*********************************
There is now a CONTENT FREEZE for Mercury while we switch to a new platform. It began on Friday, March 10 at 6pm and will end on Wednesday, March 15 at noon. No new content can be created during this time, but all material in the system as of the beginning of the freeze will be migrated to the new platform, including users and groups. Functionally the new site is identical to the old one. webteam@gatech.edu
*********************************
Title: Relay-Assisted Hybrid Fiber/mmWave-Wireless Backhaul Network
Committee:
Dr. Douglas Blough, ECE, Chair, Advisor
Dr. Ragupathy Sivakumar, ECE
Dr. Yusun Chang, ECE
Dr. Henry Owen, ECE
Dr. David Goldsman, ISyE
Abstract: The objective of this dissertation is to construct and optimize communication links between densely deployed SBSs in the grid-based 3D hybrid fiber-wireless backhaul networks for 5G communication system. At the same time, the load-balanced routing problem arises in the hybrid backhaul networks is also considered. First of all, in order to construct the long range ultra-high-speed millimeter wave backhaul links, relays are selected between a pair of source and destination base stations to achieve the highest signal-to-noise ratio at the destination. Different algorithms are proposed to select relays from a set of candidate relay locations and find a single optimal-throughput path using different relay strategies. We also design some algorithms to find the high throughput path using limited number of mmWave relay nodes and show how these algorithms can be combined to efficiently find different paths. Secondly, instead of the construction of a single path, we explore the possibility of deploying multiple relay paths between a single pair of base stations in mmWave backhaul scenario. We first formulate the problem of feasibility of multi-path construction as a constraint satisfaction problem and then transform the problem into a Boolean satisfiability problem. Different methods have been used and compared to solve this problem. Lastly, we focus on the load-balanced routing problem in our hybrid backhaul networks with grid topology. In hybrid backhaul architecture where some BSs connect with fiber to the core network and provide mmWave backhaul connections for the rest of the BSs, this architecture brings new challenges, e.g., how to prevent a large amount of traffic from becoming concentrated at certain egress BSs, thereby hurting overall backhaul performance. To address this challenge, we propose an efficient tree-based routing algorithm, which is based on a tree-like topology rooted at a fixed base station (sink node) of the network with a fixed grid topology. Moreover, we also present a variation of the algorithm that permits trade-offs between routing path length and balanced factor.
