Self-adaptive Differential Evolution based Optimized MIMO Beamforming 5G Networks

Abstract

     The industrial factory is one of the challenging environments for future wireless communication systems, where the goal is to produce products with low cost in short time. This high level of network performance is achieved by distributing massive MIMO that provides indoor networks with joint beamforming that enhances 5G network capacity and user experience as well. Judging from the importance of this topic, this study introduces a new optimization problem concerning the investigation of multi-beam antenna (MBA) coverage possibilities in 5G network for indoor environments, named Base-station Beams Distribution Problem (BBDP). This problem has an extensive number of parameters and constrains including user’s location, required data rate and number of antenna elements. Thus, BBDP can be considered as NP-hard problem, where complexity increases exponentially as its dimension increases. Therefore, it requires a special computing method that can handle it in a reasonable amount of time. In this study, several differential evolution (DE) variants have been suggested to solve the BBDP problem. The results show that among all DE variants the self-adaptive DE (jDE) can find feasible solutions and outperform the classical ones in all BBDP scenarios with coverage rate of 85% and beam diameter of 500 m.