Channel Modelling of a Multi-Cell Environment for Massive MIMO System

Abstract

In this thesis, the intricate dynamics of multi-cell Massive MIMO systems are thoroughly investigated, with particular attention to situations including spatially coupled Rician fading channels. The channel model realistically captures the complex multipath environment foundling real-world wireless communication by including both a random non- line-of-sight (NLoS) element and a fixed line-of-sight (LOS) path. The thorough examination of channel estimation approaches, such as least-squares (LS), element-wise MMSE (EW-MMSE), and minimal mean squared error (MMSE), is a major contribution to this study. With these estimate methods applied to precoding and maximum ratio (MR) combining, closed-form formulas explaining the spectral efficiency (SE) attained in the downlink (DL) and uplink (UL) may be derived. The paper also explores the behavioral analysis of SE related to various channel estimators. The comprehensive investigation’s quantitative results show a continuous pattern: the MMSE estimator performs better than other estimators in terms of SE, making it the better option. Significantly, this performance difference grows with the number of antennas, highlighting the utility of the MMSE estimator for massive MIMO with multi-cell systems via Rician fading channels against spatially correlated Rayleigh fading channels. To sum up, this thesis adds a great deal to our knowledge of Massive MIMO systems by illuminating the interactions between estimating methods, channel properties, and system performance. In addition to advancing theory, the derived closed-form formulas and numerical findings provide useful advice for optimizing wireless networks in the actual world. Through the deciphering of spatially correlated fading channels, this study establishes a strong basis for upcoming advancements in the field of next-generation wireless communication schemes