Advanced optimization and machine learning techniques for efficient wireless communication networks

Makled, Esraa Aziz Mokhtar Muhammad (2024) Advanced optimization and machine learning techniques for efficient wireless communication networks. Doctoral (PhD) thesis, Memorial University of Newfoundland.

[English] PDF - Accepted Version Available under License - The author retains copyright ownership and moral rights in this thesis. Neither the thesis nor substantial extracts from it may be printed or otherwise reproduced without the author's permission. Download (8MB)

Abstract

Wireless communications have become indispensable in modern society, driven by the proliferation of mobile devices, Internet-of-Things applications, and data-intensive services. As the world moves towards sixth-generation (6G) wireless networks, optimizing limited bandwidth and power resources is crucial to meet growing data demands. Additionally, harnessing the complete capabilities of all available wireless media, including space, air, and water, is deemed essential to ensure the seamless communications promised by the 6G wireless and beyond networks. This thesis focuses on overcoming data rate and security issues in two communication media, namely underwater and terrestrial. Acoustic is the most prominent wireless communication technology in underwater communication. In underwater acoustic networks, full-duplex (FD) and non-orthogonal multiple access (NOMA) techniques are explored to address challenges unique to the underwater environment. The goal is to enhance the data rates, reliability, and security of underwater communication systems. Power optimization is studied to maximize the sum rate or secrecy sum rate against cyber attacks. The proposed algorithms provide enhanced sum rates and security when FD and NOMA are applied with effective interference cancellation. In terrestrial communications, accurate cellular signal identification is essential for resource optimization and wireless network security. Hence, in this thesis, I provide multiple methodologies to enhance the ability to identify over-the-air signals from various technologies: global systems for mobile communications, universal mobile telecommunication systems, and long-term evolution in realtime. Morphological analysis and machine learning algorithms are proposed to achieve accurate signal detection and identification to enhance the security and efficiency of wireless communication systems. By tackling both media, this thesis aims to provide unique solutions to improve the security and data rates for future networks.

Actions (login required)

View Item