Generalized Identifiability Conditions for Blind Convolutive MIMO Separation

This correspondence deals with the problem of source separation in the case where the output of a multivariate convolutive mixture is observed: we propose novel and generalized conditions for the blind identifiability of a separating system. The results are based on higher-order statistics and are valid in the case of stationary but not necessarily i.i.d. signals. In particular, we extend recent results based on second-order statistics only. The approach relies on the use of so called reference signals. Our new results also show that only weak conditions are required on the reference signals: this is illustrated by simulations and opens up the possibility of developing new methods.

[1]  Jean-Christophe Pesquet,et al.  Source separation by quadratic contrast functions: a blind approach based on any higher-order statistics , 2005, Proceedings. (ICASSP '05). IEEE International Conference on Acoustics, Speech, and Signal Processing, 2005..

[2]  Pierre Comon,et al.  Blind Separation of Instantaneous Mixtures of Dependent Sources , 2007, ICA.

[3]  Erkki Oja,et al.  Can we solve the continuous Karhunen-Loève eigenproblem from discrete data? , 1986 .

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

[5]  J. Pesquet,et al.  Generalized contrasts for multichannel blind deconvolution of linear systems , 1997, IEEE Signal Processing Letters.

[6]  Jitendra Tugnait,et al.  Blind identifiability of FIR-MIMO systems with colored input using second order statistics , 2000, IEEE Signal Processing Letters.

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

[8]  Jitendra K. Tugnait,et al.  Identification and deconvolution of multichannel linear non-Gaussian processes using higher order statistics and inverse filter criteria , 1997, IEEE Trans. Signal Process..

[9]  Jean-Christophe Pesquet,et al.  Convolutive Blind Signal Separation Based on Asymmetrical Contrast Functions , 2007, IEEE Transactions on Signal Processing.

[10]  Jean-Christophe Pesquet,et al.  Quadratic Higher Order Criteria for Iterative Blind Separation of a MIMO Convolutive Mixture of Sources , 2007, IEEE Transactions on Signal Processing.

[11]  Chunguo Wu,et al.  A note on equivalence of proper orthogonal decomposition methods , 2003 .

[12]  Philippe Loubaton,et al.  Separation of a class of convolutive mixtures: a contrast function approach , 2001, Signal Process..

[13]  M. Zili.  On the mixed fractional Brownian motion , 2006 .

[14]  Athina P. Petropulu,et al.  A family of frequency- and time-domain contrasts for blind separation of convolutive mixtures of temporally dependent signals , 2005, IEEE Transactions on Signal Processing.

[15]  Mark Girolami,et al.  Orthogonal Series Density Estimation and the Kernel Eigenvalue Problem , 2002, Neural Computation.

[16]  P. Comon,et al.  Contrasts for multichannel blind deconvolution , 1996, IEEE Signal Processing Letters.

[17]  A. Gorokhov,et al.  Subspace-based techniques for blind separation of convolutive mixtures with temporally correlated sources , 1997 .

[18]  Marc Castella,et al.  Reference Based Contrast Functions in a Semi-blind Context , 2009, ICA.

[19]  Jean-Philippe Thiran,et al.  PCA in autocorrelation space , 2002, Object recognition supported by user interaction for service robots.

[20]  Matthias W. Seeger,et al.  Using the Nyström Method to Speed Up Kernel Machines , 2000, NIPS.

[21]  Kenneth Steiglitz,et al.  Series expansion of wide-sense stationary random processes , 1968, IEEE Trans. Inf. Theory.

[22]  Harry van Zanten,et al.  Krein's spectral theory and the Paley-Wiener expansion for fractional Brownian motion , 2005, math/0503656.

[23]  Jean-Christophe Pesquet,et al.  A quadratic MISO contrast function for blind equalization , 2004, 2004 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[24]  Jean-Philippe Thiran,et al.  Pattern recognition using higher-order local autocorrelation coefficients , 2002, Proceedings of the 12th IEEE Workshop on Neural Networks for Signal Processing.

[25]  Thomas Kühn,et al.  Optimal series representation of fractional Brownian sheets , 2002 .

[26]  Mitsuru Kawamoto,et al.  Eigenvector Algorithms Incorporated With Reference Systems for Solving Blind Deconvolution of MIMO-IIR Linear Systems , 2007, IEEE Signal Processing Letters.

[27]  Kacha Dzhaparidze,et al.  A series expansion of fractional Brownian motion , 2002 .

[28]  Pascal Bianchi,et al.  A blind source separation framework for detecting CPM sources mixed by a convolutive MIMO filter , 2006, Signal Process..

[29]  Harry van Zanten,et al.  When is a linear combination of independent fBm’s equivalent to a single fBm? , 2007 .

[30]  J. H. Zanten,et al.  Optimality of an explicit series expansion of the fractional Brownian sheet , 2005 .

[31]  Mitsuru Kawamoto,et al.  Eigenvector Algorithms Using Reference Signals , 2006, 2006 IEEE International Conference on Acoustics Speech and Signal Processing Proceedings.