A smart multi-user massive MIMO system for next G Wireless communications using evolutionary optimized antenna selection

Abstract

Massive multi-user multiple input multiple output is a very promising technique for next generation communication. It can provide further improvement to the wireless communication link performance due to relatively large number of transmitting antennas equipped at the base station. This large number has the potential to improve the performance but these systems suffer from high cost, complexity and large size. The transmit antenna selection (TAS) can be employed to solve these problems and with the objective of maximizing the achievable ergodic capacity. In this paper, The TAS problem is solved using a modified evolutionary algorithm, in particular, the chaotic binary particle swarm optimization algorithm is utilized for maximization of the total achievable capacities with reduced system complexity and minimized hardware cost. The multi-user is supported using the zero-forcing baseband beamforming. The convergence of the proposed evolutionary algorithm is proved and its performance is evaluated using numerical analysis. The presented results proved that the proposed evolutionary algorithm can achieve competitive ergodic capacities while utilizing small number of radio frequency chains. In addition, the proposed technique is better than random and maximum norm TAS. It can achieve near optimal performance that can be achieved by exhaustive search TAS but with reduced computational complexity.