AI-Predicted Driven Hierarchically-Coordinated QoS Guarantee Mechanism for LEO Walker Constellation

LI Yi; QIAO Yanzhen; Zhou Kangyan; LIU Lihao; JI Zhiyuan; ZHAO Hong

doi:doi：10.11959/j.issn.1000-0801.2025041

您当前的位置：

首页 >

文章列表页 >

AI-Predicted Driven Hierarchically-Coordinated QoS Guarantee Mechanism for LEO Walker Constellation

Studies | 更新时间：2026-01-09

- AI-Predicted Driven Hierarchically-Coordinated QoS Guarantee Mechanism for LEO Walker Constellation
- Space-Integrated-Ground Information Networks Vol. 6, Issue 4, Pages: 84-95(2025)
- 作者机构：
  
  中国星网网络系统研究院有限公司，北京 100001
- 作者简介：
- 基金信息：
- DOI：doi：10.11959/j.issn.1000-0801.2025041
  CLC： TP393
- Received：07 June 2025，
  
  Revised：2025-07-22，
  
  Published：20 December 2025
- 稿件说明：
移动端阅览
李毅,乔艳珍,周康燕等.面向LEO Walker星座的AI预测驱动分层协同QoS保障机制[J].天地一体化信息网络,2025,06(04):84-95.

LI Yi,QIAO Yanzhen,Zhou Kangyan,et al.AI-Predicted Driven Hierarchically-Coordinated QoS Guarantee Mechanism for LEO Walker Constellation[J].Space-Integrated-Ground Information Networks,2025,06(04):84-95.
李毅,乔艳珍,周康燕等.面向LEO Walker星座的AI预测驱动分层协同QoS保障机制[J].天地一体化信息网络,2025,06(04):84-95. DOI： doi：10.11959/j.issn.1000-0801.2025041.

LI Yi,QIAO Yanzhen,Zhou Kangyan,et al.AI-Predicted Driven Hierarchically-Coordinated QoS Guarantee Mechanism for LEO Walker Constellation[J].Space-Integrated-Ground Information Networks,2025,06(04):84-95. DOI： doi：10.11959/j.issn.1000-0801.2025041.

摘要

为提升LEO Walker星座的QoS保障能力，提出一种AI预测驱动分层协同QoS保障架构。该架构通过地面智能控制中心内置的融合AI引擎预测链路质量与业务需求，利用深度强化学习优化全局QoS与动态网络切片，并在卫星节点部署轻量级智能代理执行本地QoS保障。同时阐述系统架构、关键AI模型、深度强化学习优化、动态网络切片管理及分层协同控制。仿真结果表明该架构在时延、抖动、数据包传递率、资源利用率等性能上有显著提升。

Abstract

This paper proposes an AI-predicted driven hierarchically-coordinated QoS (AIP-HC-QoS) architecture for LEO Walker constellation. The architecture employs an AI engine in the ground intelligence and control center (GICC) for link quality and service demand prediction

while deep reinforcement learning optimizes global QoS and dynamic network slicing. Lightweight intelligent agents are deployed on satellite nodes for local QoS execution. This paper details the system architecture

key AI models

DRL optimization

dynamic network slicing management

and hierarchical coordination control. Simulations results indicate significant improvements in delay

jitter

PDR

resource utilization

etc.

关键词

Keywords

references

缪德山 , 邓凌越 , 孙建成 , 等 . 6G星地融合无线网络及关键技术 [J ] . 中兴通讯技术 , 2024 , 30 ( 4 ): 42 - 49 .

MIAO D S , DENG L Y , SUN J C , et al . Terrestrial and satellite network integration based 6G systemand its key technologies [J ] . ZTE Technology Journal , 2024 , 30 ( 4 ): 42 - 49 .

陆航 , 张佳鑫 , 王海旭 . 面向6G的低轨卫星通信系统移动性管理技术 [J ] . 天地一体化信息网络 , 2025 , 6 ( 1 ): 35 - 44 .

LU H , ZHANG J X , WANG H X . Mobility management technologies for 6G-oriented low earth orbit satellite communication systems [J ] . Space-Integrated-Ground Information Networks , 2025 , 6 ( 1 ): 35 - 44 .

郝子慧 , 刘敏 , 齐彧 , 等 . Starlink卫星星座项目进展与未来前景分析 [J ] . 国际太空 , 2025 ( 4 ): 60 - 66 .

HAO Z H , LIU M , QI Y , et al . Progress and future prospect analysis of Starlink satellite constellation project [J ] . Space International , 2025 ( 4 ): 60 - 66 .

GIORDANI M , POLESE M , MEZZAVILLA M , et al . Toward 6G networks: use cases and technologies [J ] . IEEE Communications Magazine , 2020 , 58 ( 3 ): 55 - 61 .

DI B Y , SONG L Y , LI Y H , et al . Ultra-dense LEO: integration of satellite access networks into 5G and beyond [J ] . IEEE Wireless Communications , 2019 , 26 ( 2 ): 62 - 69 .

王汝言 , 叶显熠 , 何鹏 , 等 . 面向非地面网络多QoS保障的低地球轨道卫星星座设计 [J ] . 通信学报 , 2023 , 44 ( 12 ): 61 - 77 .

