Hybrid Optimized Channel Compression-Reconstruction Network for RF Chain Selection and Channel Estimation in mm-Wave MIMO Systems

Abstract

A millimeter-wave (mm-wave) massive multiple input multiple output (MIMO) system is considered the safest technique to improve data rate and maintain high communication reliability for future wireless systems. Several studies have attempted to develop a model for improving the power and spectral efficiency of mmWave massive MIMO systems. However, they failed due to the inefficiency of mediating extensive communications. Therefore, this research presents an effective mmWave massive MIMO by applying standard methods to maximize the overall system performance. The main goal of this research is to select the Radio Frequency (RF) chains optimally using the Hybrid Differential Evolution Firefly Optimization (DEFO) algorithm. Then, a strong auto-encoder-driven channel compression network (CCN) and a channel reconstruction network (CRN) model are proposed to perform the channel estimation for RF chain selection. In addition, the quantum beetle swarm optimization (QBSO) algorithm is developed to tune the parameters of CCN and CRN models to accomplish higher precision and faster convergence speed. In the experimental scenario, the efficiency of the proposed model is demonstrated by evaluating and comparing the performance calculations with existing methods. The analysis verified that the proposed model accomplishes higher spectral efficiency of 5.345 bits/s/Hz and 8.67 bps/Hz/W energy efficiency, respectively, for the mmWave massive MIMO system.