High interference rejection rate achieved through an iterative signal separation

We show how high interference rejection rate can be achieved using an iterative blind signal separation technique. The proposed approach relies on weighted nonlinear functions. The weights are chosen according to a priori information on the desired signal. Closed form expression of the interference rejection rate is given via an asymptotic performance analysis. This analysis shows how the interference rejection can be improved by choosing appropriate weights of the non-linear functions. This contribution is completed by some simulations.