Frequency-domain contrast functions for separation of convolutive mixtures

This paper addresses the problem of blind separation of convolutive mixtures via contrast maximization. New frequency-domain contrast functions are constructed based on second and higher-order spectra of the observations. They allow one to separate mixtures of sources which are spatially independent, and temporally possibly non i.i.d. linear or non-linear processes. The proposed criteria provide a framework for extending to the convolutive case contrasts that have been proposed in the context of instantaneous mixtures.

[1]  Yujiro Inouye,et al.  Cumulant-based blind identification of linear multi-input-multi-output systems driven by colored inputs , 1997, IEEE Trans. Signal Process..

[2]  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..

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

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

[5]  E. Moreau,et al.  New criteria for blind signal separation , 2000, Proceedings of the Tenth IEEE Workshop on Statistical Signal and Array Processing (Cat. No.00TH8496).

[6]  Philippe Loubaton,et al.  Separation of a class of convolutive mixtures: a contrast function approach , 1999, 1999 IEEE International Conference on Acoustics, Speech, and Signal Processing. Proceedings. ICASSP99 (Cat. No.99CH36258).

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

[8]  Ananthram Swami,et al.  Multichannel ARMA processes , 1994, IEEE Trans. Signal Process..