基于长短期记忆深度Q网络的卫星网络多业务路由算法

温京龙; 张怡; 魏德宾; 潘成胜

doi:10.11959/j.issn.1000-0801.DXKX250576

您当前的位置：

首页 >

文章列表页 >

基于长短期记忆深度Q网络的卫星网络多业务路由算法

研究与开发 | 更新时间：2026-06-17

- 基于长短期记忆深度Q网络的卫星网络多业务路由算法
- Multi-service routing algorithm for satellite networks based on deep Q-network with long short-term memory
- 电信科学 2026年42卷第5期页码：48-59
- 作者机构：
  
  1.大连大学信息工程学院，辽宁大连 116622
  2.南京邮电大学通信与网络技术国家工程研究中心，江苏南京 210023
- 作者简介：
  
  温京龙（2000- ），男，大连大学硕士生，主要研究方向为卫星网络路由、边缘计算和深度强化学习。
  张怡（2000- ），女，大连大学硕士生，主要研究方向为卫星网络路由、深度强化学习。
  魏德宾（1978- ），男，博士，大连大学副教授，主要研究方向为空间信息网络传输技术、流量工程和网络优化。
  潘成胜（1962- ），男，博士，南京邮电大学教授，主要研究方向为一体化智能网络流量理论与关键技术。
- 基金信息：
  
  国家自然科学基金资助项目(61931004)
- DOI：10.11959/j.issn.1000-0801.DXKX250576
  中图分类号： TP393;TN297
- 收稿：2025-09-30，
  
  修回：2025-12-12，
  
  录用：2025-12-25，
  
  纸质出版：2026-05-20
- 稿件说明：
移动端阅览
温京龙,张怡,魏德宾等.基于长短期记忆深度Q网络的卫星网络多业务路由算法[J].电信科学,2026,42(05):48-59.

Wen Jinglong,Zhang Yi,Wei Debin,et al.Multi-service routing algorithm for satellite networks based on deep Q-network with long short-term memory[J].Telecommunications Science,2026,42(05):48-59.
温京龙,张怡,魏德宾等.基于长短期记忆深度Q网络的卫星网络多业务路由算法[J].电信科学,2026,42(05):48-59. DOI： 10.11959/j.issn.1000-0801.DXKX250576.

Wen Jinglong,Zhang Yi,Wei Debin,et al.Multi-service routing algorithm for satellite networks based on deep Q-network with long short-term memory[J].Telecommunications Science,2026,42(05):48-59. DOI： 10.11959/j.issn.1000-0801.DXKX250576.

摘要

为了解决卫星网络中多业务服务质量保障与负载均衡的协同优化难题，提出了一种融合长短期记忆网络和深度Q网络的智能多业务路由算法。该算法把长短期记忆深度Q网络作为智能体部署在当前卫星节点，将当前节点与周围节点的链路时延、带宽、丢包率、流量、网络拓扑、业务类型作为网络状态输入智能体进行训练，输出动作是下一跳节点，使用带宽、时延、丢包率的加权和与最大链路带宽利用率的加权和作为奖励函数调整动作。智能体训练收敛后，卫星进行多业务传输。通过仿真实验和性能评估，研究结果表明所提出的算法在不同性能指标上均取得了显著成效，并且在平衡网络负载方面表现优异。

Abstract

To address the challenge of collaborative optimization between multi-service quality of service (QoS) guarantee and load balancing in satellite networks

an intelligent multi-service routing algorithm integrating long short-term memory with deep Q-Network was proposed. The algorithm deployed LSTM-DQN (LDQN) as an agent at the current satellite node

where the network state input to the agent includes link latency

bandwidth

packet loss rate

traffic

network topology and service type between the current node and its neighboring nodes. The output action was the next-hop node selection for the current node. The reward function

designed to guide action adjustment

combined the weighted sum of bandwidth

latency

and packet loss rate with the weighted sum of maximum link bandwidth utilization. After the agent’s training convergence

multi-service transmission was performed in the satellite network. Simulation experiments and performance evaluations demonstrate that the proposed algorithm achieves significant improvements across various performance metrics while exhibiting outstanding load balancing capabilities.

关键词

Keywords

references

卢雪玉 , 魏雯婷 , 伏丽莹 , 等 . 基于拓扑感知的软件定义卫星网络多业务路由 [J ] . 西安电子科技大学学报(自然科学版) , 2024 , 51 ( 6 ): 10 - 24 .

Lu X Y , Wei W T , Fu L Y , et al . Topology awareness based multi-service routing for the software-defined satellite network [J ] . Journal of Xidian University (Natural Science) , 2024 , 51 ( 6 ): 10 - 24 .

王朝炜 , 杜嘉楠 , 王程 , 等 . 软件定义卫星网络中基于业务调度的混合路由算法 [J ] . 电子学报 , 2024 , 52 ( 5 ): 1506 - 1515 .

Wang C W , Du J N , Wang C , et al . A hybrid routing based on traffic scheduling in double-layer software defined satellite networks [J ] . Acta Electronica Sinica , 2024 , 52 ( 5 ): 1506 - 1515 .

Wu Y , Hu G Y , Jin F L , et al . Multi-objective optimisation in multi-QoS routing strategy for software-defined satellite network [J ] . Sensors , 2021 , 21 ( 19 ): 6356 - 6372 .

