Performance Analysis of Multiuser MIMO Scheduling With Full and Limited Feedback

In this paper, we investigate the average capacity and its scaling law of the user scheduling for the muitiuser MIMO systems exploiting multiuser and antenna diversities. We show that the limiting distribution of maximum SNR is of the Gumbel-type and average capacity scaling law follows log log K form. We also consider the effect of scheduling delay on the performance of scheduling scheme. Then, we propose two limited feedback scheduling and obtain their average capacity and show that they match the simulation results very well. In the first approach, the base station (BS) always collects feedback from the user with the best channel condition. We derive the normalized feedback load as well as the threshold value that minimizes it. The second approach employs quantization for each user, so that few bits of SNR are sent to the BS. We show that even with 1-bit quantization, the proposed quantized feedback scheduling can exploit multiuser diversity at the expense of trivial decrease of average capacity.