Improved Multiplicative Orthogonal-Group Based ICA for Separating Mixed Sub-Gaussian and Super-Gaussian Sources

Recently, the fully-multiplicative orthogonal-group ICA (OgICA) neural algorithm has been proposed, which exploits the known principle of diagonalisation of a tensor of a warped network's outputs. Unfortunately, the algorithm is only able to separate sub-Gaussian source signals. To address this...

Full description

Saved in:
Bibliographic Details
Published in2006 International Conference on Communications, Circuits and Systems Vol. 1; pp. 340 - 343
Main Authors Yalan Ye, Zhi-Lin Zhang, Shaozhi Wu, Xiaobin Zhou
Format Conference Proceeding
LanguageEnglish
Published IEEE 01.06.2006
Subjects
Application software
Blind source separation
Computer science
Computer simulation
Convergence
Independent component analysis
Iterative algorithms
Source separation
Switches
Tensile stress
Online AccessGet full text

Cover

More Information
Summary:Recently, the fully-multiplicative orthogonal-group ICA (OgICA) neural algorithm has been proposed, which exploits the known principle of diagonalisation of a tensor of a warped network's outputs. Unfortunately, the algorithm is only able to separate sub-Gaussian source signals. To address this problem, the paper proposes an improved algorithm that adopts two nonlinearities and a flexible nonlinear model switching technique. The improved OgICA algorithm can instantaneously separate not only the mixture of pure sub-Gaussian source signals, but also the mixture of super-Gaussian and sub-Gaussian source signals. Besides, the algorithm has fast convergence speed and high separation performance. The validity and effectiveness of our proposed algorithm are confirmed through extensive computer simulations
ISBN:9780780395848
0780395840
DOI:10.1109/ICCCAS.2006.284649