A Levy Flight-Based Offloading Path Decision Scheme in VANET

The calculation of vehicle tasks is restricted by its limited computing power and the shortage of RSU resources. Therefore, in this article, we propose an edge computing offloading with parked vehicle model to solve the problem of limited computing resources. We design the k-opt (Local search with k-change neighborhoods) decision algorithm based on Levy flight to find the optimal offloading path and reduce decision delay. We first use Levy Flight to simulate the natural decision-making process. Then we improve the k-opt algorithm based on the above results and dynamically select the k value to generate the offloading path. In the end, we get the shortest offloading path and decision time after multiple iterations. Our model combines cloud servers and parking vehicles to solve the problem of insufficient computing resources. The Levy-Kopt decision algorithm proposed by us improves the k-opt algorithm. According to Levy flight, k is dynamically selected to generate offloading path quickly and accurately. The simulation result shows that the algorithm can effectively reduce the error of the offloading path length of parked vehicles and the path decision delay.