Cost-Efficient UAV Deployment and Content Placement for Cellular Systems With D2D Communications

Caching content at base stations (BSs), specific user devices, and unmanned aerial vehicles (UAVs) has the potential to significantly enhance the content fetching performance of request users (RUs). In this article, the UAV deployment and content placement are jointly studied in UAV-assisted cellular systems with device-to-device communications, where UAVs, BS, and users are able to cache popular content. To reduce the system cost, i.e., the weighted sum of content fetching delay, energy consumption, and the deployment cost of UAVs, we come up with a cost function minimization problem with the joint optimization of UAV deployment, content placement, and RU association. As the formulated problem involves complicated nonconvex integer programming, we present a heuristic algorithm for the solution. Specifically, with the given initial UAV deployment strategy and content caching strategy, the original problem is reduced to a RU association problem. To deal with this one-to-many mapping problem, virtual BS and UAV nodes are introduced and Kuhn–Munkres algorithm is adopted. Thereafter, the obtained RU association strategy is adapted by examining the cache capacity constraints of BS and UAVs. To further reduce the system cost, redundant UAVs are removed via proposed cost difference-oriented greedy method. Finally, the optimal location of the remaining UAVs is determined by applying extensive search algorithm. Simulation results indicate that, compared with the scheme proposed in (Chen et al., 2019) and the case without UAV deployment, our approach saves the system cost by $10{\%}$ and $20{\%}$ , respectively.