WANG R Y , YE X Y , HE P , et al . LEO satellite constellation design with multi-QoS guarantee for non-terrestrial network [J ] . Journal on Communications , 2023 , 44 ( 12 ): 61 - 77 .

Al-HARTHI Y , El-KHAMY S E , El-DOLIL S A . A survey on QoS-aware routing protocols for LEO satellite networks: challenges, architectures, and future directions [J ] . IEEE Access , 2022 , 10 : 24870 - 24895 .

PAPA A , BIRD J N , DOBRE O A . Machine learning for QoS management in LEO satellite constellations: a survey [J ] . IEEE Access , 2022 , 10 : 100830 - 100858 .

ALSENWI M , TRAN N H , BENNIS M , et al . Intelligent resource slicing for eMBB and uRLLC coexistence in 5G and beyond: a deep reinforcement learning based approach [J ] . IEEE Transactions on Wireless Communications , 2021 , 20 ( 7 ): 4585 - 4600 .

VERBRAEKEN J , WOLTING M , KATZY J , et al . A survey on distributed machine learning [J ] . ACM Computing Surveys (CSUR) , 2020 , 53 ( 2 ): 1 - 33 .

GUIDOTTI A , VANELLI-CORALLI A , CHIANI M . Machine learning for satellite communications: a comprehensive survey [J ] . IEEE Open Journal of the Communications Society , 2021 , 2 : 1799 - 1838 .

WANG P , ZHANG J , CHEN S . LSTM-based channel prediction for LEO satellite communication systems [J ] . IEEE Wireless Communications Letters , 2021 , 10 ( 5 ): 993 - 997 .

LIU S , GAO Y , CHEN X . Spatio-temporal graph neural network for link availability prediction in LEO constellations [J ] . IEEE Journal on Selected Areas in Communications , 2023 , 41 ( 5 ): 1450 - 1463 .

WEI J , HE R , CHEN K , et al . DST-LEO: A dynamic spatio-temporal graph neural network for traffic prediction in LEO satellite constellations [C ] // Proceedings of ACM MobiHoc . NewYork : ACM Press , 2022 : 1 - 10 .

GUO K , AN K , ZHANG B , et al . Deep reinforcement learning based routing for LEO satellite networks [J ] . IEEE Transactions on Vehicular Technology , 2021 , 70 ( 11 ): 11738 - 11749 .

CUI T , ZHANG X , DONG P , et al . Deep reinforcement learning-based resource allocation in LEO satellite networks for IoT services [J ] . IEEE Internet of Things Journal , 2022 , 9 ( 17 ): 16309 - 16320 .

ORSINO A , ARANITI G , MILITANO L , et al . Network slicing for non-terrestrial networks: a survey [J ] . IEEE Access , 2021 , 9 : 128573 - 128591 .

POLESE M , JANA R , ZORZI M . An O-RAN-compliant and AI-empowered end-to-end network slicing framework for 6G [J ] . IEEE Communications Magazine , 2022 , 60 ( 5 ): 76 - 82 .

BEGA D , GRAMAGLIA M , BANCHS A , et al . EvoNetSlice: an evolutionary model for dynamic resource allocation and real-time QoS adaptation in 6G network slicing [J ] . IEEE Journal on Selected Areas in Communications , 2020 , 38 ( 8 ), 1766 - 1778 .

SAVAZZI S , NICOLI M , RAMPA V , et al . Federated learning for LEO-based earth observation [J ] . IEEE Transactions on Wireless Communications , 2020 , 19 ( 10 ): 6582 - 6595 .

HAN S , MAO H , DALLY W J . Deep compression: Compressing deep neural networks with pruning, trained quantization and huffman coding [C ] // Proceedings of International Conference on Learning Representations (ICLR) . San Diego : ICLR Press , 2016 : 1 - 12 .

YAO H , WANG C X , HUANG J , et al . A survey of channel measurements and models for satellite communication systems [J ] . IEEE Communications Surveys & Tutorials , 2021 , 23 ( 1 ): 150 - 179 .

ZHANG L , ZHAO H , HOU S , et al . A survey on artificial intelligence in non-terrestrial networks [J ] . Chinese Journal of Aeronautics , 2022 , 35 ( 11 ): 1 - 20 .

LIU J J , SHI Y P , FADLULLAH Z M , et al . Space-air-ground integrated network: a survey [J ] . IEEE Communications Surveys & Tutorials , 2018 , 20 ( 4 ): 2714 - 2741 .

SU Z , XU Q , ZHANG R . Deep reinforcement learning for dynamic resource allocation in geostationary satellite systems [J ] . IEEE Transactions on Wireless Communications , 2021 , 20 ( 3 ): 1534 - 1548 .

Views

下载量

CSCD

Alert me when the article has been cited

提交

Tools

Publicity Resources

QoS Assurance Key Technologies in Space-Ground Heterogeneous Network

Related Author

Lin LIN

Bin ZHU

Zelin WANG

Guangquan WANG

Haipeng YAO

Tao DONG

Weiying FENG

Related Institution

China Unicom Research Institute

Beijing University of Posts and Telecommunications

Aerospace Star Technology Co., Ltd.

Beihang University

⁰