车载边缘计算中基于轨迹信息的服务迁移策略

Service migration strategy based on trajectory information invehicular edge computing

面对复杂的道路交通环境和网络条件,如何在多个边缘服务器之间选择合适的目标进行迁移以及避免频繁迁移导致的高额开销是一个难题。为此,提出一种依赖轨迹信息的服务迁移策略,该策略根据车辆的轨迹信息和可迁移的边缘服务器的位置信息来优化迁移策略,从而选择更合适的目标边缘服务器。在此基础上,设计了一种基于竞争深度Q网络(dueling deep Q-network, Dueling DQN)的算法进行快速决策。仿真实验证明了该策略的有效性,与其他策略比较的结果表明,该策略可以权衡时延和迁移开销,取得最小的系统总开销。

In the face of complex road traffic environment and network conditions,it is a challenge to choose the appropriate target for migration among multiple edge servers and to avoid high cost caused by frequent migration.Therefore,a service migration strategy relying on trajectory information was proposed,which can optimize the migration strategy based on the trajectory information of vehicles and the location information of migratable edge servers,so as to select more suitable target edge servers.On this basis,an algorithm based on dueling deep Q-network(Dueling DQN)was designed for fast decision making.Simulation experiments prove the effectiveness of this strategy.The comparison results with other strategies show that the proposed strategy can trade-off the delay and migration cost to obtain the minimum total system cost.