Committee:

Dr. Douglas Blough, ECE, Chair , Advisor

Dr. Ragupathy Sivakumar, ECE

Dr. Gee-Kung Chang, ECE

Dr. Yusun Chang, ECE

Dr. Dave Goldsman, ISyE

Abstract:

The objective of this dissertation is to design and analyze the relay-assisted mmWave backhaul network in the urban environment for 5G communication systems.  First, for constructing a multi-hop relay path between a pair of BSs in the dense urban area, a novel algorithm is proposed to select relays from a set of candidate relay locations so that ultra high end-to-end throughput can be maintained to support the large backhaul traffic demand. Second, to build a relay-assisted mmWave backhaul network with a tree topology, which consists a set of pre-defined logical links between BSs, the relay selection algorithm is enhanced to not only consider controlling the intra-path mutual interference, but also have the ability to minimize the inter-path mutual interference as well. Several methods are also proposed to improve the feasibility of finding a solution where all logical links are interference-minimal. Third, upon a new interference model, through linear programming, the traffic demand of small-cell BSs can be maximized in the mmWave backhaul network in either the downlink or uplink case. An efficient distributed scheduling  algorithm is also proposed, which can be deployed easily in the practical system. Fourth, as for deploying mmWave backhaul networks along the urban roadside environment to provide 5G service to vehicles, a novel interference-minimal backhaul topology is proposed. Fifth, to enable the simulation of mmWave out-of-band backhaul networks in ns-3, the design of a new mmWave backhaul module in ns-3 is proposed. On the other hand, the existing mmWave in-band backhaul module in ns-3 is also extended to support the simulation of basic functionalities of the mmWave out-of-band backhaul network.