Abstract
In this paper we present a novel neural network topology capable of separating simultaneous signals transferred through a memoryless non-linear path. The employed neural network is a two-layer perceptron that uses parametric non-linearities in the hidden neurons. The non-linearities are formed using a mixture of sigmoidal non-linear functions and present greater adaptation towards separating complex non-linear mixed signals. Simulation results using complex forms of non-linear mixing functions prove the efficacy of the proposed algorithm when compared to similar networks that use standard non-linearities, achieving excellent separation performance and faster convergence behavior.
This is a preview of subscription content, log in via an institution.
Buying options
Tax calculation will be finalised at checkout
Purchases are for personal use only
Learn about institutional subscriptionsPreview
Unable to display preview. Download preview PDF.
References
Haykin, S., Unsupervised Adaptive Filtering, Vol. IBlind Source Separation, J. Wiley & Sons (2000).
Lee, T. W., Independent Component Analysis, Kluwer Academic Publishers (1998).
Karhunen, J., Nonlinear Independent Component Analysis, in Roberts & Everson (eds.), Independent Component Analysis: Principles and Practice, Cambridge University Press (2001).
Burel, G.: Blind separation of sources: A nonlinear neural algorithm. Neural Networks, vol 5. (1992) 937–947.
Pajunen, P., Hyvarinen, A., Karhunen, J.: Nonlinear blind source separation by self-organizing maps. ICONIP Vol 2 (1996) 1207–1210.
Deco, G., Brauer, W.: Nonlinear higher-order statistical decorrelation by volume-conserving architectures. Neural Networks, Vol 8 (1995) 525–535.
Lee, T.W., Koehler, B, Orglmeister, R.: Blind source separation of nonlinear mixing models. Neural Network for signal processing. (1997) 406–415.
Taleb, A., Jutten, C.: Nonlinear source separation: the post-nonlinear mixtures. ES ANN (1997) 279–284.
Yang, H., Amari, S., Cichocki, A.: Information theoretic approach to blind separation of sources in non-linear mixture. Signal Processing, 64 (1998) 291–300.
Author information
Authors and Affiliations
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2001 Springer-Verlag Berlin Heidelberg
About this paper
Cite this paper
Koutras, A., Dermatas, E., Kokkinakis, G. (2001). Neural Network Based Blind Source Separation of Non-linear Mixtures. In: Dorffner, G., Bischof, H., Hornik, K. (eds) Artificial Neural Networks — ICANN 2001. ICANN 2001. Lecture Notes in Computer Science, vol 2130. Springer, Berlin, Heidelberg. https://doi.org/10.1007/3-540-44668-0_79
Download citation
DOI: https://doi.org/10.1007/3-540-44668-0_79
Published:
Publisher Name: Springer, Berlin, Heidelberg
Print ISBN: 978-3-540-42486-4
Online ISBN: 978-3-540-44668-2
eBook Packages: Springer Book Archive