基于MEC-D2D的分布式协同缓存方案  

Distributed cooperative caching scheme based on MEC-D2D

作　　者：薛建彬[1] 张永刚 王泽森 张海军 XUE Jianbin;ZHANG Yonggang;WANG Zesen;ZHANG Haijun(School of Computer and Communication,Lanzhou University of Technology,Lanzhou 730050,China)

机构地区：[1]兰州理工大学计算机与通信学院,甘肃兰州730050

出　　处：《华中科技大学学报（自然科学版）》2021年第3期52-57,共6页Journal of Huazhong University of Science and Technology(Natural Science Edition)

基　　金：国家自然科学基金资助项目(61841107,61461026)。

摘　　要：针对移动边缘计算服务器存储能力有限的问题,利用终端通信技术数据共享优势提出了一种基于动态规划的移动边缘计算与终端通信(MEC-D2D)协同内容缓存放置策略.首先分析协同通信模型中的内容流行度、缓存空间有限性以及干扰限制等因素对用户内容获取平均时延的影响;然后通过建立内容分段获取模型,以降低用户内容获取平均时延为优化目标;最后将缓存问题视为完全背包问题,并采用动态规划算法设计最优缓存分配方案.仿真实验表明所提缓存策略可有效降低用户内容获取平均时延.In order to solve the problem of the limited storage capacity of server in mobile edge computing(MEC),a MEC-D2 D collaborative content cache placement strategy based on dynamic programming was proposed by taking advantage of the data-sharing advantages of device-to-device(D2 D) communication technology. Firstly,the influence of content popularity,cache space finiteness and interference limitation on the average delay of user content acquisition were analyzed. Secondly,by establishing a segmented content acquisition model, the optimization goal was to reduce the average delay of user content acquisition.Finally,the cache problem was considered as a complete knapsack problem,and the optimal cache allocation scheme was designed by the dynamic programming algorithm. Simulation experiment results show that the proposed caching strategy can reduce the average latency of user content acquisition effectively.

关 键 词：移动边缘计算 终端通信 协同缓存 动态规划 缓存策略 

分 类 号：TN929.5[电子电信—通信与信息系统]