Optimizing Offloading in MEC-Enabled Vehicular Networks Using Adaptive PSO and V2V Communication

With the advancements in mobile communication technology, intelligent transportation systems have evolved, leading to higher demands for vehicular communications. However, computation-intensive and power-consuming applications in these systems result in significant energy consumption and computation costs, posing challenges for onboard systems. To ensure a better Quality of Experience (QoE), traffic offloading and scheduling in vehicular networks become essential. This paper introduces a joint offloading strategy for Mobile Edge Computing (MEC) enabled vehicular networks using quantum particle swarm optimization. The objective is to minimize delay cost and energy consumption by efficiently assigning tasks to available service nodes, including service vehicles and nearby Road Side Units (RSUs). The offloading process involves Vehicle to Vehicle (V2V) communication, where a vehicle selection algorithm called Fuzzified Relay Learning model is utilized to determine the optimal offloading decision sequence. To optimize task delay and energy consumption, the proposed approach utilizes an adaptive particle swarm optimization algorithm. This involves introducing a crossover operator to exchange information among individuals to maintain population diversity, and defining a crossover probability to enhance search capability and convergence speed. Additionally, an adaptive shrinkage expansion factor is designed to improve local search accuracy during later iterations. Simulation results demonstrate the effectiveness of the proposed joint offloading strategy in reducing system overhead and task completion delay across various system parameters. By implementing this approach, vehicular networks can achieve better efficiency and performance, meeting the demanding requirements of intelligent transportation systems.