基于DTN的车载云计算卸载算法  

DTN-based computation offloading algorithms for vehicular clouds

作　　者：李波[1] 黄鑫[1] 薛端[1] 侯严严 裴以建[1] LI Bo, HUANG Xin, XUE Duan, HOU Yan-yan, PEI Yi-jian(School of Information Science and Engineering, Yunnan University, Kunming 650500, Chin)

机构地区：[1]云南大学信息学院,云南昆明650500

出　　处：《云南大学学报（自然科学版）》2018年第2期215-221,共7页Journal of Yunnan University(Natural Sciences Edition)

基　　金：国家自然科学基金(61562092)

摘　　要：车载云计算卸载存在计算资源受限和网络通信中断等问题.提出了2种基于DTN的适用于时延容忍型应用的计算卸载算法:基于DTN的直接传递算法和基于DTN的泛洪算法.通过仿真实验对这2种算法和不采用DTN的直接传递算法的计算卸载性能进行了测试、分析和比较.结果表明基于DTN的计算卸载算法适用于网络拓扑结构不断变化的车载云计算环境,能够提高卸载任务的成功率,缩短任务完成时间.There are some problems such as limitation of computing resources and interruptions of network connections in vehicular cloud computing environments. In this paper,two DTN-based computational offloading algorithms（ i.e.,DTN-based direct transfer algorithm and DTN-based Flooding algorithm） were proposed for delay-tolerant applications in vehicular clouds.The performance of these two algorithms and the direct transfer algorithm without DTN was tested,analyzed and compared via simulation.The results show that the DTN algorithm is suitable for the on-board cloud computing environment,which can improve the success rate of unloading tasks and shorten the completion time.

关 键 词：车载云 车载自组织云 计算卸载 卸载算法 延迟容忍网络 

分 类 号：TP316.4[自动化与计算机技术—计算机软件与理论]