Task Distribution in Vehicular Edge Computing
Loading...
Authors
Zaki, Amr M.
Date
2025-05-06
Type
thesis
Language
eng
Keyword
Autonomous Vehicles , Vehicular Edge Computing , Cooperative Perception , Task Offloading , Reinforcement Learning , Multi-Agent , Service Migration
Alternative Title
Abstract
Advancements in autonomous vehicles (AVs) and Cooperative Intelligent Transportation Systems (C-ITS) have driven the emergence of numerous traffic management services aimed at improving road safety and enhancing traffic efficiency In contrast to traditional cloud computing, Vehicular Edge Computing (VEC) enables real-time data processing by offloading computationally intensive services to the network’s edge. This approach significantly reduces latency and improves responsiveness in real-time vehicular applications. However, the deployment and operation of VEC systems are fraught with challenges due to the stringent task requirements, the complexities of urban environments, and the resource constraints of both users and edge nodes. These limitations complicate the effective execution of delay-sensitive tasks. Additionally, the dynamic nature of vehicle mobility, along with the uneven spatial distribution of vehicles and the heterogeneity of edge nodes, presents significant challenges in maintaining continuous service and task execution across the network. In this work, we propose the Joint Offloading and Migration for Cooperative Transportation Services (JOMC) framework, which addresses these challenges in VEC environments To handle the complexity of the problem and the real-time requirements, JOMC integrates Deep Reinforcement Learning (DRL) to efficiently handle decisionmaking in Vehicular Edge Computing (VEC). The framework adopts a decentralized solution approach to respect user privacy, while employing auction mechanisms to align with market dynamics. JOMC tackles the joint problem of task offloading and migration for C-ITS allowing for efficient management of the complexities inherent in modern VEC systems. JOMC addresses the core complexities of offloading and migration in C-ITS, including stringent quality requirements, communication uncertainties, limited resources, and maintaining service continuity amid uneven load distributions. It enables adaptive and reliable decision-making in dynamic vehicular environments JOMC is designed to navigate the intricate communication challenges of modern vehicular networks, ensuring service continuity, incentivizing fair participation, and enhancing system reliability in C-ITS.
Description
Citation
Publisher
License
Queen's University's Thesis/Dissertation Non-Exclusive License for Deposit to QSpace and Library and Archives Canada
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
Attribution 4.0 International
Intellectual Property Guidelines at Queen's University
Copying and Preserving Your Thesis
This publication is made available by the authority of the copyright owner solely for the purpose of private study and research and may not be copied or reproduced except as permitted by the copyright laws without written authority from the copyright owner.
Attribution 4.0 International