计算机科学

计算机科学 ›› 2026, Vol. 53 ›› Issue (7): 354-362.doi: 10.11896/jsjkx.250500060

• 计算机网络 • 上一篇    下一篇

面向时变拓扑与动态异构资源的星间多方协同计算卸载算法

商科峰1,2, 张丹1, 颛孙盈3, 李丹丹3, 刘言4,5,6, 朱凯歌4,5,6   

  1. 1 西南电子技术研究所 成都 610036
    2 智能测控与天基信息应用实验室 成都 610036
    3 北京邮电大学计算机学院(国家示范性软件学院) 北京 100876
    4 上海卫星互联网研究院有限公司 上海 201210
    5 上海市卫星互联网重点实验室 上海 201210
    6 卫星互联网全国重点实验室 上海 201210
  • 收稿日期:2025-05-15 修回日期:2025-09-23 出版日期:2026-07-15 发布日期:2026-07-10
  • 通讯作者: 李丹丹(dandl@bupt.edu.cn)
  • 作者简介:(shangkefeng@126.com)

Multi-party Inter-satellite Collaborative Computing Offloading Algorithm for Time-varying Topologies and Dynamic Heterogeneous Resources

SHANG Kefeng1,2, ZHANG Dan1, ZHUAN Sunying3, LI Dandan3, LIU Yan4,5,6, ZHU Kaige4,5,6   

  1. 1 Southwest China Institute of Electronic Technology,Chengdu 610036,China
    2 Intelligent TT&C and Space-based Application Laboratory,Chengdu 610036,China
    3 School of Computer Science(National Pilot Software Engineering School),Beijing University of Posts and Telecommunications,Beijing 100876,China
    4 Shanghai Satellite Network Research Institute Co.,Ltd.,Shanghai 201210,China
    5 Shanghai Key Laboratory of Satellite Network,Shanghai 201210,China
    6 State Key Laboratory of Satellite Network,Shanghai 201210,China
  • Received:2025-05-15 Revised:2025-09-23 Published:2026-07-15 Online:2026-07-10
  • About author:SHANG Kefeng,born in 1987,postgra-duate.His main research interests include aerospace tracking and control,constellation networking and applications.
    LI Dandan,born in 1987,Ph.D,asso-ciate professor,is a member of CCF(No.98186M).Her main research interests include next-generation Internet and edge intelligence.

PDF (PC)

20

摘要： 星地协同计算卸载场景中,由卫星先行完成部分计算任务,再将剩余任务卸载至地面节点接续完成。这种场景通常需要在星地之间传输大量数据,为资源有限的卫星带来了高昂成本。星间协同计算卸载可以在不依赖地面计算设备的情况下完成任务。然而,目前研究未能全面考量卫星网络拓扑的时变性以及卫星资源的动态性和异构性,降低了任务的成功率。因此,提出一种面向卫星时变拓扑与动态异构资源的星间多方协同计算卸载算法。具体而言,对于每个时隙内的任务集合,为使卸载算法适应时变的卫星拓扑和动态异构的卫星资源,算法首先收集当前时隙卫星网络的拓扑状态、各卫星的资源情况以及与地面的连接时长等关键信息。随后,以最小化任务时延和最大化任务成功率为优化目标,为每个任务选定执行任务的卫星及协同计算的邻接卫星。基于STK工具采集每个时隙的卫星网络拓扑、连接时间等数据,实验结果表明,与基线算法相比,所提出的算法具有更高的任务成功率和更低的任务时延。

