Adaptive processing of blind source separation through 'ICA with OS'

Blind Source Separation is a problem whose solution is vital in numerous applications in Communications .We are proposing a multistage procedure to separate N original sources from N instantaneous mixtures. The goal is to extract the parameters of the unknown mixture in a deflation approach. In each stage of the procedure a novel cost function is applied . The cost function is derived from the properties of the cdf (cumulative density function) to perform an appropriate independent measure by means of order statistics (unbiased estimator of the cdf).The keypoint of this contribution is the adaptive algorithm applied to optimize our Cost Function using gradient descent techniques. 1.INTRODUCTION The procesing model in a generic NxN scenario is shown in figure 1, where linear instantaneus mixtures of N independent unknown sources are collected by the sensors. A first step performs spatial decorrelation in order to decrease the number of parameters to be found. The global resultant mixture of the spatial decorrelation pre-processing: V = DH is an unknown orthogonal matrix, therefore the second step performs ICA (Independent Component Analysis) in order to update the matrix B until the separation is carried out when B=V.