Explicit data caching at mobile edge networks: A cross-layer perspective

To fully exploit the increasingly dense deployments of mobile base station (BS) infrastructure, this paper proposes a novel caching paradigm called explicit caching, which stores selective contents at BSs and exposes such contents to users for volunteer browsing and download. We formulate the explicit caching problem as a joint content recommendation and delivery problem, which aims to maximize the expected quality-of-experience (QoE) per user over the long term. Optimal algorithms are given to solve the problem with varying degrees of cross-layer information. The impacts of user interest profile, cache storage space, BS backhaul capacity, cross-layer information, user mobility, and user mobility prediction on the system performance are analyzed and discussed.