关键词: 低轨卫星星座, 卫星网络, 深度强化学习, 协同计算, 计算卸载

Abstract: In the scenario of satellite-ground collaborative computing offloading,the satellite first computes part of the tasks and then offloads the remaining tasks to the ground for completion.This scenario typically requires the transmission of a large amount of data between the satellite and the ground,which incurs high costs for satellites with limited resources.Inter-satellite collaborative computing offloading can complete tasks without relying on ground computing devices.However,current research has not comprehensively considered the time-varying nature of satellite network topology,the dynamic and heterogeneous nature of satellite resources,which reduces the success rate of tasks.Therefore,an inter-satellite multi-party collaborative computing offloading algorithm for time-varying satellite topology and dynamic heterogeneous resources is proposed.Specifically,for the task set in each time slot,to make the offloading algorithm adapt to the time-varying satellite topology and dynamic heterogeneous satellite resources,the algorithm first collects key information such as the current time slot's satellite network topology,the resource status of each satellite,and the connection duration with the ground.Subsequently,the ground center optimizes task allocation with the dual objectives of minimizing task latency and maximizing task success rate.The primary objectives are to minimize task delay and maximize the task success rate for each task.Experimentally,dynamic topological data of the satellite network and connection time data are collected using the STK tool.The results show that,compared with the baseline algorithms,the proposed algorithm achieves a higher task success rate and lower task delay.

Key words: Low earth orbit satellite constellation, Satellite network, Deep reinforcement learning, Collaborative computing, Computing offloading

中图分类号: 

