无人机接替场景中相似度感知的边缘计算缓存算法

Edge Computing Caching Algorithm for Similarity Aware in UAV Succession Scenario

无人机边缘计算将无人机的移动性和边缘计算的源端处理优势相结合,可以在基础设施薄弱区域中提供计算、传输和存储服务,近两年得到了广泛关注。然而,无人机平台的计算、能量和存储容量有限,面对不断增长的内容服务请求,无人机边缘节点只能缓存部分流行度较高的内容。此外,为了保证服务的连续性,无人机之间往往需要接替提供缓存服务。现有的边缘缓存算法大多根据内容的请求历史来预测流行度,使得接替无人机上对新请求的内容服务启动缓慢,缓存命中率较低,限制了边缘计算缓存的性能。在此背景下,充分考虑无人机的移动性和内容请求的动态性,设计了一种针对接替无人机的动态缓存架构,并综合考虑接替无人机上缓存内容的流行度、相似性以及接替距离,提出了一种相似度感知缓存算法(Similarity-Aware Caching,SAC)。仿真结果表明,该缓存算法可以有效提升无人机接替场景中的缓存命中率和接替成功率。

UAV(Unmanned Aerial Vehicles)edge computing,by combing the mobility of UAVs with the advantages of edge computing in source-end processing,can provide computing,transmission,and storage services in areas with weak infrastructure,and received widespread attention in the past two years.However,the UAV platform has limited computing,energy and storage capacity.In the face of increasing content service requests,UAV edge nodes can only cache some of the more popular content.In addition,in order to ensure the continuity of services,UAVs often need to replace each other to provide caching service.Existing edge caching algorithms mostly predict the popularity in accordance with the request history of the content,as a result,the content service of the new request on the successor UAV starts slowly,the cache hit rate is low,which limits the performance of the edge computing cache.In this context,by fully considering the mobility of the UAV and the dynamics of the content request,a dynamic cache architecture for the successor UAV is designed.At the same time,by comprehensively considering the popularity,similarity and replacement distance of the cached content on the successor UAV,a SAC(similarity-aware caching)algorithm is proposed.The simulation results show that the caching algorithm can effectively improve the cache hit rate and success rate of UAV replacement scenarios.