Analytical Modeling for Caching Enabled UE-to-Network Relay in Cellular Networks

User equipment (UE)-to-network relay is a promising approach to assist UEs, especially cell-edge UEs, to obtain a better connectivity experience. By leveraging the proximity communication, specific UEs can serve as relays to assist transmissions between UEs and base stations (BSs). However, it also imposes a significant challenge on backhaul pressure. On the other hand, caching is considered as an efficient way to reduce data traffic and alleviate the burden of the backhaul networks. In this paper, we jointly consider UE-to-network relay and caching. We propose a tractable analytical framework to model caching enabled UE-to-network relay in cellular networks. Besides, we employ a four-level hierarchical mode selection scheme, where a request is satisfied with priority order by: 1) self-request mode, 2) device request mode, 3) UE-to-network relay mode, and 4) cellular request mode. Based on stochastic geometry, we derive tractable expressions for performance metrics, such as cache hit probability, successful transmission probability, and cache-aided throughput. The accuracy of analytical results is validated by simulations with the 3rd generation partnership project evaluation methodology. The results also show that the distance threshold should be set carefully to satisfy user communication requirements. Moreover, we find that the UE intensity and content popularity jointly have significant impacts on caching policy choices.