MULTI-OBJECTIVE OPTIMAL ROUTING SCHEMES FOR HIGH MOBILITY VEHICULAR NETWORKS: A PATH TO EFFICIENCY

Abstract

Technological advancements in wireless communication networks have enabled futuristic applications that support massive device access and pervasive communications. Moreover, vehicular networks in Intelligent Transportation Systems (ITS) require efficient communication and routing schemes to accommodate Electric and Flying Vehicles (EnFVs). A centralized approach is often flawed due to the high mobility and dynamic nature of device movement. Therefore, efficient and novel solutions are required to provide connectivity to EnFVs without any centrally connected unit. Our main focus in this study is to enable a faster, better, and improved communication platform for EnFVs, support a wide range of applications.

This dissertation provides an in-depth examination of EnFVs within ITS, emphasizing the necessity for a unified approach to tackle the unique challenges they pose. Moreover, this study thoroughly analyzes the role of Artificial Intelligence (AI), specifically Genetic Algorithms (GAs), in optimizing communication decision-making for high-mobility vehicles. This comprehensive work extensively reviews existing solutions and the background of GAs, highlighting the relevance of multi-objective optimization algorithms.

Communication and routing issues in EnFVs are examined from various angles. A novel multi-objective routing scheme addresses the diverse constraints and goals of EnFV networks, aiming to improve packet routing performance and efficiency. Our novel scheme prioritizes energy and transmission rate for routing decisions while focusing on vehicle connectivity time. The Genetic Algorithms employed identify the optimal solution for multi-objective routing problems. Significant findings include an optimized routing scheme that outperforms current solutions, achieving over 90% packet delivery ratio, extended connectivity time, reduced average hop distance, and efficient energy use. The research also explores the potential of Genetic Algorithms in solving complex optimization problems in EnFVs, demonstrating their effectiveness in dynamic routing scenarios.

Further enhancements to the solution improve route discovery methods, making the process lighter and more suitable for high-mobility UAV networks. The dissertation concludes with recommendations for future research to further improve the efficiency and effectiveness of routing algorithms in EnFV networks, aim for seamless integration into modern transportation systems and advance the field of EnFVs.