BLIND IDENTIFICATION AND SIGNAL SEPARATION OF 2-INPUT 2-OUTPUT FIR SYSTEMS

This paper presents a new kind of higher order statistics (HOS) based approaches for blind identification and signal separation of two-input two-output FIR systems. The approaches consist of eigenanalysis of hi gher order spectrum matrices of measurements at different frequency sl ices and then exploiting the structural information of the coupling sy stem to eliminate the identification indeterminacy. It is proved that under some weak restrictions, our approaches can guarantee the uniquen ess of solution and lead to the desired solution, which is quite diffe rent from other existing methods.