Study Of MEC-based Task Offloading Strategy In C-V2X Internet Of Vehicle

Intelligent traffic system is rapidly promoting the development and construction of smart cities in China.It leads continuous iterative update of the road traffic safety system,and makes people’s driving environment more safe,environmental protection,and smoother.It hence provides higher quality services to users.The intelligent transportation system uses intelligent networked vehicles as nodes to form a C-V2X vehicle networking communication system.In the perception layer,lot of sensors,such as speed sensor,acceleration sensor,GPS,tire pressure sensor etc.,are used as the sensing nodes.The communication protocols between electronic control units(ECUs)are established via a controller area network(CAN),and the collected data from the sensors are fused.Thus,the intercommunication,interconnection,information transmission and sharing between all devices in the C-V2X vehicle networking communication system are realized.At present,the different high-speed mobility of intelligent networked vehicle nodes will produce big data flow computational tasks with different attributes,which may cause problems,such as delay jitter during information transmission and offloading,large computational energy consumption and system overhead and so on.To solve these problems,developing mobile edge computing technology applied to C-V2X internet of vehicles can reduce delay jitter,computational energy consumption and system overhead in the computational tasks offloading process,in order to improve the scheduling efficiency of the C-V2X vehicle network.This has become a current social research hots pot.The main research work of this thesis as follows.This thesis integrates C-V2X Internet of Vehicles and mobile edge computing technologies in chapter 3,and proposes a task offloading optimization strategy based on simulated annealing algorithm for C-V2X internet of vehicles to reduce time delay,computational energy consumption and system overhead.Firstly,a proportional task offloading model is studied in this thesis,aiming to minimize the total system overhead under collaborative offloading.Depending on the task processing priority,the high priority task is divided into two parts according to a different proportion.The coordinated offloading calculation is performed by the local,MEC servers.Secondly,a task offloading strategy based on simulated annealing algorithm is developed according to the offloading computation method of the tasks.The offloading scale factor is analyzed and optimized with the aid of globally searching for the optimal offloading scale factor.Finally,the minimizing of the system overhead is transformed into the allocation convex optimization of power and computational resource in the process of task offloading and updating.Lagrange multiplier method is used to obtain the optimal solution under the constraints of multiple variables.Compared with the adaptive genetic algorithm,the proposed task offloading strategy and resource allocation method based on the simulated annealing algorithm effectively reduce the time delay,power consumption and system overhead.As the calculation task volume increases,the time delay,power consumption and system overhead are maximally decreased by 6.35%,92.27% and 91.7%.As the CPU cycles of the calculation task increases,the delay,power consumption and system overhead are maximally reduced by 19.61%,94.39% and 89.88% respectively.Compared with the Q-learning offloading strategy,the proposed task offloading and resource allocation method reduces the system overhead by 60.96% and accelerates the convergence.This thesis aims to make full use of the excessive intelligent networked vehicle node resources on the idle road side in chapter 4,in order and reduce the delay and overhead of the C-V2X IOV communication system.We analyze and establish a C-V2X IOV task offloading optimization strategy based on the artificial fish swarm algorithms and harmony search algorithms.Firstly,the computation task generated by the intelligent vehicle node is divided into three parts according to a certain proportion,and the coordinated offloading calculation is performed by the local,edge and target vehicle nodes.Secondly,the dependences of V2I offloading and V2V offloading order on the delay and energy consumption of the system are modeled respectively in C-V2X IOV communication system.Finally,aiming to minimize time delay and overhead of the C-V2X IOV communication system,an optimization scheme based on artificial fish swarm algorithm and harmony search algorithms is established in order to optimize multi-hop V2V transport routing and task offloading scale factor in the two different modes described above.Simulation results show that when multi-hop transmission is carried out between vehicle nodes,as the amount of tasks continues to increase,the offloading delay of communication between vehicle nodes is reduced by 4.06% using the proposed artificial fish swarm algorithm and harmony search algorithms,compared with the commonly used distributed algorithms.Compared with the hybrid multi-hop edge computing offloading algorithm,with the continuous increase in the amount of tasks,the optimized task offloading strategy based on artificial fish swarm algorithm and harmony algorithm proposed in this thesis reduces the total system overhead by 6.98%.