Source separation in post-nonlinear mixtures

We address the problem of separation of mutually independent sources in nonlinear mixtures. First, we propose theoretical results and prove that in the general case, it is not possible to separate the sources without nonlinear distortion. Therefore, we focus our work on specific nonlinear mixtures known as post-nonlinear mixtures. These mixtures constituted by a linear instantaneous mixture (linear memoryless channel) followed by an unknown and invertible memoryless nonlinear distortion, are realistic models in many situations and emphasize interesting properties i.e., in such nonlinear mixtures, sources can be estimated with the same indeterminacies as in instantaneous linear mixtures. The separation structure of nonlinear mixtures is a two-stage system, namely, a nonlinear stage followed by a linear stage, the parameters of which are updated to minimize an output independence criterion expressed as a mutual information criterion. The minimization of this criterion requires knowledge or estimation of source densities or of their log-derivatives. A first algorithm based on a Gram-Charlier expansion of densities is proposed. Unfortunately, it fails for hard nonlinear mixtures. A second algorithm based on an adaptive estimation of the log-derivative of densities leads to very good performance, even with hard nonlinearities. Experiments are proposed to illustrate these results.

[1]  G. Darmois,et al.  Analyse générale des liaisons stochastiques: etude particulière de l'analyse factorielle linéaire , 1953 .

[2]  W. J. Hall,et al.  ON CHARACTERIZATION OF THE GAMMA DISTRIBUTION. , 1968 .

[3]  Christian Jutten,et al.  Detection de grandeurs primitives dans un message composite par une architecture de calcul neuromime , 1985 .

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

[5]  J. J. Lacoume,et al.  Sources indentification: a solution based on the cumulants , 1988, Fourth Annual ASSP Workshop on Spectrum Estimation and Modeling.

[6]  Jean-Francois Cardoso,et al.  Source separation using higher order moments , 1989, International Conference on Acoustics, Speech, and Signal Processing,.

[7]  Kurt Hornik,et al.  Multilayer feedforward networks are universal approximators , 1989, Neural Networks.

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

[9]  Dinh Tuan Pham,et al.  Separation of a mixture of independent sources through a maximum likelihood approach , 1992 .

[10]  Dirk Van Compernolle,et al.  Feedforward and feedback in a symmetric adaptive noise canceler : “stability analysis in a simplified case” , 1992 .

[11]  Gilles Burel,et al.  Blind separation of sources: A nonlinear neural algorithm , 1992, Neural Networks.

[12]  Eric Moreau,et al.  New self-adaptative algorithms for source separation based on contrast functions , 1993, [1993 Proceedings] IEEE Signal Processing Workshop on Higher-Order Statistics.

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

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

[15]  Gustavo Deco,et al.  Nonlinear higher-order statistical decorrelation by volume-conserving neural architectures , 1995, Neural Networks.

[16]  Christian Jutten,et al.  Blind source separation for convolutive mixtures , 1995, Signal Process..

[17]  Nathalie Delfosse,et al.  Séparation aveugle adaptative de mélanges convolutifs , 1995 .

[18]  Dimitrios Hatzinakos,et al.  Blind identification of LTI-ZMNL-LTI nonlinear channel models , 1995, IEEE Trans. Signal Process..

[19]  D. Pham.  Séparation aveugle de sources via une analyse en composantes indépendantes , 1995 .

[20]  Nathalie Delfosse,et al.  Adaptive blind separation of independent sources: A deflation approach , 1995, Signal Process..

[21]  Philippe Loubaton,et al.  Subspace method for blind separation of sources in convolutive mixture , 1996, 1996 8th European Signal Processing Conference (EUSIPCO 1996).

[22]  Jean-François Cardoso,et al.  Equivariant adaptive source separation , 1996, IEEE Trans. Signal Process..

[23]  Juha Karhunen,et al.  Neural approaches to independent component analysis and source separation , 1996, ESANN.

[24]  Philippe Loubaton,et al.  Second order blind identification of convolutive mixtures with temporally correlated sources: A subspace based approach , 1996, 1996 8th European Signal Processing Conference (EUSIPCO 1996).

[25]  Ehud Weinstein,et al.  Multichannel signal separation: methods and analysis , 1996, IEEE Trans. Signal Process..

[26]  H. Yangy,et al.  Information Back-propagation for Blind Separation of Sources in Non-linear Mixture , 1997 .

[27]  Yannick Deville,et al.  Optimization of the asymptotic performance of time-domain convolutive source separation algorithms , 1997, ESANN.

[28]  Christian Jutten,et al.  Nonlinear source separation: the post-nonlinear mixtures , 1997, ESANN.

[29]  Shun-ichi Amari,et al.  Adaptive Online Learning Algorithms for Blind Separation: Maximum Entropy and Minimum Mutual Information , 1997, Neural Computation.

[30]  Christian Jutten,et al.  Entropy Optimization - Application to Blind Source Separation , 1997, ICANN.

[31]  Shun-ichi Amari,et al.  Nonlinearity and separation capability: further justification for the ICA algorithm with mixture of densities , 1997, ESANN.

[32]  C. Jutten,et al.  Séparation de sources. Application à la séparation de signaux et de brouilleurs dans un satellite de télécommunications , 1997 .

[33]  Philippe Garat,et al.  Blind separation of mixture of independent sources through a quasi-maximum likelihood approach , 1997, IEEE Trans. Signal Process..

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

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

[36]  Adrian M. Ionescu,et al.  High Performance Magnetic Field Smart Sensor Arrays with Source Separation , 1998 .

[37]  Philippe Loubaton,et al.  Adaptive subspace algorithm for blind separation of independent sources in convolutive mixture , 2000, IEEE Trans. Signal Process..