GEO卫星多波束动态频率-功率联合分配研究Research on Joint Allocation for Multi-beam Dynamic Frequency-power of GEO Satellite
朱浩宁,李静涛,董赞扬,聂高峰
摘要(Abstract):
针对地球同步轨道(GEO)多波束卫星面临的频率、功率、干扰约束,建立以总服务率最大化为目标的系统优化模型,提出一种基于深度强化学习(DRL)的分支深度Q网络(DQN)算法,实现多波束动态频率-功率联合分配。其中,将联合动作拆分为起始频率槽、频率槽数量和功率档位3个子决策,以有效降低所提DQN动作空间维度。仿真结果表明:分支DQN算法服务率达86.5%,较传统近端策略优化(PPO)算法提升约5.0%,因此可应用于GEO卫星的频率规划与功率调整,以提高系统服务满足度。
关键词(KeyWords): 地球同步轨道卫星;多波束;频率-功率联合分配;深度Q网络
基金项目(Foundation):
作者(Author): 朱浩宁,李静涛,董赞扬,聂高峰
参考文献(References):
- [1] Rothblatt M A.Satellite communication and spectrum allocation [J].American Journal of International Law,1982,76(1):56-77.
- [2] Chen R,Long W X,Wang B Q,et al.Multibeam high throughput satellite:hardware foundation,resource allocation,and precoding:arXiv:2508.00800 [R].Ithaca,New York:Cornell University,2025.
- [3] Vázquez M ??,Perez-neira A,Christopoulos D,et al.Precoding in multibeam satellite communications:pre sent and future challenges [J].IEEE Wireless Communications,2016,23(6):88-95.
- [4] Takahashi M,Kawamoto Y,Kato N,et al.Adaptive power resource allocation with multi-beam directivity control in high-throughput satellite communication systems[J].IEEE Wireless Communications Letters,2019,8(4):1248-1251.
- [5] Zhang H,Ren D,Jiang F.A beam search-based channel allocation method for interference mitigation of NGSO satellites with multi-beam antennas [J].Aerospace,2022,9(4):177.
- [6] Luis J J G,Guerster M,Delportillo I,et al.Deep reinforcement learning for continuous power allocation in flexible high throughput satellites [C]//2019 IEEE Cognitive Communications for Aerospace Applications Workshop (CCAAW),2019:1-4.
- [7] Cocco G,Angelone M,Pèrez-neira A I.Co-channel interference cancelation at the user terminal in multibeam satellite systems[J].International Journal of Satellite Communications and Networking,2017,35(1):45-65.
- [8] Jing Y,Yang Z,Zhao Y,et al.Energy-efficient routing based on a genetic algorithm for satellite laser communication [J].Optics Express,2023,31(5):8682-8695.
- [9] Abbott R J,Campbell M L,Krenz W C.A sustainable genetic algorithm for satellite resource allocation[J].Telematics and Informatics,1995,12(3/4):141-159.
- [10] Zhao B,Liu J,Wei Z,et al.A deep reinforcement learning based approach for energy-efficient channel allocation in satellite internet of things[J].IEEE Access,2020,8:62197-62206.
- [11] Silva P R,Abreu I S,Forte P A,et al.Genetic algorithms for satellite launcher attitude controller design[J].Inteligencia Artificial,2019,22(63):150-161.
- [12] Lagunas E,Sharma S K,Maleki S,et al.Resource allocation for cognitive satellite communications with incumbent terrestrial networks [J].IEEE Transactions on Cognitive Communications and Networking,2015,1(3):305-317.
- [13] Qiu C,Yao H,Yu F R,et al.Deep Q-learning aided networking,caching,and computing resources allocation in software-defined satellite-terrestrial networks[J].IEEE Transactions on Vehicular Technology,2019,68(6):5871-5883.
- [14] Li Z,Xie Z,Liang X.Dynamic channel reservation strategy based on DQN algorithm for multi-service LEO satellite communication system [J].IEEE Wireless Communications Letters,2020,10(4):770-774.
- [15] Li X F,Chen J,Cai X T,et al.Satellite online sche duling algorithm based on proximal policy [C]//The 3rd International Conference on Artificial Intelligence in China,2022:100-108.
- [16] Liu X.Mission schedule of agile satellites based on proximal policy optimization algorithm:arXiv:2007.02352 [R].Ithaca,New York:Cornell University,2020.
- [17] Xie X,Fan K,Deng W,et al.Multi-satellite beam hopping and power allocation using deep reinforcement learning:arXiv:2501.02309 [R].Ithaca,New York:Cornell University,2025.