面向交互式视频业务的MEO-LEO分层协同卫星路由策略

王波; 卢锐; 万锦辉; 王吉平

doi:10.11959/j.issn.1000-436x.TXXB260125

您当前的位置：

首页 >

文章列表页 >

面向交互式视频业务的MEO-LEO分层协同卫星路由策略

更新时间：2026-05-14

- 面向交互式视频业务的MEO-LEO分层协同卫星路由策略
- MEO–LEO hierarchical collaborative satellite routing strategy for interactive video services
- 通信学报 2026年
- 作者机构：
  
  1.桂林电子科技大学信息与通信学院，广西桂林 541004
  2.广西科技信息网络中心，广西南宁 530012
- 作者简介：
  
  万锦辉，邮箱：wlzx@kjt.gxzf.gov.cn
- 基金信息：
  
  面向6G的LEO卫星时空能量预测与绿色调度优化研究(C25GAH30)
- DOI：10.11959/j.issn.1000-436x.TXXB260125
  中图分类号： TN927.2
- 收稿：2026-03-18，
  
  修回：2026-05-13，
  
  录用：2026-05-14，
- 稿件说明：
移动端阅览
王波, 卢锐, 万锦辉, 等. 面向交互式视频业务的MEO-LEO分层协同卫星路由策略[J/OL]. 通信学报, 2026.

Wang Bo, Lu Rui, Wan Jinhui, et al. MEO–LEO hierarchical collaborative satellite routing strategy for interactive video services[J/OL]. Journal on Communications, 2026.
王波, 卢锐, 万锦辉, 等. 面向交互式视频业务的MEO-LEO分层协同卫星路由策略[J/OL]. 通信学报, 2026. DOI： 10.11959/j.issn.1000-436x.TXXB260125.

Wang Bo, Lu Rui, Wan Jinhui, et al. MEO–LEO hierarchical collaborative satellite routing strategy for interactive video services[J/OL]. Journal on Communications, 2026. DOI： 10.11959/j.issn.1000-436x.TXXB260125.

摘要

针对低轨卫星网络中交互式视频业务的低时延、低抖动和会话连续性需求，面向强时变拓扑与能量时空异构条件下路由决策复杂、链路切换频繁等问题，提出MEO–LEO分层协同路由策略H-GMRL。采用GNN与Double-DQN进行多约束全局路径规划，结合MARL与触发式平滑切换实现局部修复。仿真表明，所提策略平均端到端时延最高降低12.1%，时延抖动最高降低45.3%，分组丢包率最高降低44.8%，平均路径TTL最高提升25.9%，可增强高负载场景下交互式视频业务的连续稳定承载能力。

Abstract

To meet the requirements of interactive video services in low Earth orbit satellite networks for low latency

low jitter

and session continuity

a hierarchical collaborative routing strategy

termed H-GMRL

was proposed for medium Earth orbit (MEO)–low Earth orbit (LEO) satellite networks to address complex routing decisions and frequent link switching under highly dynamic topologies and spatiotemporally heterogeneous energy conditions. Graph neural networks (GNNs) and Double Deep Q-Network (Double-DQN) were adopted for multi-constraint global path planning

while multi-agent reinforcement learning (MARL) together with a trigger-based smooth switching mechanism was used for local repair. Simulation results show that the proposed strategy reduces the average end-to-end delay by up to 12.1%

decreases delay jitter by up to 45.3%

lowers the packet loss rate by up to 44.8%

and increases the average path time to live (TTL) by up to 25.9%. It provides more stable and continuous support for interactive video services under high-load scenarios.

关键词

Keywords

references

Zhu X , Jiang C . Creating efficient integrated satellite-terrestrial networks in the 6G era [J ] . IEEE Wireless Communications , 2022 , 29 ( 4 ): 154 - 160 .

Wang X , Yu R , Yang D , et al . Space booking: enabling performance-critical applications in broadband satellite networks [C ] // Proceedings of the IEEE International Conference on Distributed Computing Systems (ICDCS) . Glasgow : IEEE , 2025 : 989 - 999 .

Zhao N , Long X , Wang J . A multi-constraint optimal routing algorithm in LEO satellite networks [J ] . Wireless Networks , 2021 , 27 ( 4 ): 1107 - 1117 .

Wang C , Liu Y , Zhang X . Energy-efficient routing in LEO satellite networks for extending satellites’ battery lifetime [J ] . Journal of Communications and Networks , 2022 , 24 ( 3 ): 45 - 56 .

Li Y , Zhang H , Qi Y , et al . Available energy routing algorithm considering QoS requirements for LEO satellite networks [J ] . Computer Communications , 2024 , 188 : 120 - 132 .

Xu G , Zhao Y , Ran 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 .

Xiang J , He X , Zhao Y , et al . Distributed dynamic routing for LEO satellite networks with temporal graph convolutions and imitation acceleration [J ] . IEEE Communications Letters , 2025 , 29 ( 11 ): 2521 - 2525 .

Shi Y , Wang W , Zhu X , et al . Low Earth orbit satellite network routing algorithm based on graph neural networks and deep Q-network [J ] . Applied Sciences , 2024 , 14 ( 9 ): 3840 .

Ran Y , Ding Y , Chen S , 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 .

Zhao J , Pan J . Low-latency live video streaming over a low-Earth-orbit satellite network [C ] // Proceedings of the 15th ACM Multimedia Systems Conference (MMSys’24) . New York : ACM , 2024 : 100 - 108 .

Li S , Wu Q , Wang R . Dynamic discrete topology design and routing for satellite-terrestrial integrated networks [J ] . IEEE/ACM Transactions on Networking , 2024 , 32 ( 5 ): 3840 - 3853 .

Zhang S , Liu A , Han C , et al . Federated graph reinforcement learning-driven adaptive routing strategy for mega LEO satellite networks [J ] . IEEE Wireless Communications Letters , 2025 , 14 ( 10 ): 3344 - 3348 .

Li Q , Wang S , Ma X , et al . Battery-aware energy optimization for satellite edge computing [J ] . IEEE Transactions on Services Computing , 2024 , 17 ( 2 ): 437 - 451 .

Baccari S , Mostacciuolo E , Tipaldi M , et al . A deep reinforcement learning approach for energy management in low Earth orbit satellite electrical power systems [J ] . Electronics , 2025 , 14 : 3110 .

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

基于RL-WGAN的5G网络异常数据生成方法

FTTR C-WAN架构下多AP协同的Co-EDCA机制研究

基于IPv6+的智能车联算网调度方案设计与实现

水下无线传感器网络节点定位方案综述

面向未知网络威胁的网络要地自适应防御模型