Affinity-Based User Clustering for Efficient Edge Caching in Content-Centric Cellular Networks

Network densification by small cell networks has become the main alternative for mobile network operators to deal with an ever-increasing traffic growth. In this context, Named Data Networking (NDN), an evolution of Content-Centric Networking (CCN), emerges as an alternative to improve data offloading by the promotion of in-network caching. In this paper, we propose a user clustering scheme that takes advantage of the affinity among users with respect to frequency of content requisition and common interest for content for a more efficient edge caching. The proposed strategy is evaluated in a varied set of scenarios, including different cache sizes, communication models, or concentration levels of content popularity. Simulation results show that the proposed strategy increases both cache hit ratio and data offloading in Content-Centric Cellular Networks.