论文信息 - Reference Based Contrast Functions in a Semi-blind Context

Reference Based Contrast Functions in a Semi-blind Context

We deal with blind signal extraction in the framework of a convolutive mixture of independent sources. Considering so-called reference signals, we generalize former identifiability conditions. Based on this result, we propose to incorporate some a priori information in the references. We show the validity of reference based contrast functions in two semi-blind situations. The results are confirmed by computer simulations.

Marc Castella | Eric Moreau

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

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

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

[4] 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..

[5] 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.