Joint Precoding Scheme with Enhanced Spectral Efficiency for Cell-Free Massive MIMO Downlink

Abstract

The cell-free massive multiple-input multiple-output (MIMO) system is a promising technology for B5G and 6G, it owns large spectral and power efficiency. However, the precoding scheme can affect the system downlink spectral efficiency obviously. Maximum-ratio transmission (MRT) is the most common precoding scheme used in cell-free massive MIMO systems, which maximizes the power of the desired signal, but is unable to suppress interference. Centralized zero-forcing (CZF) is a precoding scheme with strong interference cancellation ability, but can not support too many users because of the high computational complexity and limited front-haul capacity. In order to suppress interference and satisfy the requirement of system scalability, we propose a joint precoding scheme with enhanced spectral efficiency, named Scalable Joint Zero-Forcing Precoding (SJZF), which deploys different zero-forcing precoding schemes with the degree of inter-APs interference. We derive a closed-form expression of the achievable spectral efficiency (SE) for SJZF, under the assumption of independent rayleigh fading channel and pilot contamination. Based on the characteristics of the SE expression, we propose a novel random pilot assignment method, and a max-min fairness power optimization algorithm, to improve the system's SE obviously. Numerical results show that the SE of our SJZF outperforms other conventional precoding schemes, approaching CZF, but have affordable computation complexity.