Improved Precoding Scheme for MmWave Large-Scale MIMO Systems

Abstract

Millimeter wave (mmWave) system has been considered as a key technology in 5G, due to its large available bandwidth and high spectrum efficiency. For such system, the traditional full digital precoding is unrealistic due to the costs and complexity of radio-frequency(RF) chains. To overcome this difficulty, hybrid precoding is proposed. However, most existing hybrid precoding still involves high energy consumption. In this paper, we propose an improved hybrid precoding scheme based on energy-efficient switch and inverter (SI)-based precoding architecture. The improved hybrid precoding is aimed at adaptively updated the probability distribution of elements in hybrid precoder by minimizing cross entropy(CE), and reach a higher achievable sum rate by weighting. Simulation results prove that the proposed improved hybrid precoding scheme performs better than before.