Dynamic Power Allocation for Satellite-Terrestrial Communications with Cooperative Beam Hopping

In this paper, a multi-satellite cooperative beam hopping (BH) scenario in a low Earth orbit (LEO) satellite constellation is studied for satellite-terrestrial communications. Generally, the geographically non-uniform and time-varying packet arrival process, as well as the interference caused by dense beams, presents significant challenges in throughput maximization. To address these issues, an optimization problem is formulated to maximize long-term throughput in a multi-satellite cooperation scenario by determining the BH pattern and dynamic power allocation, while satisfying both peak and average power constraints. Subsequently, the original long-term problem is converted into decisions for each time slot through the Lyapunov drift-plus-penalty method. Moreover, the non-convex problem is solved with the successive convex approximation algorithm. Simulation results indicate that the proposed method can effectively improve throughput compared with the existing baseline, which demonstrates the effectiveness of the proposed method.