LMMSE Receivers in Uplink Massive MIMO Systems With Correlated Rician Fading

Abstract

We carry out a theoretical analysis of the uplink \n(UL) of a massive MIMO system with per-user channel correlation \nand Rician fading, using two processing approaches. Firstly, \nwe examine the linear-minimum-mean-square-error receiver under \ntraining-based imperfect channel estimates. Secondly, we \npropose a statistical combining technique that is more suitable \nin environments with strong Line-of-Sight (LoS) components.We \nderive closed-form asymptotic approximations of the UL spectral \nefficiency (SE) attained by each combining scheme in single \nand multi-cell settings, as a function of the system parameters. \nThese expressions are insightful in how different factors such as \nLoS propagation conditions and pilot contamination impact the \noverall system performance. Furthermore, they are exploited \nto determine the optimal number of training symbols which \nis shown to be of significant interest at low Rician factors. \nThe study and numerical results substantiate that stronger LoS \nsignals lead to better performances, and under such conditions, \nthe statistical combining entails higher SE gains than the \nconventional receiver.