Separation of post-nonlinear mixtures using ACE and temporal decorrelation

ABSTRACTWe propose an efficient method based on the concept ofmaximal correlation that reduces the post-nonlinear blindsource separation problem (PNL BSS) to a linear BSS prob-lem. For this we apply the Alternating Conditional Expec-tation (ACE) algorithm – a powerful technique from non-parametricstatistics –toapproximatelyinvertthe (post-)non-linear functions. Interestingly, in the framework of the ACEmethod convergence can be proven and in the PNL BSSscenario the optimal transformation found by ACE will co-incide with the desired inverse functions. After the non-linearities have been removed by ACE, temporal decorrela-tion (TD) allows us to recover the source signals. An ex-cellent performance underlines the validity of our approachand demonstrates the ACE-TD method on realistic exam-ples.1. INTRODUCTIONBlind source separation (BSS) research has mainly beenfocused on variants of linear ICA and temporal decorrela-tion methods (see e.g. [14, 6, 5, 7, 1, 2, 13, 29, 22, 12]).Linear BSS assumes that at time

[1]  J. Friedman,et al.  Estimating Optimal Transformations for Multiple Regression and Correlation. , 1985 .

[2]  Christian Jutten,et al.  Blind separation of sources, part I: An adaptive algorithm based on neuromimetic architecture , 1991, Signal Process..

[3]  W. Härdle.  Applied Nonparametric Regression , 1992 .

[4]  J. Cardoso,et al.  Blind beamforming for non-gaussian signals , 1993 .

[5]  Pierre Comon,et al.  Independent component analysis, A new concept? , 1994, Signal Process..

[6]  Terrence J. Sejnowski,et al.  An Information-Maximization Approach to Blind Separation and Blind Deconvolution , 1995, Neural Computation.

[7]  Gustavo Deco,et al.  Linear redundancy reduction learning , 1995, Neural Networks.

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

[9]  A. Hyvärinen,et al.  Nonlinear Blind Source Separation by Self-Organizing Maps , 1996 .

[10]  Juha Karhunen,et al.  A Maximum Likelihood Approach to Nonlinear Blind Source Separation , 1997, ICANN.

[11]  D. Chakrabarti,et al.  A fast fixed - point algorithm for independent component analysis , 1997 .

[12]  Juan K. Lin,et al.  Faithful Representation of Separable Distributions , 1997, Neural Computation.

[13]  Te-Won Lee,et al.  Blind source separation of nonlinear mixing models , 1997, Neural Networks for Signal Processing VII. Proceedings of the 1997 IEEE Signal Processing Society Workshop.

[14]  Andreas Ziehe,et al.  TDSEP — an efficient algorithm for blind separation using time structure , 1998 .

[15]  Andrzej Cichocki,et al.  Information-theoretic approach to blind separation of sources in non-linear mixture , 1998, Signal Process..

[16]  Andreas Ziehe,et al.  Combining time-delayed decorrelation and ICA: towards solving the cocktail party problem , 1998, Proceedings of the 1998 IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP '98 (Cat. No.98CH36181).

[17]  Ali Mansour,et al.  Blind Separation of Sources , 1999 .

[18]  Christian Jutten,et al.  Source separation in post-nonlinear mixtures , 1999, IEEE Trans. Signal Process..

[19]  A. Yeredor.  Blind separation of Gaussian sources via second-order statistics with asymptotically optimal weighting , 2000, IEEE Signal Processing Letters.

[20]  Pei Ling Lai,et al.  Ica Using Kernel Canonical Correlation Analysis , 2000 .

[21]  Juha Karhunen,et al.  Nonlinear Independent Component Analysis Using Ensemble Learning: Experiments and Discussion , 2000 .

[22]  Terrence J. Sejnowski,et al.  Nonlinear blind source separation using kernel feature spaces , 2001 .

[23]  Constance de Koning,et al.  Editors , 2003, Annals of Emergency Medicine.