Separation of sources using and competitive simulated annealing learning

C.G. Puntoneta 1 A. Mansour 2, C. Bauer 3,  E. Lang 4

1- department of Architecture and Computer Technology,
University of Granada, Granada, Spain

2- Bio-Mimetic Control Research Center (RIKEN),
Nagoya-ship Japan

3- department of Biophysics, University of Regensburg,
Regensburg, Germany

Received 10 January 2001; accepted 22 October 2001

Abstract
This paper presents a new adaptive procedure for the linear and non-linear separation
of signals with non-uniform, symmetrical probability distributions, based
on both simulated annealing and competitive learning methods by means of
a neural network, considering the properties of the vectorial spaces of
sources and mixtures, and using a multiple linearization in the mixture
space. The main characteristics of the method are its simplicity and the
rapid convergence experimentally validated by the separation of many kinds
of signals, such as speech or biomedical data.

 @ 2002 Elsevier Science B.V. All rights reserved.
 
Keywords.. Blind separation; Independent component analysis;
Simulated annealing; Competitive learning; Neural networks