Adaptive Score Functions for Maximum Likelihood ICA

We propose Blind Source Separation (BSS) techniques that are applicable to a wide class of source distributions that may be skewed or symmetric and may even have zero kurtosis. Skewed distributions are encountered in many important application areas such as communications and biomedical signal processing. The methods stem from maximum likelihood approach. Powerful parametric models based on the Extended Generalized Lambda Distribution (EGLD) and the Pearson system are employed in finding the score function. Model parameters are adaptively estimated using conventional moments or linear combinations of order statistics (L-moments). The developed methods are compared with the existing methods quantitatively. Simulation examples demonstrate the good performance of the proposed methods in the cases where the standard Independent Component Analysis (ICA) methods perform poorly.

[1]  J. Hosking.  L‐Moments: Analysis and Estimation of Distributions Using Linear Combinations of Order Statistics , 1990 .

[2]  Jean-Francois Cardoso,et al.  Blind signal separation: statistical principles , 1998, Proc. IEEE.

[3]  J. Cardoso.  Infomax and maximum likelihood for blind source separation , 1997, IEEE Signal Processing Letters.

[4]  Shun-ichi Amari,et al.  Adaptive blind signal processing-neural network approaches , 1998, Proc. IEEE.

[5]  Visa Koivunen,et al.  Source Distribution Adaptive Maximum Likelihood Estimation of ICA Model , 2000 .

[6]  M. Kendall,et al.  Kendall's advanced theory of statistics , 1995 .

[7]  Dinh-Tuan Pham,et al.  Blind separation of instantaneous mixture of sources via an independent component analysis , 1996, IEEE Trans. Signal Process..

[8]  Noboru Murata,et al.  A stable and robust ICA algorithm based on t-distribution and generalized Gaussian distribution models , 1999, Neural Networks for Signal Processing IX: Proceedings of the 1999 IEEE Signal Processing Society Workshop (Cat. No.98TH8468).

[9]  Edward J. Dudewicz,et al.  The extended generalized lambda distribution (EGLD) system for fitting distributions to data with moments, II: tables , 1996 .

[10]  Visa Koivunen,et al.  Pearson System based Method for Blind Separation , 2000 .

[11]  Te-Won Lee,et al.  Independent Component Analysis , 1998, Springer US.

[12]  S. Amari,et al.  Flexible Independent Component Analysis , 1998, Neural Networks for Signal Processing VIII. Proceedings of the 1998 IEEE Signal Processing Society Workshop (Cat. No.98TH8378).

[13]  Aapo Hyvärinen,et al.  Fast and robust fixed-point algorithms for independent component analysis , 1999, IEEE Trans. Neural Networks.

[14]  Sergio Cruces,et al.  Novel blind source separation algorithms using cumulants , 2000, 2000 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings (Cat. No.00CH37100).

[15]  Shun-ichi Amari,et al.  Natural Gradient Works Efficiently in Learning , 1998, Neural Computation.

[16]  M. Kendall,et al.  Kendall's Advanced Theory of Statistics: Volume 1 Distribution Theory , 1987 .

[17]  E. Dudewicz,et al.  The extended generalized lambda distribution system for fitting distributions to data: history, completion of theory, tables, applications, the “final word” on moment fits , 1996 .

[18]  Aapo Hyvärinen,et al.  The Fixed-Point Algorithm and Maximum Likelihood Estimation for Independent Component Analysis , 1999, Neural Processing Letters.

[19]  Andrzej Cichocki,et al.  A New Learning Algorithm for Blind Signal Separation , 1995, NIPS.

[20]  G. Stewart,et al.  Matrix Perturbation Theory , 1990 .