Dynamic Beam Hopping for Coverage Enhancement in Multi-Beam Satellite System Based on Determinantal Point Process Learning

As channel information changes rapidly in satellite communication systems, how to effectively improve the communication quality while enhancing coverage has become an important issue needs to be solved. A multi-beam coverage enhancement beam hopping (BH) scheme is designed in this paper, aiming to improve the channel quality of high-speed mobile terminals in low earth orbit (LEO) satellite communication system. Through the joint optimization of precoder and beam scheduling, system throughput and system delay is balanced. However, since the problem is non-convex, we adopt a data-driven determinantal point process (DPP) learning method to simplify the beam scheduling problem and design a precoder by taking advantage of the angular domain channel property. Simulation results show that compared with single-beam scheme, the proposed adaptive multi-beam coverage enhancement scheme increases the system throughput by 6% while reduces the average delay by 60%, and the DPP-based BH scheme using the improved diversity model shortens the calculation time by over 54%.