论文信息 - Hierarchic Nonlinear PCA Algorithms for Neural Blind Source Separation

Hierarchic Nonlinear PCA Algorithms for Neural Blind Source Separation

Blind separation of unknown sources from their mixtures is currently a timely research topic in statistical signal processing and unsupervised neural learning. We have recently shown that certain symmetric nonlinear PCA type neural algorithms can be successfully applied to this problem. In this paper, we extend our previous results. We show that several sequential either robust or nonlinear PCA type learning algorithms perform well in the separation problem, too, and may have advantages in certain situations. We also modify some of our algorithms so that they can be used for simultaneous separation of both sub-Gaussian and super-Gaussian sources.

P. Pajunen | J. Karhunen | J. Karhunen | P. Pajunen

[1] Terence D. Sanger,et al. Optimal unsupervised learning in a single-layer linear feedforward neural network , 1989, Neural Networks.

[2] Ehud Weinstein,et al. New criteria for blind deconvolution of nonminimum phase systems (channels) , 1990, IEEE Trans. Inf. Theory.

[3] Juha Karhunen,et al. Generalizations of principal component analysis, optimization problems, and neural networks , 1995, Neural Networks.

[4] Andrzej Cichocki,et al. Neural networks for optimization and signal processing , 1993 .

[5] Bernd Freisleben,et al. A Hierarchical Learning Rule for Independent Component Analysis , 1996, ICANN.

[6] Juha Karhunen,et al. A Unified Neural Bigradient Algorithm for robust PCA and MCA , 1996, Int. J. Neural Syst..

[7] Colin Fyfe,et al. Stochastic ICA Contrast Maximisation Using Oja's Nonlinear PCA Algorithm , 1997, Int. J. Neural Syst..

[8] Juha Karhunen,et al. Nonlinear PCA type approaches for source separation and independent component analysis , 1995, Proceedings of ICNN'95 - International Conference on Neural Networks.