Research On Resource Allocation In Internet Of Vehicles Based On Mobile Edge Computing

With the emergence of a large number of computational resource-intensive applications and various content delivery services,there is an explosion of data growth in the internet of vehicle(Io V).In order to solve the shortage of local computation resources,mobile edge computing(MEC),which could offload the task of vehicles to the MEC server,is used in the vehicle network to reduce the energy consumption of the whole network and guarantee the execution delay.Moreover,in order to reduce the end to end delay,caching technology is considered as a potential solution to reduce the content transmission dela.This thesis focuses on the task offloading in the Io V and the resource management scheme in the content caching.In this thesis,the task offloading problem in Io V is studied.Firstly,according to the attribute of the task,the tasks are divided into offloadable and unoffloadable subtasks.A dynamic partial offloading model is proposed for the offloadable subtask.We aim to minimize the network cost of Io V,which includes energy consumption and dropped packets.Secondly,a computation resource allocation method is designed for the vehicle to optimize the computation resource allocation of the MEC server,which allocates the computation resources of MEC server to subtasks.By this way,the processing time of each subtask in the MEC server is the same.In addition,a Lyapnuov-based dynamic offloading decision algorithm(LDOD)is proposed to minimize the overall network cost,by optimizing the offloading decision of offloadable subtasks,energy consumption and the corresponding packet drop strategy while ensuring the queue stability.Finally,based on the simulation results,the effectiveness of the proposed algorithm is verified in the aspects of average energy consumption and average packet drop rate.Furthermore,in order to minimize the content fetching delay for vehicles,the content cache optimization scheme based on V2 V collaboration in Io V is studied in this thesis.Firstly,the optimization problem is decomposed into two subproblems,namely,the vehicle matching optimization and content caching optimization.Secondly,a delay-aware vehicle matching algorithm(DVM)is proposed to optimize the vehicle matching problem.The mode,e.g.the V2 I mode or the V2 V mode,which provides lower delay of fetching contents will be selected.Then,based on the results of vehicle matching,the content caching scheme is optimized in two network scenarios.In this thesis,we propose a dynamic programming based content caching algorithm(DPCC)to optimize caching decisions in the scenario which does not consider handover vehicles during the content fetching process.Moreover,in the scenario which considers handover vehicles during the content fetching process,the content is divided into chunks and the caching decision optimized by the dynamic programming method in combination with the DPCC algorithm according to different situations.Finally,in the simulation results,the actual scene is simulated and the effectiveness of the proposed algorithm is verified.