Cooperative caching strategy based on cluster and social interest in mobile edge network

Abstract

Mobile edge caching can deliver contents directly without the backhaul link, which can effectively solve the problem of spectrum scarcity caused by huge mobile data traffic. In this paper, different from the existing user-centric clustering algorithms, a distributed caching algorithm is proposed based on content providers (CPs), which can form a CPs cluster as large as possible to satisfy the UE requirements. The cache capacity in the cluster formed by this algorithm is collectively used to provide higher content hit probability and diversity. Furthermore, considering the impact of social interests on the performance of caching strategies, a closed-form expression of the network hit ratio of the entire cache is derived on the basis of random geometry theory. Then, a network hit ratio maximization optimization problem is constructed and solved. The simulation results show that the proposed strategy has superior data offloading performance than other cooperative caching strategies.