Two blind algorithms of MMSE filter based on second-order statistics in MIMO systems with multiple interferences

In this paper, we consider an MIMO system with multiple interferences. The minimum mean-square error (MMSE) criterion is the best trade off between the rejection of interference and the preservation of user signal in terms of bit error rate (BER). The solution of MMSE filter is known but it requires channel state information which is to be estimated by pilot signals. The estimation tends to be inaccurate or time consuming when the interferences are large. As such, an accurate blind algorithm under heavy interferences would be an attractive alternative approach. In this paper two blind algorithms are presented, one for strong signal and one for strong interferences. The deviation of the mean-square error (MSE) between the ideal MMSE filter and the estimated MMSE filter by the blind algorithms are shown by simulation to be within 0.001.