Spectral-Efficiency Analysis of Massive MIMO Systems in Centralized and Distributed Schemes

TL;DR

This paper develops a comprehensive channel model and analyzes the spectral efficiency of massive MIMO systems in centralized and distributed setups, providing insights into optimal antenna configurations and validating results with simulations.

Contribution

It introduces a novel closed-form channel model for both C-MIMO and D-MIMO, and derives exact asymptotic spectral efficiency expressions for these configurations.

Findings

Optimal antenna array radius for D-MIMO is approximately the cell radius divided by 1.31.

The proposed model accurately predicts spectral efficiency in various environmental conditions.

Simulation results confirm the theoretical analysis and optimal configurations.

Abstract

This paper analyzes the spectral efficiency of massive multiple-input multiple-output (MIMO) systems in both centralized and distributed configurations, referred to as C-MIMO and D-MIMO, respectively. By accounting for real environmental parameters and antenna characteristics, namely, path loss, shadowing effect, multi-path fading and antenna correlation, a novel comprehensive channel model is first proposed in closed-form, which is applicable to both types of MIMO schemes. Then, based on the proposed model, the asymptotic behavior of the spectral efficiency of the MIMO channel under both the centralized and distributed configurations is analyzed and compared in exact forms, by exploiting the theory of very long random vectors. Afterwards, a case study is performed by applying the obtained results into MIMO networks with circular coverage. In such a case, it is attested that for the D-MIMO of cell radius $r_{\mathrm{c}}$ and circular antenna array of radius~$r_{\mathrm{a}}$, the optimal value of~$r_{\mathrm{a}}$ that maximizes the average spectral efficiency is accurately established by $r_{\mathrm{a}}^{\mathrm{opt}}=r_{\mathrm{c}}/1.31$. Monte Carlo simulation results corroborate the developed spectral-efficiency analysis.