Date of Award

3-20-2020

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Information Science and Systems Engineering

First Advisor

Hussain Al-Rizzo

Abstract

Massive Multiple-Input Multiple-Output (MIMO) is a key promising technology for 5G networks that allows a significant increase in spectral and energy efficiency without increasing the spectrum and/or the number of cell sites. However, the computational time and resources required to rigorously analyze the capacity offered by a base station equipped with a large number of antenna elements increases with the increase of the number of antenna elements. A novel approach is presented for modeling large microstrip antenna arrays in base stations of massive MIMO systems. The approach subdivides an M×N array into columns, rows, rectangular, or square subarrays, each consisting of a group of elements. The coupling is rigorously taken into account within each subarray, but it is ignored between subarrays using the array factor. It is shown that the difference in the capacity evaluated using rigorous electromagnetic simulations and the proposed approach is less than 0.79% for both Suburban Macrocell model of the Extended Spatial Channel Model (SCME/SMa) outdoor propagation model and 3D independent identically distributed (i.i.d/3D) model with a significant reduction of 44% and 50%, respectively, in computational time as compared to the full-wave antenna array modeling approach. The algorithm has been demonstrated for open- and closed-loop MIMO systems using several antenna array topologies. Furthermore, the Coefficient of Variation (CoV) is proposed as an alternative performance metric to assess the effects of handset orientation on the capacity of MIMO systems. The 3-D simulated radiation patterns of a base station and handset along with their associated scattering parameters are combined in two anisotropic propagation environments. The capacity is evaluated as the handset rotates about the X-Y-Z axes using standard Euler’s angles. The CoV is numerically derived by rotating the handset over the Euler angles in each direction every 15o about each axis over a full sphere where each rotation involves the creation of numerous instances of the propagation environment depending on the statistical robustness of the results sought. Three array geometries are examined using two different propagation models. The results suggest that the proposed CoV is an effective and versatile metric in selecting the best antenna system design.

Download

Share

COinS