| With the continuous development of networking technology,more and more applications and additional services have been deployed to the vehicle to provide a better driving experience.However,the limited computing power of the vehicle makes it difficult to complete the processing of large particle computing tasks in demanded delay.Using edge computing to offload the large particle computing tasks to the edge nodes such as the road side unit and base station to improve the network task processing capacity,is becoming an important research direction.However,during the offloading process,there are problems such as service interruption in highly dynamic scenarios and rapid decisionmaking demand under the influence of multiple factors.In view of the above problems,this paper proposes an offloading decision method for edge computing of IoV based on the improved SAC.Besides,this paper designs and implements a prototype system for mobile edge computing offloading in the scenario of IoV.The specific work is as follows.(1)Decision method of edge computing offloading of IoV based on improved SAC.Aiming at the problem of task offloading interruption caused by vehicular mobility,the IoV edge computing offloading model is constructed.Tasks are divided vertically,and multi-node collaborative offloading is used to reduce the impact of vehicular dynamics.Aiming at the problem of offloading decision,an offloading decision optimization method based on the improved SAC algorithm is designed to shorten the time window of vehicular decision and calculation and ensure the continuity of vehicle tasks.Compared with Q-learning and DQN algorithms,the improved SAC algorithm can reduce the offloading delay by approximately 60%and can generate offloading decision within 70ms.(2)Design and implementation of mobile edge computing offloading simulation prototype system in the scenario of IoV.Based on the above key technologies,the simulation system is designed in general and detail.Based on the Django framework,JavaScript,Python and other programming languages are used to develop the mobile edge computing offloading simulation prototype system,which realizes the functions of IoV edge environment construction,algorithm configuration,offloading setting,simulation result display,etc.,providing users with the ondemand configuration of simulation parameters and simulation statistics,and verified through the test. |
