Energy Efficiency Optimization of Cell-Free Massive MIMO with Zero Forcing Precoding

Abstract

Recently cell-free massive multiple-input multiple-output (CF-MMIMO) systems have attracted a lot of interest. It has been considered one of the main keys of current and next-generation wireless communications networks like 5G and 6G. Effectively, it can handle demand growth and maintain better spectral efficiency (SE). However, the power consumption of a large number of access points (APs) significantly influences the performance which is one of the main points in CF-MMIMO. In this research, improving energy efficiency (EE) is considered while maintaining a high quality of service in the downlink (DL) transmission by optimizing the power allocation of users. This is achieved through the use of the sequential convex approximation method to optimize DL power control coefficients with zero-forcing (ZF) precoding. Moreover, the power consumption of the backhaul lines is minimized by implementing a novel dynamic AP selection method. This method ensures that each user is assigned appropriate APs without quality-of-service degradation and reliable spectral efficiency. Reducing overall power, which leads to improving the EE, is essential for cost savings, better coverage, and carbon emissions. The outcome of the optimized power allocation demonstrates a 43.7% improvement in energy efficiency when comparing the ZF with the conjugate beamforming approach. The results of the suggested AP selection demonstrate a 20.8% improvement in EE compared to the scenario without AP selection. Additionally, there is a 7.2% enhancement in EE compared to a previous study that used fixed AP selection with a signal-to-noise ratio of 10 dB. There is a tradeoff in the total SE when AP selection is used since it tends to decrease. This degradation can be effectively controlled through the careful selection of suitable APs.