Nonlinear Blind Source Separation Using Hybrid Neural Networks

  • Chun-Hou Zheng
  • Zhi-Kai Huang
  • Michael R. Lyu
  • Tat-Ming Lok
Part of the Lecture Notes in Computer Science book series (LNCS, volume 3971)


This paper proposes a novel algorithm based on minimizing mutual information for a special case of nonlinear blind source separation: post-nonlinear blind source separation. A network composed of a set of radial basis function (RBF) networks, a set of multilayer perceptron and a linear network is used as a demixing system to separate sources in post-nonlinear mixtures. The experimental results show that our proposed method is effective, and they also show that the local character of the RBF network’s units allows a significant speedup in the training of the system.


Mutual Information Radial Basis Function Independent Component Analysis Speech Signal Independent Component Analysis 
These keywords were added by machine and not by the authors. This process is experimental and the keywords may be updated as the learning algorithm improves.


Unable to display preview. Download preview PDF.

Unable to display preview. Download preview PDF.


  1. 1.
    Hyvärinen, A., Karhunen, J., Oja, E.: Independent Component Analysis. J. Wiley, New York (2001)CrossRefGoogle Scholar
  2. 2.
    Hyvärinen, A., Pajunen, P.: Nonlinear Independent Component Analysis: Existence and Uniqueness Results. Neural Networks 12(3), 429–439 (1999)CrossRefGoogle Scholar
  3. 3.
    Huang, D.S.: Systematic Theory of Neural Networks for Pattern Recognition. Publishing House of Electronic Industry of China, Beijing (1996)Google Scholar
  4. 4.
    Huang, D.S.: The United Adaptive Learning Algorithm for the Link Weights and the Shape Parameters in RBFN for Pattern Recognition. International Journal of Pattern Recognition and Artificial Intelligence 11(6), 873–888 (1997)CrossRefGoogle Scholar
  5. 5.
    Taleb, A., Jutten, C.: Source Separation in Post- nonlinear Mixtures. IEEE Trans. Signal Processing 47, 2807–2820 (1999)CrossRefGoogle Scholar
  6. 6.
    Martinez, D., Bray, A.: Nonlinear Blind Source Separation Using Kernels. IEEE Trans. Neural Networks 14(1), 228–235 (2003)CrossRefGoogle Scholar
  7. 7.
    Almeuda, L.B.: MISEP – Linear and Nonlinear ICA Based on Mutual Information. Journal of Machine Learning Research 4(2), 1297–1318 (2003)CrossRefGoogle Scholar

Copyright information

© Springer-Verlag Berlin Heidelberg 2006

Authors and Affiliations

  • Chun-Hou Zheng
    • 1
    • 2
  • Zhi-Kai Huang
    • 1
    • 2
  • Michael R. Lyu
    • 3
  • Tat-Ming Lok
    • 4
  1. 1.Intelligent Computing Lab, Institute of Intelligent MachinesChinese Academy of SciencesHefeiChina
  2. 2.Department of AutomationUniversity of Science and Technology of China 
  3. 3.Computer Science & Engineering Dept.The Chinese University of Hong KongHong Kong
  4. 4.Information Engineering Dept.The Chinese University of Hong KongShatin, Hong Kong

Personalised recommendations