A Partial Learning-Based Detection Scheme for Massive MIMO

Massive multiple-input multiple-output (MIMO) is a promising key technology for the fifth-generation (5G) and future mobile wireless network. Although maximum likelihood (ML) detection can get the best detection performance with the lowest bit error rate (BER), its computational complexity significantly increases as the number of antennas increases. Thus, based on neural networks, in this letter we propose a new partial learning (PL)-based detection scheme. Theoretical analyses and simulation results showed that the proposed PL-based detection scheme can achieve low BER with low computational complexity.

I. Introduction

The data rate is expected to be increased by roughly 1000 times for the fifth-generation (5G) wireless networks as compared with that of the fourth-generation (4G) wireless networks [1]. High spectrum efficiency and sufficient spatial freedom make massive multiple-input multiple-output (MIMO) one of the key technologies for 5G and future mobile wireless networks [2], [3].