Research On Joint Optimization Of Task Offloadingand Resource Allocation For Software–Defined Vehicular Networking

Vehicular ad hoc networks(VANETs)have drawn much attention from academia and industry as key enablers of intelligent transportation systems(ITSs).The incipient vehicular applications and the explosive growth of data generally lead to an increase in demand for communication,computation,and storage resources,as well as stringent performance requirements on latency and wireless network capacity.To address these challenges,vehicular edge computing(VEC)is envisioned as a promising paradigm that extends the computation capability to the edge of the vehicular network,allowing nodes such as roadside units(RSUs)and on-board units(OBUs)in vehicles to perform services with location awareness and low delay requirements.Furthermore,it alleviates the bandwidth congestion caused by the large number of data requests on the network.On-board edge task offloading is the key technology to realize the vision of on-board edge tasks processing.However,the strong mobility of vehicle nodes and frequent topology changes bring new challenges to on-board task offloading and task processing.The main work of the thesis is as follows:Firstly,a task offloading decision model is designed using game theory to optimize the task processing capability in VEC environment.In this scenario,the mobile node connects to a set of edge servers through testing and decides which edge server node should be selected for offloading in order to optimize the sequence of computing tasks.In view of this,the task offloading decision process is modeled as a game model.Secondly,VEC is incorporated with device-to-device(D2D)task offloading in VANETs to benefit from proximity gain and to optimize computation offloading.A software-defined network(SDN)inside the vehicular edge network(SD-VEN)is developed to rectify most problems in the vehicular network.The SD-VEN controller handles the D2 D pairing of vehicles and coordinates with the edge cloud for flexible allocation of computing resources among vehicles and mobile users.The aim is to optimize the number of devices the vehicular network can support under the communication and computational limitations.We formulate the task offloading and associated resource allocation problem as a game with mixed strategies.Thirdly,the concept of resource pooling is employed to enhance system capacity by pooling the edge computing resources available to the vehicle and its passenger devices together.The evaluation discusses how the GT approach performs compared to random offloading,equal offloading,VEC offloading,and vehicular nearby peer computing offloading.The effects of task offloading on participant utility were analyzed qualitatively and quantitatively.Experimental results shows that equilibrium is reached efficiently if the number of tasks offloaded remains constant.The baseline approach is outperformed if we compare the overall task round-trip times in the SD-VEN system.In addition,in the simulations of dynamic task offloading,the experimental results show that the game model still converges to equilibrium,and once the game model reaches a near stable state,only a few policy changes are needed.In summary,the paper models the task offloading decision process as a game model,It is the first time to introduce D2 D into the vehicular edge computing network,which can enhance the capacity of the network system and improve the utilization of edge resources without additional cost.Then,the SDN network is introduced into this scenario,and the optimal task offloading and global resource management of the global network can be realized through the awareness of the network interaction state by the controller.Both the theoretical and experimental results show that the proposed game-based vehicle task offloading strategy can play an important role in optimizing the decision of task offloading.Numerical simulation results show that the average processing time of the proposed model is 25% higher than that of the existing schemes.