A neural network for the blind separation of non-Gaussian sources

Author

Freisleben, Bernd ; Hagen, Claudia ; Borschbach, Markus

Author_Institution

Dept. of Electr. Eng. & Comput. Sci., Siegen Univ., Germany

Volume

2

fYear

1998

fDate

4-9 May 1998

Firstpage

837

Abstract

In this paper, a two-layer neural network is presented, which organizes itself to perform blind source separation. The inputs to the network are prewhitened linear mixtures of unknown independent source signals. An unsupervised nonlinear Hebbian learning rule is used for training the network. After convergence, the network is able to extract the source signals from the mixtures, provided that the source signals do not have Gaussian distributions

Keywords

Hebbian learning; convergence; feedforward neural nets; signal detection; unsupervised learning; blind source separation; convergence; independent component analysis; multilayer neural network; nonGaussian sources; nonlinear Hebbian learning; prewhitened linear mixtures; signal extraction; unsupervised learning; Array signal processing; Blind source separation; Computer science; Convergence; Independent component analysis; Neural networks; Principal component analysis; Proposals; Source separation; Unsupervised learning;

fLanguage

English

Publisher

ieee

Conference_Titel

Neural Networks Proceedings, 1998. IEEE World Congress on Computational Intelligence. The 1998 IEEE International Joint Conference on

Conference_Location

Anchorage, AK

ISSN

1098-7576

Print_ISBN

0-7803-4859-1

Type

conf

DOI

10.1109/IJCNN.1998.685876

Filename

685876