Hierarchical Distributed Q-learning-based resource allocation and UBS control in SATIN

This paper proposes an algorithm to overcome the limitations of traditional ground-based stations (GBS) in providing communication services to the satellite-air-terrestrial integrated network (SATIN). An algorithm integrates low earth orbit (LEO) satellites and unmanned aerial vehicle base stations (UBS) to create a dynamic communication network with extensive coverage. Challenges arise due to LEO propagation delay and computational complexity, and cross-tier channel interference issues with efficient use of frequency resources through integrated access and backhaul (IAB) [1]. To overcome these challenges, this research proposes the hierarchical distributed Q-learning algorithm that maximizes the network sum rate through resource allocation and UBS control. To overcome these challenges, this research proposes the hierarchical distributed Q-learning algorithm that maximizes the network sum rate through resource allocation and UBS control.