Analyzing Hit Probability of Spatial Correlated Caching for Heterogeneous Mobile Edge Computing

In recent years, edge servers are expected to provide not only computing resources but also caching resources for the purpose of enabling direct content delivery and offloading the data traffic from backhaul network. To overcome the restriction of limited caching resources in practice, diverse devices can be used to cache contents, which form a heterogeneous mobile edge caching network. In this paper, we study a spatial correlated caching model for heterogeneous edge caching network consisting of two kinds of caching devices: edge server and helper. A content placement strategy based on modified hard-core point process is proposed for caching helpers to provide efficient caching resources while reducing the redundancy of caching contents. The performance of the proposed strategy is analyzed using tool of stochastic geometry, in terms of hit probability, considering the effect of helper density and exclusion radius of edge servers. As validated by extensive numerical results, we are able to derive the upper bound and lower bound of hit probability and find the optimal exclusion radius of edge server to achieve a relatively high hit probability with fewer caching helpers.