Multiuser MISO precoding for sum-rate maximization under multiple power constraints

This paper is concerned with linear precoding designs in a multiuser multiple-input single-output system. With the design objective of maximizing the system sum-rate, we take into consideration multiple linear power constraints at the base-station, including sum, per-antenna, and interference power constraints. We then propose two mean-squared error (MSE)-based precoders, namely minimum MSE (MMSE) and iterative minimization of weighted MSE (IWMMSE) precoders. Both proposed precoding designs are obtained by specialized iterative algorithms. To enforce the multiple power constraints, a certain set of auxiliary variables are introduced and updated iteratively at each algorithm. The proposed precoders are then given in closed-form at each iteration. Convergence of both proposed algorithms is then proved and verified by numerical simulations. Simulation results also show significant enhancements in sum-rate performance by the proposed precoding designs over zero-forcing precoding.