  • TP389.1
[1]ZHANG S,CUI G,LONG Y,et al.Joint computing and communication resource allocation for satellite communication networks with edge computing[J].China Communications,2021,18(7):236-252.
[2]SUN J,CHEN X,LI Z,et al.Joint Optimization of Multiple Resources for Distributed Service Deployment in Satellite Edge Computing Networks[J].IEEE Internet of Things Journal,2025,12(3):2359-2372.
[3]SUN Y H,WANG Y,ZHAO L,et al.Review of Digital Twin Based Satellite Network Mobile Edge Computing[J].Computer Science,2024,51(12):12-19.
[4]JIA M,ZHANG L,WU J,et al.Deep Multi-Agent Reinforcement Learning for Task Offloading and Resource Allocation in Satellite Edge Computing[J].IEEE Internet of Things Journal,2025,12(4):3832-3845.
[5]WANG Z X,PENG Q L,SUN R Y,et al.Delay and Energy-aware Task Offloading Approach for Orbit Edge Computing[J].Computer Science,2024,51(S1):754-762.
[6]GEIST A,CRUM G,BREWER C,et al.Nasaspacecube next-generation artificial-intelligence computing for stp-h9-scenic on iss[C]//Small Satellite Conference.2023.
[7]ZHENG H Q,ZHAGN J S,CHEN X.Deployment Optimization and Computing Offloading of Space-Air-Ground Integrated Mobile Edge Computing System[J].Computer Science,2023,50(2):69-79.
[8]YANG J,ZHANG Y,XIAO Z,et al.Joint Access Selection and Computation Offloading in LEO Ubiquitous Edge Computing Networks:An Alternating DRL-Based Approach[J].IEEE Transactions on Cognitive Communications and Networking,2025,11(3):1870-1886.
[9]ZHANG H,ZHAO H,LIU R,et al.Collaborative task offloading optimization for satellite mobile edge computing using multi-agent deep reinforcement learning[J].IEEE Transactions on Vehicular Technology,2024,73(10):15483-15498.
[10]ZHANG H,LIU R,KAUSHIK A,et al.Satellite edge computing with collaborative computation offloading:An intelligent deep deterministic policy gradient approach[J].IEEE Internet of Things Journal,2023,10(10):9092-9107.
[11]CAO H,PENG Y,WANG H,et al.Multi-Satellite Cooperative Computing Task Offloading Strategy Based on Deep Reinforcement Learning[C]//2024 4th International Conference on Computer Communication and Artificial Intelligence(CCAI).IEEE,2024:464-471.
[12]FEI H,ZHANG X,LONG J,et al.Towards multi-satellite collaborative computing via task scheduling based on genetic algorithm[J].Aerospace,2023,10(2):95.
[13]WU H,YANG X,BU Z.Task offloading with service migration for satellite edge computing:A deep reinforcement learning approach[J].IEEE Access,2024,12:25844-25856.
[14]TANG Q,FEI Z,LI B.Distributed deep learning for cooperative computation offloading in low earth orbit satellite networks[J].China Communications,2022,19(4):230-243.
[15]SONG Y,LI X,JI H,et al.Energy-aware task offloading and resource allocation in the intelligent LEO satellite network[C]//2022 IEEE 33rd Annual International Symposium on Personal,Indoor and Mobile Radio Communications(PIMRC).IEEE,2022:481-486.
[16]WANG R,ZHU W,LIU G,et al.Collaborative computation offloading and resource allocation insatellite edge computing[C]//GLOBECOM 2022－2022 IEEE Global Communications Confe-rence.IEEE,2022:5625-5630.
[17]HU Y,GONG W.An on-orbit task-offloading strategy based on satellite edge computing[J].Sensors,2023,23(9):4271.
[18]ZHANG Y,CHEN C,LIU L,et al.Aerial edge computing on orbit:A task offloading and allocation scheme[J].IEEE Tran-sactions on Network Science and Engineering,2022,10(1):275-285.
[19]PENG S,HOU X,SHEN Z,et al.Collaborative satellite computing through adaptivednn task splitting and offloading[C]//2024 IEEE Symposium on Computers and Communications(ISCC).IEEE,2024:1-6.
[20]HE P,HU J,FAN X,et al.Load-balanced collaborative offloa-ding for LEO satellite networks[J].IEEE Internet of Things Journal,2023,10(21):19075-19086.
[21]CHAI F,ZHANG Q,YAO H,et al.Joint multi-task offloading and resource allocation for mobile edge computing systems in satellite IoT[J].IEEE Transactions on Vehicular Technology,2023,72(6):7783-7795.
[22]CHEN J H,KUO W C,LIAO W.SpaceEdge:Optimizing service latency and sustainability for space-centric task offloading in LEO satellite networks[J].IEEE Transactions on Wireless Communications,2024,23(10):15435-15446.
[23]FAN H,YANG Z,ZHANG X,et al.A novel multi-satellite and multi-task scheduling method basedon task network graph aggregation[J].Expert Systems with Applications,2022,205:117565.
[24]QIU L,MENG W,HAN S,et al.Priority-Aware General Packet Offloading in Multi-Layer Dense Satellite Networks[C]//ICC 2024－IEEE International Conference on Communications.IEEE,2024:1849-1854.
[25]OSBAND I,BLUNDELL C,PRITZEL A,et al.Deep exploration via bootstrapped DQN[C]//Advances in Neural Information Processing Systems.2016.
[26]VAN HASSELT H,GUEZ A,SILVER D.Deep reinforcement learning with double q-learning[C]//Proceedings of the AAAI Conference on Artificial Intelligence.2016.
[27]WANG Z,SCHAUL T,HESSEL M,et al.Dueling network architectures for deep reinforcement learning[C]//International Conference on Machine Learning.PMLR,2016:1995-2003.
[28]SUTTON R S,MCALLESTER D,SINGH S,et al.Policy gradient methods for reinforcement learning with function approximation[C]//Advances in Neural Information Processing Systems.1999.
[29]LILLICRAP T P,HUNT J J,PRITZEL A,et al.Continuouscontrol with deep reinforcement learning[J].arXiv:1509.02971,2015.
[30]CHENG C A,XIE T,JIANG N,et al.Adversarially trained actor critic for offline reinforcement learning[C]//International Conference on Machine Learning.PMLR,2022:3852-3878.
[31]SCHULMAN J,LEVINE S,ABBEEL P,et al.Trust region po-licy optimization[C]//International Conference on Machine Learning.PMLR,2015:1889-1897.
[32]SCHULMAN J,WOLSKI F,DHARIWAL P,et al.Proximal policy optimization algorithms[J].arXiv:1707.06347,2017.
[33]MATTHIESEN B,RAZMI N,LEYVA-MAYORGA I,et al.Federated learning in satellite constellations[J].IEEE Network,2024,38(2):232-239.
[34]VALLATR.Pingouin:statistics in Python[J].Journal of Open Source Software,2018,3(31):1026.
Viewed
Full text


Abstract

Cited
  Shared   
  Discussed   
No Suggested Reading articles found!
渝ICP备16013121号-4
地址：重庆市两江新区洪湖西路18号    邮编：401121  
电话：023-63500828（编辑部） 023-67039612（会议/专题） 023-67039623（财务/发票）
E-mail：jsjkx12@163.com
本系统由北京玛格泰克科技发展有限公司设计开发