Efficiency of the spatial signal—interference separator constructed using the mean-square error criterion

Abstract

We analyze the efficiency of interference suppression in the adaptive spatial signal—interference separator for radio communication channels which was constructed using the mean-square error (MSE) criterion. We consider three algorithms: the minimized MSE (MMSE) algorithm, the MMSE algorithm with orthonormalized input signals, and the algorithm for direct inversion of the covariance matrix of input processes (DIM). It is shown that these algorithms are asymptotically optimal and are equivalent to each other. In nonstationary communication channels, the MMSE algorithm is much less efficient than the other two, because of the finite averaging time in the correlators, and the DIM algorithm has somewhat better characteristics of interference suppression. The efficiency of these algorithms is studied by both numerical simulations and hardware modeling.