gradient algorithm. A new RLS-based algorithm is developed using this classical interpretation. The algorithm provides improved on-line separation speed under the same steady state error compared to the natural gradient algorithm without requiring pre-whitening
Keywords :
adaptive filters; blind source separation; gradient methods; least squares approximations; RLS-based algorithm; blind signal separation; classical adaptive filtering; natural gradient algorithm; on-line separation; steady state error; Adaptive filters; Blind source separation; Cost function; Electronic mail; Information theory; Maximum likelihood estimation; Real time systems; Source separation; Steady-state; System-on-a-chip;
Conference_Titel :
System-on-Chip for Real-Time Applications, The 6th International Workshop on
Conference_Location :
Cairo
Print_ISBN :
1-4244-0898-9
DOI :
10.1109/IWSOC.2006.348258
