Antenna subset selection in MU large-scale MIMO systems

Using a large number of transmit and receive antennas can significantly improve system performance at the price of higher hardware costs. For a large-scale MIMO system, an efficient approach to reduce the hardware cost is antenna selection. In this paper, we propose two fast antenna selection algorithms for wireless large-scale MIMO systems. The first algorithm is aiming to maximize the downlink sum-rate at a low computation complexity. Simulation results show that the proposed algorithm achieves nearly 2 dB gain in sum-capacity than the norm based selection (NBS) algorithm and has lower computational complexity than the optimal exhaust search scheme. The second algorithm minimizes the number of selected transmit antennas while maintaining a pre-defined user rate requirement, so as to reduce the hardware cost to the extreme. The performance of the proposed scheme is verified by simulation.