Research On Task Offloading Algorithm Of Mobile Edge Computing Enabling Internet Of Vehicles

In recent years,the intelligent transportation industry has developed vigorously,and a number of in-vehicle applications have emerged to improve the driving experience and safety.However,most of these application services are computationally intensive or latency-sensitive,which brings a severe test to the storage and computing capabilities of mobile vehicles.On the other hand,the traditional offloading solution based on cloud services cannot meet the requirements of low-latency in-vehicle applications because the cloud server is too far away from the moving vehicle.By introducing mobile edge computing technology into the Internet of Vehicles and deploying mobile edge servers at the edge of the network,the capabilities of vehicles in terms of communication,storage and computing can be significantly improved.This thesis studies the optimization of task offloading algorithm based on mobile edge computing in the Internet of Vehicles environment.The main research contents are as follows:First,in the multi-vehicle single-server scenario,the impact of the real-time change of the distance between the vehicle and the server on the uplink data transmission rate and the impact of the task offloading decision on the system bandwidth allocation are modeled.In the delay constraint,it is considered that the unloading delay should be less than the time when the vehicle leaves the coverage of the roadside unit to ensure that the unloading of the vehicle task is not interrupted.The joint task offloading decision and computing resource allocation problem is formulated as a system utility maximization problem.Since this problem is NP-hard,it is converted into multiple local optimization problems to solve,and a numerical method based computing resource allocation is proposed in the local optimization problem.Algorithms are used to determine computing resource allocation,and a heuristic task offloading algorithm is proposed to obtain offloading decisions.The simulation results show that,compared with the random unloading algorithm,the unified and independent unloading algorithm and the adaptive genetic algorithm,the offloading algorithm proposed in this thesis can effectively reduce the task processing delay and energy consumption,and improve the total utility of the system.Secondly,in the multi-vehicle and multi-server scenario,for the scenario in which the moving vehicle can select the server for partial unloading through multiple roadside units,the queuing of tasks in the server is modeled,and the construction is based on the task processing delay and server load balancing.A user satisfaction utility function is proposed,and the joint selection decision,offloading rate and computing resource allocation problem are formulated as a system utility maximization problem under delay constraint,and then a partial task offloading algorithm based on PSO is proposed to solve the offloading decision problem,and use The Lagrangian method obtains the optimal unloading rate and computing resource allocation results.The simulation results show that the algorithm proposed in this thesis can effectively reduce the task processing delay and achieve the effect of multi-server load balancing.