Research and Development of Ai-Based Resource Allocation Methods for Aerial and Terrestrial Base Stations in Communication Networks

This study examines artificial intelligence–based approaches for efficient resource allocation in hybrid aerial and terrestrial base station networks. As wireless traffic demand and service diversity continue to increase, conventional optimization techniques become less effective in dynamic and large-scale environments. Therefore, AI methods such as deep reinforcement learning, supervised learning, and graph neural networks are employed to optimize spectrum sharing, power control, user association, and UAV trajectory planning. Furthermore, the research considers cross-layer constraints including backhaul capacity, energy consumption, and quality-of-service requirements. The results demonstrate that AI-driven strategies enhance system throughput, fairness, adaptability, and energy efficiency while maintaining reliable performance in time-varying network conditions.