Resource Allocation in Internet of Vehicles Mobile Edge Computing Scenarios

With the rapid development of 5G/6G communications and intelligent transportation systems, the Internet of Vehicles (IoV) requires real-time processing of massive data to meet low-latency and high-reliability demands. Traditional cloud computing struggles to adapt to IoV scenarios due to high transmission latency and dynamic network conditions, while Mobile Edge Computing (MEC), by deploying computing resources at the network edge (e.g., roadside units, onboard terminals), emerges as a key solution. Dynamic resource allocation, which involves multi-objective optimization, collaborative scheduling, and security assurance, is complicated by issues including task priority disparities, resource heterogeneity, and vehicle mobility. To address these issues, this paper proposes a dynamic resource allocation framework based on deep reinforcement learning (DRL), jointly optimizing computation offloading, communication resource scheduling, and energy consumption control to balance low latency and high energy efficiency. A distributed collaboration mechanism and lightweight security protocols are designed to incentivize resource sharing and mitigate malicious attacks. This study demonstrates that the proposed strategy reduces task processing latency by 20-30% and edge node energy consumption by over 15% in peak congestion scenarios, while achieving a critical task completion rate exceeding 95%. These results provide theoretical and practical insights for edge computing resource management in intelligent transportation systems.