Joint User Association and Caching Placement for Cache-Enabled UAVs in Cellular Networks

Cache-enabled unmanned aerial vehicles (UAVs) are considered for caching popular contents and providing downlink data offloading in cellular networks in this paper. In this context, we formulate a joint optimization problem of user association and caching placement for minimizing content acquisition delay with consideration of the energy limitation of UAVs. A matching theory based algorithm is proposed for solving it. More particularly, a three-dimensional (3D) matching problem is formulated to model the relationships among UAVs, users and contents. We decompose it into two two-dimensional (2D) matching problems and provide a low-complexity two-step algorithm. Simulation results demonstrate that: i) the proposed cache-enabled UAV framework is capable of outperforming the conventional UAVassisted cellular networks without cache in terms of content acquisition delay; and ii) the proposed algorithm can achieve a significant reduction of content acquisition delay against two benchmark algorithms without joint optimization of user association and caching placement.