Subspace beamforming via block SVD for MIMO systems

In slow-fading multiple-input-multiple-output (MIMO) channels, the weak subchannels play the major role in the degradation of the total performance of the system. In this paper, we propose a method for decoupling strong and weak subchannels of a MIMO channel based on a novel block singular value decomposition (BSVD) algorithm which has a near-optimum performance and less computational cost than the optimum SVD method. The proposed scheme enables us to code across the dominant subchannel of rank n. To evaluate the performance of the proposed scheme, we consider the dominant subchannel of rank 2 in conjunction with Alamouti space-time coding to employ both power gain and diversity gain of MIMO channels. Then we compare it with the optimum SVD method as well as the other space-time codes. Simulation results verify the superiority of the proposed subspace beamforming method.