Login | Register

Unmanned Aerial Vehicles for 5G and Beyond: Optimization and Deep Learning


Unmanned Aerial Vehicles for 5G and Beyond: Optimization and Deep Learning

Shoukry, Moataz (2020) Unmanned Aerial Vehicles for 5G and Beyond: Optimization and Deep Learning. PhD thesis, Concordia University.

[thumbnail of Shoukry_PhD_S2021.pdf]
Text (application/pdf)
Shoukry_PhD_S2021.pdf - Accepted Version


Aerial platforms and, more precisely, Unmanned Aerial Vehicles (UAVs) or drones augmented with ubiquitous computing, processing, and wireless communication technologies are expected to play an important role in next-generation cellular networks. The flexibility and controllable mobility of UAVs render them suitable to be part of access. Nonetheless, combined terrestrial and UAV communication improving network coverage and Quality of Service by leveraging line-of-sight communication as well as minimizing the delay and age-of-information for UAV-to-ground communication. Despite its numerous advantages, the deployment of UAVs faces different challenges with respect to wireless networks, ranging from radio resource management to UAVs’ trajectory under energy limitation constraint and minimal knowledge of the environment. To this end, this dissertation aims to address the challenges in the efficient deployment of UAVs in future networks under various performance metrics. The key goal of this dissertation is to provide the analytical foundations for deployment, learning, in-depth analysis, and optimization of UAV-assisted wireless communication networks. Towards achieving this goal, this dissertation makes significant contributions to several areas of UAV-assisted wireless communication networks within the contexts of static environments as well as high mobility environments. For the deployment of UAVs in static environments such as Internet of Things (IoT) wireless networks, various tools such as optimization theory and machine learning frameworks are employed to enable UAV trajectory design under different scenarios and performance metrics. Results demonstrate the effectiveness of the proposed designs. In particular, UAVs adapt their mobility and altitude to enable reliable and energy-efficient communication, to maximize service for IoT applications, and to maintain the freshness of information. For the deployment of UAVs in high mobility environments such as vehicular networks, unique design challenges are considered and carefully handled to guarantee the effective performance of the UAV. Particularly, the high mobility of the vehicles leads to distinct network conditions and changes the network topology. The challenge here is that designing an efficient deployment of UAVs while considering the complex and dynamic network conditions is not a trivial task. This challenge was addressed through comprehensive studies that led to effective, robust, and high-performance solutions. Different performance metrics such as coverage, age of information, throughput, and Quality of Service were evaluated and compared with other approaches. Results shed light on the trade-offs in the vehicular network such as throughput-latency when exploiting UAV mobility for service. The findings in this dissertation highlight key guidelines for the effective design of UAV-assisted wireless communication networks. More insights on the efficient deployment of UAVs in static and high mobility environments are provided in order to assist and enhance communication in future networks while considering the unique features of UAVs such as their flight time, mobility, energy budget, and altitude.

Divisions:Concordia University > Gina Cody School of Engineering and Computer Science > Concordia Institute for Information Systems Engineering
Item Type:Thesis (PhD)
Authors:Shoukry, Moataz
Institution:Concordia University
Degree Name:Ph. D.
Program:Information and Systems Engineering
Date:15 December 2020
Thesis Supervisor(s):Assi, Chadi and Ghrayeb, Ali and Sharafeddine, Sanaa
ID Code:988009
Deposited By: Moataz Shoukry
Deposited On:29 Jun 2021 20:58
Last Modified:29 Jun 2021 20:58
All items in Spectrum are protected by copyright, with all rights reserved. The use of items is governed by Spectrum's terms of access.

Repository Staff Only: item control page

Downloads per month over past year

Research related to the current document (at the CORE website)
- Research related to the current document (at the CORE website)
Back to top Back to top