Multichannel blind deconvolution of arbitrary signals: adaptive algorithms and stability analyses

We derive three computationally-efficient algorithms for the multichannel blind deconvolution of arbitrary non-Gaussian source mixtures. Two of the algorithms are spatio-temporal extensions of previously-derived blind signal separation algorithms that combine kurtosis-based contrast function optimization with output whitening. We analyze the local stability properties of the algorithms to determine the constraints on the step sine parameters to allow the separation and deconvolution of arbitrary signal mixtures. These analyses and supporting simulations illustrate the abilities of the new algorithms to effectively deconvolve and separate mixtures of arbitrarily-distributed sources.

[1]  Sun-Yuan Kung,et al.  Adaptive paraunitary filter banks for principal and minor subspace analysis , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

[2]  Scott C. Douglas,et al.  Self-stabilized gradient algorithms for blind source separation with orthogonality constraints , 2000, IEEE Trans. Neural Networks Learn. Syst..

[3]  S. Amari,et al.  Multichannel blind separation and deconvolution of sources with arbitrary distributions , 1997, Neural Networks for Signal Processing VII. Proceedings of the 1997 IEEE Signal Processing Society Workshop.

[4]  Y. Inouye.  Criteria for blind deconvolution of multichannel linear time-invariant systems , 1998, IEEE Trans. Signal Process..

[5]  C. Papadias,et al.  A constant modulus algorithm for multiuser signal separation in presence of delay spread using antenna arrays , 1997, IEEE Signal Processing Letters.

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

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

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

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

[10]  S.C. Douglas,et al.  Multichannel blind deconvolution and equalization using the natural gradient , 1997, First IEEE Signal Processing Workshop on Signal Processing Advances in Wireless Communications.

[11]  Andrzej Cichocki,et al.  Self-whitening algorithms for adaptive equalization and deconvolution , 1999, IEEE Trans. Signal Process..