Multichannel adaptive blind equalization with CRIMNO-MSE technique

Abstract

A novel adaptive blind equalization algorithm for multichannel systems is presented. The cost function is an augmented version of CRIMNO, which also includes the mean square error (MSE) criterion between the outputs of the multichannel equalizers. It is shown that by forcing the outputs of the equalizers to be equal, as it should be during perfect equalization, the convergence of the CRIMNO-MSE technique is faster than that of the individual single-channel CRIMNO algorithms. Two additional hybrid schemes, which use a finite length memory window to form the MSE of the outputs of the equalizers, are developed. These schemes are called the CRIMNO-MSEM and the CMA-MSEM, where the final M denotes the memory length. Simulation results justify the performance of the new algorithms against single-channel CRIMNO algorithms.<>