A New Blind Adaptive Signal Separation and Equalization Algorithm for MIMO Convolutive Systems

In this paper, the problem of adaptive blind source separation and equalization for Multiple-Input Multiple-Output (MIMO) convolutive systems is addressed. We propose a new approach based on the constant modulus (CM) and second order criteria, from which a reduced complexity version is derived. The proposed algorithm allows to find an equalizer with a desired delay, which makes it possible to remove the delay ambiguity inherent to blind equalization of MIMO convolutive systems. Furthermore, it was shown that when the delay is chosen in the middle of the interval of all possible delays, we get a better performance. Moreover, a recursive implementation of the proposed criterion via the Recursive Least Square (RLS) algorithm is also presented. The efficiency of the proposed algorithm over the classical constant modulus algorithm is illustrated by simulations.