Transmit Antenna Selection by Group Combination Precoding In Massive MIMO System

In this paper, we propose a new algorithm to select transmit antennas for precoding in a Massive MIMO system. At the Base Station (BS), the antennas are divided into subgroups with smaller dimensions. Then, they are selected based on the evaluation of the corresponding channel capacities. The groups, which give the best sum capacity, are selected for precoding process. To lower the complexity of antenna selection, we adopt the Singular Value Decomposition of channel matrices to compute the channel capacities. After the antennas are selected, the transmitted signal vector is precoded by the conventional Zero Forcing (ZF) and Minimum Mean Square Error (MMSE) precoders before being transmitted to the users. Numerical and simulation results show that the proposed transmit antenna selection algorithm has remarkably lower complexity than the optimal one, yet at the cost of small degradation in the bit error rate (BER) performance of the system.