Deep Reinforcement Learning-Based Satellite-Ground Links Scheduling for Mega Satellite Constellations
- Resource Type
- Conference
- Authors
- Yang, Bo; Xi, Chao; Li, Gong; Liu, Peizhang; Zhu, Ruijie
- Source
- 2022 Asia Communications and Photonics Conference (ACP) Asia Communications and Photonics Conference (ACP), 2022. :1284-1288 Nov, 2022
- Subject
- Communication, Networking and Broadcast Technologies
Photonics and Electrooptics
Deep learning
Satellite constellations
Satellite antennas
Satellites
Scheduling algorithms
Simulation
Switches
deep reinforcement learning (DRL)
mega satellite constellations
satellite-ground links scheduling
- Language
We propose a satellite-ground links scheduling algorithm (DRSGLS) based on deep reinforcement learning. Simulation results show that the DRSGLS algorithm outperforms the other two algorithms in average resource utilization, average switch times, and scheduling runtime.