Date of Award

5-31-2020

Document Type

Dissertation

Degree Name

Doctor of Philosophy in Electrical Engineering - (Ph.D.)

Department

Electrical and Computer Engineering

First Advisor

Nirwan Ansari

Second Advisor

Abdallah Khreishah

Third Advisor

Qing Liu

Fourth Advisor

Yuanqiu Luo

Fifth Advisor

Roberto Rojas-Cessa

Abstract

The fifth generation wireless network is designed to accommodate enormous traffic demands for the next decade and to satisfy varying quality of service for different users. Drone-mounted base stations (DBSs) characterized by high mobility and low cost intrinsic attributes can be deployed to enhance the network capacity. In-band full-duplex (IBFD) is a promising technology for future wireless communications that can potentially enhance the spectrum efficiency and the throughput capacity. Therefore, the following issues have been identified and investigated in this dissertation in order to achieve high spectrum efficiency and high user quality of service.

First, the problem of deploying DBSs is studied. Deploying more DBSs may increase the total throughput of the network but at the expense of the operation cost. The droNe-mounted bAse station PlacEment (NAPE) problem with consideration of IBFD communications and DBS backhaul is then formulated. The objective is to minimize the number of deployed DBSs while maximizing the total throughput of the network by incorporating IBFD-enabled communications for both access links and backhaul links via DBSs as relay nodes. A heuristic algorithm is proposed to solve the NAPE problem, and its performance is evaluated via extensive simulations.

Second, the 3-D DBS placement problem is investigated as the communication efficiency is greatly affected by the positions of DBSs. Then, the DBS placement with IBFD communications (DSP-IBFD) problem for downlink communications is formulated, and two heuristic algorithms are proposed to solve the DSP-IBFD problem based on different DBS placement strategies. The performance of the proposed algorithms are demonstrated via extensive simulations.

Third, the potential benefits of jointly optimizing the radio resource assignment and 3-D DBS placement are explored, upon which the Drone-mounted Base Station Placement with IBFD communications (DBSP-IBFD) problem is formulated. Since the DBSP-IBFD problem is NP-hard, it is then decomposed into two sub-problems: the joint bandwidth, power allocation and UE association problem and the DBS placement problem. A 1/2(1-/2^{l}})-approximation algorithm is proposed to solve the DBSP-IBFD problem based on the solutions to the two sub-problems, where l is the number of simulation runs. Simulation results demonstrate that the throughput of the proposed approximation algorithm is superior to benchmark algorithms.

Fourth, the uplink communications is studied as the mobile users need to transmit and receive data to and from base stations. The Backhaul-aware Uplink communications in a full-duplex DBS-aided HetNet (BUD) problem is investigated with the objective to maximize the total throughput of the network while minimizing the number of deployed DBSs. Since the BUD problem is NP-hard, it is then decomposed into three sub-problems: the joint UE association, power and bandwidth assignment problem, the DBS placement problem and the problem of determining the number of DBSs to be deployed. The AA-BUD algorithm is proposed to solve the BUD problem with guaranteed performance based on the solutions to the three sub-problems, and its performance is demonstrated via extensive simulations.

The future work comprises two parts. First, a DBS can be used to provide both communications and computing services to users. Thus, how to minimize the average latency of all users in a DBS-aided mobile edge computing network requires further investigation. Second, the short flying time of a drone limits the deployment and the performance of DBSs. Free space optics (FSO) can be utilized as the backhaul link and the energizer to provision both communication and energy to a DBS. How to optimize the charging efficiency while maximizing the total throughput of the network requires further investigation.

Share

COinS