Low Earth orbit satellite switching and recovery method based on generic congestion control algorithm

• Published in: Scientific reports Vol. 15; no. 1; pp. 25341 - 15
• Main Authors: Chen, Hao, Yu, Hang, Zhang, Hui, Chen, Huancheng, Kong, Xiangsen, Li, Huaichuan, Qing, Demao
• Format: Journal Article
• Language: English
• Published: London Nature Publishing Group UK 14.07.2025; Nature Publishing Group; Nature Portfolio
• Subjects: 639/705/117; 639/766/1130; Aging; Algorithms; Artificial satellites; Communications networks; Congestion control; Control algorithms; Data transmission; Deep learning; Earth orbits; Efficiency; Energy consumption; Genetic algorithms; Ground stations; Humanities and Social Sciences; Internet access; Internet of Things; Linear programming; Link switching; Low Earth orbit satellite network; Low earth orbit satellites; multidisciplinary; Network congestion; Neural network; Neural networks; Optimization; Quality of service; Satellite communications; Satellites; Science; Science (multidisciplinary); Congestion control; Network congestion; Neural network; Low Earth orbit satellite network; Link switching
• ISSN: 2045-2322; 2045-2322
• DOI: 10.1038/s41598-025-10649-z

With the rapid development of low Earth orbit (LEO) satellite communication technology, its applications in mobile communication and data transmission are becoming increasingly widespread. To improve the user experience for video services and reduce communication interruptions and network congestion during link switching, this study proposes a LEO satellite switching recovery method based on an enhanced congestion control algorithm. The method includes a neural network-based attitude control strategy, a handover detection mechanism incorporating the satellite’s remaining service time, and a rate recovery algorithm combining congestion control with time series prediction. Simulation results show that the proposed method reduces the communication interruption rate from 6.48 to 2.13%, increases the minimum throughput from 257 kbps to 2017 kbps, and shortens the average recovery time from 3200 ms to 1800 ms. Moreover, the false detection rate decreases by 71.6%, and the detection accuracy improves to 96.32%. Under conditions of system stability exceeding 95%, the method maintains good real-time performance, with average processing times of 5.23 s (training) and 4.58 s (validation). These results demonstrate the effectiveness of the proposed method in improving switching and recovery performance in LEO satellite networks and provide theoretical and technical support for stable and efficient satellite system operations.