Research On Collaborative Offloading Strategy Between Multi-access Edge Computing And Cloud Computing Under Vehicular Internet Environment

The growth of Vehicle to Everything(V2X)has greatly improved user experience and traffic efficiency.By combining Multi-access Edge Computing(MEC)and the V2 X,the vehicles can offload some applications to MEC for computing so as to compensate well for the shortage of computation resources.However,due to the limited computation resources,the MEC couldn't meet the demand of the high-density vehicle scenario.At the same time,current researches on computation offloading in the V2 X mainly focus on the minimization of system delay and energy consumption,without considering the different importance of driving safety and passenger experience for different applications.In view of the above shortages,this paper respectively researches the collaborative computation offloading of MEC and CC in the quasi-static scenario and the mobile scenario,and the main contents are as follows:(1)A collaborative offloading strategy of MEC and CC based on the offloading priority in quasi-static scenario is proposed.In the quasi-static scenario with low speed in the city,each application has a priority factor to represent its offloading priority and the factor reflects the influence of the application on driving safety and passenger experience while MEC and CC are collaborated for computation offloading.And an Improved Discrete Artificial Bee Colony Algorithm(IDABC)is proposed to solve the optimization problem,then the simulation results show that the offloading strategy is effective to prioritize the applications with higher offloading priority to improve the driving safety and passenger influence.(2)A collaborative offloading strategy in the mobile scenario is proposed,where the mobility management is realized and all applications can be arbitrarily segmented.For the collaborative computation offloading of MEC and CC in the urban road with high speed,each application has a certain priority factor and is classified into safety-ralated or nonsafety-related application in order to improve the driving safety of the whole system,and each base station should reserve certain spectrum and computation resources for adjacent base station to meet the mobility management of crossing-base-station applications while each application can be arbitrarily segmented to be processed in parallel and make full use of the resources of MEC and CC.An Modefied Whale Optimization Algorithm based on the double greedy criteria is proposed to solve the optimization problem,and the simulation results show that the proposed collaborative offloading strategy has a good effect on offloading,could improve the driving safety especially for the crossing-base-station applications and max the total number of applications that can be successfully offloaded.