面向5G车联网场景的移动任务动态卸载策略研究

由于车辆自身的高速移动性和资源有限性等特征，使得采用传统通信和计算手段的车联网场景无法满足用户日益增长的数据计算需求和体验质量需求。采用5G和边缘计算技术构建的新型车联网架构可以满足以上需求，但由于网络结构的变化，需设计适合新场景下的车辆任务通信和计算策略。针对5G车联网场景下的移动车辆任务动态卸载问题进行研究，提出了对应的动态任务分配策略和卸载调度低时延算法。车辆会根据提出的策略和算法将未完成的计算任务卸载到相应的 MEC 服务器或车辆上，并且计算结果将通过边缘服务器之间的联合通信或直接从被选择接受卸载任务的附近空闲车辆上直接返回给车主。仿真结果表明，所提出的策略和算法在优化卸载延迟方面具有良好的性能，并提高了用户体验质量。

Research on dynamic offloading strategy of mobile task for 5G vehicle networks

Due to the characteristics of high-speed mobility and limited resources of the vehicle itself, the vehicle networks scenarios using traditional communication and computing methods cannot meet the increasing data computing needs and the quality of experience(QoE) required by users. New vehicle networks scenarios built with 5G and mobile edge computing(MEC) technologies can meet the above requirements, but due to changes in network structure, it is necessary to design vehicle task communication and computing strategies suitable for the new scenarios. This paper studied the dynamic offloading of mobile vehicle tasks in the 5G vehicle networks scenario, and proposed a corresponding dynamic task allocation strategy and a low-latency algorithm for offloading scheduling. The vehicle would offload the unfinished computing tasks to the corresponding MEC server or vehicle according to the proposed strategy and algorithm. And the calculation result would be directly returned to the vehicle owner through the joint communication between the edge servers or directly from the nearby idle vehicle selected to accept the offloading task. Simulation results show that the proposed strategy and algorithm have good performance in optimizing offload latency and improve the quality of user experience.