Delay minimization in NTNs: Deployment and caching optimization for satellite- and cache-aided UAV

This paper explores satellite- and cache-aided UAV communications for content delivery in non-terrestrial networks, demonstrating significant potential to offer widespread connectivity and high capacity. Specifically, the UAV provides caching to alleviate backhaul congestion, while the satellite supports the UAV's backhaul link. A problem is formulated to minimize content transmission delay. To address this problem, we design a user clustering and UAV deployment method based on an improved K-means algorithm, which determines the minimum number of UAVs and their deployment locations while ensuring QoS of users. Then, using linear programming relaxation and interior point methods, we obtain the optimal cache placement strategy. Simulations analyze the effects of Zipf parameter, backhaul bandwidth, and UAV altitude on system performance. The results and analysis can provide guidance for real-world network deployment.