Date of Award

8-2025

Document Type

Dissertation

Degree Name

Doctor of Philosophy (PhD)

Department

Electrical Engineering

First Advisor

Naima Kaabouch

Abstract

The increasing demand for high-throughput, low-latency, and reliable wireless communication has led to significant advancements of cellular wireless technologies since the early ‘80s. However, in dynamic and interference-prone environments—particularly those involving Unmanned Aerial Vehicles (UAVs)—signal degradation due to pathloss, multipath propagation, co-channel interference, and weather-induced attenuation continues to undermine performance. These factors contribute to elevated bit error rates (BER), inefficient spectrum usage, and unreliable communication links, thereby limiting the practical utility of wireless systems in aerial applications. Traditional mitigation techniques, while effective under certain conditions, often fail to adapt to rapidly changing wireless environments. This shortcoming underscores the need for intelligent, context-aware optimization strategies capable of dynamic adaptation.This dissertation aims to advance mobile wireless communication, with an emphasis on UAV-to-ground control station (GCS) links, through the development and application of Artificial Intelligence (AI)-based optimization frameworks. These frameworks are designed to counteract the sub-optimization mechanisms arising from channel impairments and operational variability. The research investigates how environmental and system-level factors—such as UAV altitude, velocity, frequency selection, and modulation order—interact to degrade signal quality and increase BER. A central problem addressed in this work is the lack of a comprehensive, real-time optimization strategy that can simultaneously manage multiple degradation factors while dynamically reconfiguring key communication parameters. The research problem is approached through a multi-phase methodology. In Phase 1, detailed models are developed to characterize the individual and combined effects of pathloss, multipath propagation, interference, and weather-induced attenuation on signal quality and BER. These models are implemented and validated using MATLAB® simulations. In Phase 2, seven state-of-the-art metaheuristic algorithms—Particle Swarm Optimization (PSO), Ant Colony Optimization (ACO), Gray Wolf Optimization (GWO), Genetic Algorithm (GA), Chicken Swarm Optimization (CSO), Whale Optimization Algorithm (WOA), and Elephant Herd Optimization (EHO)—are employed to optimize critical communication parameters, including carrier frequency, transmit power, modulation scheme, UAV speed, and altitude. The goal is to minimize BER while satisfying practical constraints. Phase 3 evaluates the performance of these algorithms across two metrics: convergence time and processing time, enabling a comparative analysis of their effectiveness and efficiency in real-time operational contexts. This dissertation makes several key contributions. First, it presents a quantitative analysis of how pathloss affects BER and signal quality in UAV communications, incorporating a variety of operational parameters such as distance, frequency, and altitude. Second, it delivers a comprehensive examination of multipath propagation, assessing its effects across multiple modulation schemes, channel conditions, and Doppler scenarios. Third, it characterizes how different interference levels and frequency overlaps influence bandwidth availability and system robustness. Fourth, it models the impact of atmospheric weather conditions—such as fog, rain, and cloud density—on signal attenuation and link reliability, offering insights into how these conditions affect UAV-GCS communication under realistic constraints. Finally, the dissertation proposes a unified AI-based multi-parameter optimization framework, capable of dynamically tuning key communication parameters to maintain low BER and high reliability in rapidly varying aerial environments. The framework's performance is rigorously evaluated using the metaheuristic algorithms, providing practical guidelines for selecting the most effective optimization strategy based on application-specific constraints. By integrating signal degradation modeling with intelligent optimization techniques, this work addresses a critical gap between theoretical models and their operational deployment. Unlike existing approaches that either oversimplify the communication environment or focus on isolated impairments, the proposed framework accounts for the complex interplay between various degradation factors. It enables real-time, environment-aware reconfiguration of communication systems, thus enhancing the spectral efficiency, resilience, and quality of service of UAV-supported wireless networks. The findings of this research have strong implications for the design of next-generation mobile communication systems, particularly in mission-critical applications such as disaster response, aerial surveillance, and autonomous logistics. In conclusion, this dissertation presents a robust, scalable, and adaptive solution to one of the most pressing challenges in UAV wireless communication—real-time BER minimization under diverse and unpredictable channel conditions. The AI-based optimization framework developed herein offers a viable path forward for the deployment of reliable and efficient UAV communication systems in real-world environments, paving the way for more intelligent and autonomous aerial network architecture.

Download

Share

COinS