Generalized MMSE Beamforming for Downlink MIMO Systems

In this paper, a generalized MMSE beamforming is proposed for downlink multiple input multiple output (MIMO) systems. Unlike the conventional MMSE beamforming, cellular environments are considered where each user is randomly distributed in a cell and has the different temporal correlation of fading channel. We apply the proposed generalized MMSE beamforming to two important downlink MIMO scenarios, which are multiuser MIMO (MU-MIMO) and multicell MIMO (MC-MIMO). We derive the closed-form expressions of the generalized MMSE beamforming for both scenarios using convex optimization technique. The derived beamforming solution captures the impacts of the random geometry and different temporal correlations of users. Numerical results verify that the generalized MMSE beamforming achieves lower BER and higher average sum-rate than the previous beamforming schemes developed for MU-MIMO and MC-MIMO including the conventional MMSE beamforming.