Joint precoding and transmit antenna selection with power allocation in multiuser MIMO systems under limited feedback

This paper considers a joint quantized precoding and transmit antenna selection with power allocation problem in the downlink of multiuser multi-input multi-output (MIMO) systems with limited feedback. Our objective is to maximize the capacity, which leads to a mixed integer programming (MIP) problem. To resolve this complicated MIP problem with reasonable cost, we propose a new heterogeneous genetic algorithm (HGA), where each chromosome is divided into a bit string for precoding vector selection, an integer string for transmit antenna selection, and a real number string for power allocation. In addition, new crossover and mutation operations are employed to accommodate these new chromosomes. Conducted simulations show that the proposed HGA provides superior performance compared with previous works.