| 摘要: |
| 作为5G中的一种重要模型，雾无线接入网络（Fog Radio Access Network，F-RAN）通过设备到设备通信和无线中继等技术获得了显著的性能增益，而边缘设备中合适的缓存则可以让内容缓存用户（Caching Users，CUs）向内容请求用户（Requesting Users,RUs）直接发送缓存内容，有效减小前传链路的负担和下载延迟。考虑一个F-RAN模型下用户发出请求并获得交付的场景，将每个CU的内容请求队列建模为独立的M/D/1模型，分析导出CUs缓存命中率和平均下载延迟关于内容缓存与交付方案的表达式，证明CUs缓存命中率与内容统计分布之间的联系有助于实现前者的近似最优解。针对在一段时间内的期望视角下建立的优化问题，提出了基于统计分布的算法并注意了执行时的交付控制。仿真结果表明，相较于现有缓存策略，优化内容整体统计分布的方案能够最大化CUs缓存命中率，同时减小平均下载延迟。 |
| 关键词: 5G 雾无线接入网络（F-RAN） 雾计算 终端直通（D2D） 内容缓存 |
| DOI：10.20079/j.issn.1001-893x.220419009 |
|
| 基金项目:国家重点研发计划（2020YFB1806606）；国家自然科学基金资助项目（61801460） |
|
| Content Caching and Delivery Policy with Frequency Distribution in Fog Radio Access Networks |
| CHEN Haoyu,HU Honglin |
| ((1.Shanghai Advanced Research Institute,Chinese Academy of Sciences,Shanghai 201210,China;2.School of Electronic,Electrical and Communication Engineering,University of Chinese Academy of Sciences,Beijing 100049,China)) |
| Abstract: |
| As an important model in 5G,the Fog Radio Access Network(F-RAN) achieves significant performance gains through technique among device-to-device(D2D) communications and wireless relays.Especially,caching appropriately in the edge devices allows content caching users(CUs) to send content to content requesting users(RUs) directly,which effectively reduces the burden of the fronthaul and download delay.The authors consider a scenario where users send content requests and get delivered under the F-RAN model.By modeling the content request queue at each CU as an independent M/D/1 queue model,the authors analyze and derive the cache hit probability of CUs and average download delay expressions under content caching and delivery policies.It is proved that the relationship between the cache hit probability of CUs and content frequency distribution can help realizing the approximate optimal solution of the former.On this basis,the authors establish an optimization problem under the expectation perspective over some time and propose a frequency distribution(FD)-based algorithm with delivery control when implementing to solve it.The simulation results show that compared with the existing caching policies,the policy optimizing the FD of all content can maximize the cache hit probability of CUs and reduce the average download delay. |
| Key words: 5G fog radio access network(F-RAN) fog computing device-to-device(D2D) content caching |
