面向高移动性车联网场景的预测卸载决策算法  

Predictive Offloading Decision Algorithm for High Mobility Vehicular Network Scenarios

作　　者：彭维平[1] 王戈 宋成[1] 阎俊豪 PENG Weiping;WANG Ge;SONG Cheng;YAN Junhao(College of Computer Science and Technology,Henan Polytechnic University,Jiaozuo 454000,China)

机构地区：[1]河南理工大学计算机科学与技术学院,焦作454000

出　　处：《北京邮电大学学报》2023年第2期43-49,共7页Journal of Beijing University of Posts and Telecommunications

基　　金：国家重点研发计划项目(2018YFC0604502);河南省青年骨干教师计划项目(2019GGJS061)。

摘　　要：针对车联网场景下高移动性车辆在不同边缘服务器之间频繁切换导致任务卸载失败率高的问题,提出了一种新的预测卸载决策算法。首先,构建了本地、边缘服务器和云服务器的计算模型,并基于计算任务量大小、最大容忍时延、服务器资源等约束条件,预测任务的卸载方式;其次,针对边缘服务器的卸载方式,利用长短期记忆网络构建车辆位置预测模型,生成可用于卸载的边缘服务器集合;最后,采用改进的蚁群算法在多个边缘服务器之间实现最优的任务分配。仿真实验结果表明,所提算法可提高任务的完成率和资源利用率。To solve the issue of the high failure rate of task offloading caused by frequent switching between different edge servers of highly mobile vehicles in the scenario of the Internet of vehicles,a novel predictive offloading decision algorithm is proposed.First,the local server edge server and cloud server computing models are constructed,and the offloading mode of tasks is predetermined based on the constraints of the size of computing tasks,maximum latency tolerance,and server resources.Then,a vehicle location prediction model is constructed by generating edge servers based on a long short-term memory network,which can be used for offloading.Finally,the improved ant colony algorithm is employed to optimize the offloading task allocation among multi-edge servers.Simulation results show that the proposed algorithm improves task completion rate and resource utilization rate.

关 键 词：车联网 边缘计算卸载 位置预测 长短期记忆网络 蚁群算法 

分 类 号：TN929.5[电子电信—通信与信息系统]