A Random Beamforming Technique in Multiuser Multi-Antenna OFDM Systems For Large System Capacity and Fairness Among Users

We propose a random beamforming technique in multiuser multi-antenna orthogonal frequency division multiplexing (OFDM) systems for a large system capacity and fairness among users. In the proposed technique, we allocate several beams on each subcarrier to users based on signal-to-interference plus noise ratio (SINR) and all the remaining beams on each subcarrier by the proportional fairness (PF) algorithm. We also use the random unitary matrices corresponding to the beams and subcarriers where each user with low capacity has its highest SINR in the next time slot. We show that compared to the conventional techniques, such as the SINR-based scheduling scheme and the PF scheduling scheme, the proposed technique achieves high system capacity with about the same variance of user capacity.