Jiang W J , Zong P . QoS routing algorithm based on traffic classification in LEO satellite networks [C ] // Proceedings of the 2011 Eighth International Conference on Wireless and Optical Communications Networks . Piscataway : IEEE Press , 2011 : 1 - 5 .

张慧卿 , 彭宏利 . 基于Pareto多目标遗传的LEO卫星网络多业务QoS路由算法 [J ] . 电子设计工程 , 2017 , 25 ( 20 ): 116 - 120 .

Zhang H Q , Peng H L . LEO network multi-service QoS routing using Pareto multi-object genetic algorithm [J ] . Electronic Design Engineering , 2017 , 25 ( 20 ): 116 - 120 .

Xie T J . SDSN: software-defined space networking: architecture and routing algorithm [J ] . Mobile Networks and Applications , 2019 , 24 ( 5 ): 1542 - 1554 .

Jiao J , Yang P L , Du Z , et al . Clustered multi-criteria routing algorithm for mega low earth orbit satellite constellations [J ] . IEEE Transactions on Vehicular Technology , 2024 , 73 ( 9 ): 13790 - 13803 .

王奎宇 , 宋晓勤 , 缪娟娟 , 等 . 基于SDN的高性能QoS保障低轨道卫星星间路由算法 [J ] . 计算机工程 , 2022 , 48 ( 5 ): 185 - 190, 199 .

Wang K Y , Song X Q , Miao J J , et al . SDN-based high-performance and QoS guaranteed inter-satellite routing algorithm for low-earth orbit satellites [J ] . Computer Engineering , 2022 , 48 ( 5 ): 185 - 190, 199 .

Wang F , Jiang D , Wang Z , et al . Fuzzy-CNN based multi-task routing for integrated satellite-terrestrial networks [J ] . IEEE Transactions on Vehicular Technology , 2021 , 71 ( 2 ): 1913 - 1926 .

Wei W T , Fu L Y , Gu H X , et al . Iris: toward intelligent reliable routing for software-defined satellite networks [J ] . IEEE Transactions on Communications , 2025 , 73 ( 1 ): 454 - 468 .

汪昊 , 冉泳屹 , 赵雷 , 等 . 基于深度图强化学习的低轨卫星网络动态路由算法 [J ] . 重庆邮电大学学报(自然科学版) , 2023 , 35 ( 4 ): 596 - 605 .

Wang H , Ran Y Y , Zhao L , et al . Dynamic routing algorithm for LEO satellite networks based on deep graph reinforcement learning [J ] . Journal of Chongqing University of Posts and Telecommunications (Natural Science Edition) , 2023 , 35 ( 4 ): 596 - 605 .

Dong F , Song J , Zhang Y , et al . DRL-based load-balancing routing scheme for 6G space-air-ground integrated networks [J ] . Remote Sensing , 2023 , 15 ( 11 ): 2801 - 2821 .

李新桐 , 张亚生 . 一种适用于低轨卫星的SDN网络人工智能路由方法 [J ] . 电子测量技术 , 2020 , 43 ( 22 ): 109 - 114 .

Li X T , Zhang Y S . Artificial intelligence routing method for SDN network suitable for LEO satellites [J ] . Electronic Measurement Technology , 2020 , 43 ( 22 ): 109 - 114 .

Wang C , Wang H , Wang W . A two-hops state-aware routing strategy based on deep reinforcement learning for LEO satellite networks [J ] . Electronics , 2019 , 8 ( 9 ): 920 - 936 .

Xu G L , Zhao Y Y , Ran Y Y , et al . Spatial location aided fully-distributed dynamic routing for large-scale LEO satellite networks [J ] . IEEE Communications Letters , 2022 , 26 ( 12 ): 3034 - 3038 .

Chu K , Cheng S , Zhu L . A robust routing strategy based on deep reinforcement learning for mega satellite constellations [J ] . Electronics Letters , 2023 , 59 ( 11 ): 12820 - 12823 .

Shi Y , Yuan Z , Zhu X , et al . An adaptive routing algorithm for inter-satellite networks based on the combination of multipath transmission and Q-Learning [J ] . Processes , 2023 , 11 ( 1 ): 167 - 187 .

翟凡妮 . SDN中基于强化学习的多业务智能QoS路由方法 [J ] . 智能物联技术 , 2025 , 57 ( 1 ): 44 - 47 .

Zhai F N . Multi service intelligent QoS routing method based on reinforcement learning in SDN [J ] . Technology of IoT&AI , 2025 , 57 ( 1 ): 44 - 47 .

Wang M , Di X Q , Jiang Y M , et al . End-to-end stochastic QoS performance under multi-layered satellite network [C ] // Proceedings of International Conference on Space Information Networks . Singapore : Springer , 2017 : 182 - 201 .

Wei D B , Cao Y , Teng Q X , et al . Software defined network-based multi service routing algorithm for spatial information network [J ] . Journal of Physics: Conference Series , 2023 , 2558 ( 1 ): 012002 .

浏览量

下载量

CSCD

文章被引用时，请邮件提醒。

提交

工具集

关联资源

星地边缘计算网络中基于 CA3C的异构任务卸载方法研究

一种面向资源高效利用的无蜂窝RAN分层协同资源分配算法

基于深度强化学习的数据传输策略优化研究

面向异构算力互联的智算网络关键技术研究

基于时频变换与动态图注意力的多模态序列推荐方法