Capacity maximization for zero-forcing MIMO-OFDMA downlink systems with multiuser diversity

This paper considers the optimal resource allocation for a multiuser MIMO-OFDMA downlink system using zero-forcing SDMA, with the objective of maximizing the system capacity. We provide the general mathematical formulation of the capacity maximization problem whose resource allocation includes user selection, power allocation and receive signal subspace selection. While the optimal solution of this problem is difficult to obtain, we consider two special cases: user selection based only and antenna selection based only optimization, and develop the optimal solution by reformulating the combinatorial optimization problem into a convex problem. Our proposed algorithms for these special cases can be extended to develop a suboptimal algorithm for the original general optimization problem which allows subspace selection. Moreover, we develop reduced-complexity algorithms for these problems which incur negligible capacity loss. Numerical results and analysis are provided to evaluate the system capacity and to study the effect of the number of users and the receive antennas per user on the performance. Comparisons on the capacity achieved by user selection only, antenna selection only and suboptimal subspace selection are discussed.