A fast algorithm for blind separation of sources based on the cross-cumulant and Levenberg-Marquardt method

Abstract

A new algorithm for the instantaneous mixture of the blind separation of sources problem is derived. This algorithm deals with multi-input multi-output (MIMO) channels. The cost function proposed in this paper can be considered as an extension of that previously proposed by us for only two sources and two sensors. The cost function is based on the cross-cumulant 2/spl times/2 and it is minimized using the Levenberg-Marquardt method. Generally, algorithm convergence is attained in less than 50 iterations and the experimental results are satisfactory even with five stationary sources and two non-stationary sources.