Content Caching Policy with Edge Caching User Classification in Fog Radio Access Networks

With the rapid growth of mobile devices and applications, fog radio access network (F-RAN) has been proposed as a promising paradigm for the 5th generation of mobile networks. Meanwhile, F-RAN can be further enhanced by edge caching underpinned by device-to-device (D2D) communications, where the download delay can be reduced by edge caching users (EUs) providing content though D2D links to content requesting users (RUs). In this paper, to mitigate the impact of limited storage at each EU, we propose an EU classification based content caching policy for F-RAN, where the EUs are divided into two groups, each group caching a different content set. To maximize the cache hit probability at EUs, we present a caching policy selection algorithm that allows the fog radio access point (F-AP) to choose between the proposed EU classification based content caching policy and the conventional probability based caching policy. By modeling the content request queue at each EU as an independent M/D/1 queue model at EUs, we analyze the average download delay under the proposed caching policy selection algorithm. The simulation results show that the proposed algorithm can significantly increase the cache hit probability of EUs and can also reduce the average download delay for RUs.