On the optimality of downlink OFDMA MIMO systems

Motivated by the increasing popularity of OFDMA and MIMO, this paper studies the capacity of MIMO enabled OFDMA system under a generic multiuser multicarrier framework. We show that in terms of total system capacity, OFDMA is the optimal downlink scheme for independent decoding receivers. To achieve the capacity bound however, one must solve a multiuser subcarrier allocation and the optimal power allocation jointly. The computational cost for finding the optimal solution is exponential with respect to the number subcarriers and polynomial with respect to the number of users. To alleviate the complexity, we propose two suboptimal subcarrier allocation criteria, namely, the product-criterion and the sum-criterion, for OFDMA MIMO systems. The computation complexity of the suboptimal approaches grows only linearly with the number of users and the number of subcarriers. Numerical results are presented to compare the two subcarrier allocation criteria with the optimal criterion. Specifically, it is observed that the product-criterion is near optimal in high SNR regions while the sum-criterion approaches optimal in low SNR regions.