A New Concept for Separability Problems in Blind Source Separation

The goal of blind source separation (BSS) lies in recovering the original independent sources of a mixed random vector without knowing the mixing structure. A key ingredient for performing BSS successfully is to know the indeterminacies of the problemthat is, to know how the separating model relates to the original mixing model (separability). For linear BSS, Comon (1994) showed using the Darmois-Skitovitch theorem that the linear mixing matrix can be found except for permutation and scaling. In this work, a much simpler, direct proof for linear separability is given. The idea is based on the fact that a random vector is independent if and only if the Hessian of its logarithmic density (resp. characteristic function) is diagonal everywhere. This property is then exploited to propose a new algorithm for performing BSS. Furthermore, first ideas of how to generalize separability results based on Hessian diagonalization to more complicated nonlinear models are studied in the setting of postnonlinear BSS.

[1]  E. Lukács,et al.  A Property of the Normal Distribution , 1954 .

[2]  Fabian J. Theis,et al.  Linear Geometric ICA: Fundamentals and Algorithms , 2003, Neural Computation.

[3]  Christian Jutten,et al.  Space or time adaptive signal processing by neural network models , 1987 .

[4]  Fabian J. Theis,et al.  Nonlinear Geometric ICA , 2003 .

[5]  Juha Karhunen,et al.  Advances in blind source separation (BSS) and independent component analysis (ICA) for nonlinear mixtures , 2004, Int. J. Neural Syst..

[6]  Andrzej Cichocki,et al.  Adaptive blind signal and image processing , 2002 .

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

[8]  Aapo Hyvärinen,et al.  Nonlinear independent component analysis: Existence and uniqueness results , 1999, Neural Networks.

[9]  Andrzej Cichocki,et al.  Adaptive Blind Signal and Image Processing - Learning Algorithms and Applications , 2002 .

[10]  V. Koivunen,et al.  Identifiability and Separability of Linear Ica Models Revisited , 2003 .

[11]  Fabian J. Theis,et al.  Separability of analytic postnonlinear blind source separation with bounded sources , 2004, ESANN.

[12]  Eric Moulines,et al.  A blind source separation technique using second-order statistics , 1997, IEEE Trans. Signal Process..

[13]  Elmar Lang,et al.  Analyzing biomedical signals with probabilistic ICA and kernel-based source density estimation , 2001 .

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

[15]  Pierre Comon.  Independent component analysis - a new concept? signal processing , 1994 .

[16]  Fabian J. Theis,et al.  On model identifiability in analytic postnonlinear ICA , 2005, Neurocomputing.

[17]  Christian Jutten,et al.  A geometric approach for separating post non-linear mixtures , 2002, 2002 11th European Signal Processing Conference.

[18]  Germán Mato,et al.  Asynchronous States and the Emergence of Synchrony in Large Networks of Interacting Excitatory and Inhibitory Neurons , 2003, Neural Computation.

[19]  Lang Tong,et al.  Indeterminacy and identifiability of blind identification , 1991 .

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

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

[22]  Arie Yeredor,et al.  Blind source separation via the second characteristic function , 2000, Signal Process..

[23]  Fabian J. Theis,et al.  On the use of sparse signal decomposition in the analysis of multi-channel surface electromyograms , 2006, Signal Process..

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

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

[26]  A. M. Kagan.  New Classes of Dependent Random Variables and a Generalization of the Darmois–Skitovich Theorem to Several Forms , 1989 .

[27]  Juan K. Lin.  Factorizing Multivariate Function Classes , 1997, NIPS.