Abstract
We study the performance of antenna selection-based multiple-input-multiple-output networks with large-but-finite number of transmit antennas and receivers. Considering the continuous and bursty communication scenarios with different users' data request probabilities, we develop an efficient antenna selection scheme using genetic algorithms (GAs). As demonstrated, the proposed algorithm is generic in the sense that it can be used in the cases with different objective functions, precoding methods, levels of available channel state information, and channel models. Our results show that the proposed GA-based algorithm reaches (almost) the same throughput as the exhaustive search-based optimal approach, with substantially less implementation complexity.