An Energy-Efficient Hybrid Precoding Based on MPBIL Algorithm for mmWave Massive Mimo Systems

Abstract

To reduce power consumption while ensuring the spectral efficiency of millimeter wave (mmWave) massive multiple-input multiple-output (MIMO) systems, we propose an energy-efficient hybrid precoding based on the improved population based incremental learning (PBIL) algorithm which introduced the random mutation (called as MPBIL hybrid precoding). Specifically, the MPBIL hybrid precoding using switches and inverters in analog part, which can greatly reduce energy consumption. Additionally, we utilize the proposed MPBIL algorithm to iteratively search for the best analog beamformer. The number of iterations is greatly reduced because that the search efficiency of the proposed MPBIL algorithm is significantly improved by introducing the random mutation which increase the diversity of the population. As a result, the proposed MPBIL-based hybrid precoding has lower complexity and higher energy efficiency than some other traditional algorithms.