Abstract
With the rapid expansion of satellite communication applications, the flexibility in utilizing on-board resources becomes increasingly significant. In this letter, we consider a multi-satellite system that combines beam hopping (BH) and non-orthogonal multiple access (NOMA) techniques. Satellites employ BH and NOMA to flexibly schedule terminals and allocate power. To make full use of BH and NOMA techniques, a joint optimization problem of beamforming, subchannel, and power allocation is proposed. It is a non-convex optimization problem with mixed variables. However, the combination of discrete and continuous variables becomes the bottleneck for providing an effective solution with limited system information. To deal with this challenge, we propose a novel intelligent-optimization coupling framework. This framework integrates proximal policy optimization (PPO) and geometric programming (GP) algorithms. PPO handles beamforming and subchannel allocation, while GP addresses power allocation, with the two iteratively solving the joint optimization problem. Extensive simulation results demonstrate the superiority of the proposed scheme compared to the conventional schemes.
Funding Information
  • National Natural Science Foundation of China (62271093)
  • National Key Laboratory of Advanced Communication Networks Foundation Project (HHX25641X004)

This publication has 14 references indexed in Scilit: