激光星间链路是低地球轨道卫星网络中实现高速率、高带宽、低功耗的有效方法,目前已经有很多针对激光星间链路的路由算法,但是大多数算法都假设激光星间链路是稳定的,没有考虑到因激光星间链路抖动而引起的丢包。本文提出了一种基于缓存与深度强化学习相结合的星间激光链路动态路由算法。该算法融合了分布式多智能体深度强化学习框架与双倍速率同步动态随机存储器缓存机制,通过计算主备路径并结合缓存重传显著降低了链路中断时的丢包率、时延以及实现了负载均衡。仿真结果表明,在简单场景下,我们提出的路由算法表现很好,在复杂场景下相较于传统深度Q网络和最短路径算法,其在丢包率、平均时延及负载均衡方面分别提升了15.14%、7.95%和58.25%,且表现与简单场景基本一致,有效增强了激光星间链路下低轨星座网络的鲁棒性。
Laser inter-satellite link (LISL) is an effective method to achieve high speed, high bandwidth, and low power consumption in low Earth orbit (LEO) satellite networks, and there are many routing algorithms for LISL, but most of the algorithms assume that the LISL is stable, and do not take into account the packet loss due to the LISL jitter. In this paper, we propose a Caching Based Dynamic Routing algorithm (CBDR) for interstellar laser links based on the combination of caching and deep reinforcement learning (DRL). The algorithm integrates a distributed multi-intelligence deep reinforcement learning framework with a double-rate synchronous dynamic random memory (DDR) caching mechanism, which significantly reduces the packet loss rate and delay during link interruption and achieves load balancing by calculating the primary and backup paths and combining with cache retransmission. Simulation results show that CBDR performs well in simple scenarios. Compared with traditional deep Q learning and shortest path algorithms, CBDR improves packet loss rate, average delay, and load balancing by 15.14%, 7.95%, and 58.25%, respectively, in complex scenarios, and its performance is basically consistent with that in simple scenarios, effectively enhancing the robustness of LEO constellation networks under LISL.
[1] Su Y T, Liu Y Q, Zhou Y Q, et al.Broadband LEO satellite communications: Architectures and key technologies[J]. IEEE Wireless Communications, 2019, 26(2): 55-61. DOI: 10.1109/MWC.2019.1800299.
[2] 邹恒光, 惠腾飞, 翟盛华, 等. 卫星通信技术发展综述[J]. 空间电子技术, 2025, 22(S1): 1-19.
[3] Chan V W S. Free space optical communication and network architecture-fundamentals, history and looking forward[J]. IEEE Journal of Selected Topics in Quantum Electronics, 2026, 32(1: Advances in Free Space Laser Communications): 9900120. DOI: 10.1109/JSTQE.2025.3600154.
[4] Chaudhry A U, Yanikomeroglu H.Laser intersatellite links in a starlink constellation: A classification and analysis[J]. IEEE Vehicular Technology Magazine, 2021, 16(2): 48-56. DOI: 10.1109/MVT.2021.3063706.
[5] Bhattacharjee D, Madoery P G, Chaudhry A U, et al.On-demand routing in LEO mega-constellations with dynamic laser inter-satellite links[J]. IEEE Transactions on Aerospace and Electronic Systems, 2024, 60(5): 7089-7105. DOI: 10.1109/TAES.2024.3415571.
[6] 李锐, 林宝军, 刘迎春, 等. 激光星间链路发展综述: 现状、趋势、展望[J]. 红外与激光工程, 2023, 52(3): 20220393. DOI: 10.3788/IRLA20220393.
[7] Kaymak Y, Rojas-Cessa R, Feng J H, et al.A survey on acquisition, tracking, and pointing mechanisms for mobile free-space optical communications[J]. IEEE Communications Surveys & Tutorials, 2018, 20(2): 1104-1123. DOI: 10.1109/COMST.2018.2804323.
[8] Kaushal H, Kaddoum G.Optical communication in space: Challenges and mitigation techniques[J]. IEEE Communications Surveys & Tutorials, 2017, 19(1): 57-96. DOI: 10.1109/COMST.2016.2603518.
[9] Tang K Y, Sun X B.Congestion control and routing optimization for LEO satellite networks based on deep reinforcement learning[C]//2024 IEEE 7th International Conference on Automation, Electronics and Electrical Engineering (AUTEEE). December 27-29, 2024, Shenyang, China. IEEE, 2024: 746-749. DOI: 10.1109/AUTEEE62881.2024.10869716.
[10] Zuo P L, Wang C, Yao Z, et al.An intelligent routing algorithm for LEO satellites based on deep reinforcement learning[C]//2021 IEEE 94th Vehicular Technology Conference (VTC2021-Fall). September 27-30, 2021, Norman, OK, USA. IEEE, 2021: 1-5. DOI: 10.1109/VTC2021-Fall52928.2021.9625325.
[11] Ding Z L, Liu H J, Tian F, et al.SDDRL-SR: A high-reliability satellite routing algorithm based on deep reinforcement learning[C]//2024 IEEE 99th Vehicular Technology Conference (VTC2024-Spring). June 24-27, 2024, Singapore, Singapore. IEEE, 2024: 1-6. DOI: 10.1109/VTC2024-Spring62846. 2024. 10683203.
[12] 刘兆洋, 潘必韬. 基于强化学习的光网络自适应高效率RWA算法[J]. 光通信研究, 2024(5): 66-73. DOI: 10.13756/j.gtxyj.2024.240024.
[13] Ran Y Y, Ding Y J, Chen S W, et al.Fully-distributed dynamic packet routing for LEO satellite networks: A GNN-enhanced multi-agent reinforcement learning approach[J]. IEEE Transactions on Vehicular Technology, 2025, 74(3): 5229-5234. DOI: 10.1109/TVT. 2024.3499933.
[14] 汪红光, 江逸茗, 刘侠君, 等. 基于强化学习的软件定义卫星互联网自适应负载均衡策略[J/OL]. 计算机科学, 2025: 1-9. (2025-07-07)[2025-10-22]. https://kns.cnki.net/KCMS/detail/detail.aspx?filename=JSJA2025070400I&dbname=CJFD&dbcode=CJFQ.
[16] Lyu Y F, Hu H, Fan R F, et al.Dynamic routing for integrated satellite-terrestrial networks: A constrained multi-agent reinforcement learning approach[J]. IEEE Journal on Selected Areas in Communications, 2024, 42(5): 1204-1218. DOI: 10.1109/JSAC.2024.3365869.
[17] Li Y B, Chen Y.Propagation modeling and analysis for terahertz inter-satellite communications using FDTD methods[C]//2021 IEEE International Conference on Communications Workshops (ICC Workshops). June 14-23, 2021, Montreal, QC, Canada. IEEE, 2021: 1-6. DOI: 10.1109/ICCWorkshops 50388. 2021.9473712.
[18] Zhang S B, Liu A J, Han C, et al.GRLR: routing with graph neural network and reinforcement learning for mega LEO satellite constellations[J]. IEEE Transactions on Vehicular Technology, 2025, 74(2): 3225-3237. DOI: 10.1109/TVT.2024.3471658.
