User-centric satellite edge computing architecture for task offloading optimization

doi:10.11772/j.issn.1001-9081.2025070842

Journal of Computer Applications ›› 2026, Vol. 46 ›› Issue (8): 2524-2532.DOI: 10.11772/j.issn.1001-9081.2025070842

• Advanced computing • Previous Articles Next Articles

User-centric satellite edge computing architecture for task offloading optimization

Juan CHEN¹(), Yujie CHEN¹, Zongling WU², Di TIAN¹, Jie ZHONG¹

^1.School of Computer and Software Engineering，Xihua University，Chengdu Sichuan 610039，China
^2.School of Information Science and Technology，Southwest Jiaotong University，Chengdu Sichuan 611756，China

Received:2025-07-28 Revised:2025-10-07 Accepted:2025-10-09 Online:2025-11-05 Published:2026-08-10
Contact: Juan CHEN
About author:CHEN Yujie， born in 2000， M. S. candidate. His research interests include edge computing， cloud computing.
WU Zongling， born in 1987， Ph. D.， engineer. His research interests include embedded internet of things.
TIAN Di， born in 2001， M. S. candidate. His research interests include air-space-ground integrated network.
ZHONG Jie， born in 1999， M. S. candidate. His research interests include satellite edge computing， mobile edge computing.
Supported by:
Special Project of Ministry of Education(HZKY20220578);Provincial University Key Laboratory Open Fund(KL2411)

用户为中心的卫星边缘计算架构优化任务卸载

陈娟¹(), 陈玉杰¹, 吴宗玲², 田谛¹, 钟杰¹

^1.西华大学计算机与软件工程学院，成都 610039
^2.西南交通大学信息科学与技术学院，成都 611756

通讯作者: 陈娟
作者简介:陈娟（1986—），女，山西朔州人，讲师，博士，CCF会员，主要研究方向：智能边缘计算、深度强化学习
陈玉杰（2000—），男，四川威远人，硕士研究生，主要研究方向：边缘计算、云计算
吴宗玲（1987—），男，重庆人，工程师，博士，主要研究方向：嵌入式物联网
田谛（2001—），男，四川德阳人，硕士研究生，主要研究方向：空天地一体化网络
钟杰（1999—），男，四川广安人，硕士研究生，主要研究方向：卫星边缘计算、移动边缘计算。
基金资助:
教育部专项(HZKY20220578);省级高校重点实验室开放基金资助项目(KL2411);省级高校重点实验室开放基金资助项目(2024WYJ02)

Abstract

Abstract:

Satellite Edge Computing （SEC） improves computational efficiency for remote users by offloading tasks to satellite servers. However， SEC system performance may be affected by wireless transmission quality， since there is signal interference among User Equipments （UEs）. To this end， we proposed a User-Centric SEC （UCSEC） architecture to allocate Access Point （AP） clusters dynamically according to the real-time location of users and network conditions. To handle dynamic task offloading in UCSEC， we designed a Master-Slave multi-agent Proximal Policy Optimization with Gated Recurrent Unit （GRU） and Multi-head Attention mechanism （MSPPO_GMA） algorithm. The algorithm adopted a centralized control-distributed execution framework. The master agent， deployed on the SEC server， monitored global resource constraints and coordinated slave agents， while slave agents deployed on the UE executed task offloading and resource allocation according to network states. In MSPPO_GMA algorithm， GRU captured temporal correlations of network states， while multi-head attention focused on critical information of task offloading and resource allocation. Experimental results demonstrate that compared with MAPPO， MATD3 and other algorithms， MSPPO_GMA algorithm reduces task completion latency by 3.7% to 13.6% and UE energy consumption by 6.5% to 13.2%.

Key words: Satellite Edge Computing (SEC), multi-agent, deep reinforcement learning, task offloading, resource allocation

摘要：

卫星边缘计算（SEC）通过将任务卸载至卫星服务器提高远程用户的计算效率；然而，由于用户设备（UE）间的信号干扰，无线传输质量可能会影响SEC系统的性能，因此，提出一种以用户为中心的SEC（UCSEC）架构，根据用户实时位置和网络状况动态分配接入点（AP）集群。针对UCSEC中的动态任务卸载问题，设计一种基于门控循环单元（GRU）与多头注意力机制的主从式多智能体近端策略优化（MSPPO_GMA）算法。该算法采用集中控制-分布执行框架，即部署在SEC服务器上的主智能体负责监控全局资源约束并协调从智能体，而部署在UE上的从智能体根据网络状态执行任务卸载和资源分配。在MSPPO_GMA算法中，GRU捕捉网络状态的时间相关性，而多头注意力机制关注任务卸载和资源分配的重要信息。实验结果表明，与MAPPO和MATD3等算法相比，MSPPO_GMA算法的任务完成延迟降低了3.7%~13.6%，UE能耗降低了6.5%~13.2%。

关键词: 卫星边缘计算, 多智能体, 深度强化学习, 任务卸载, 资源分配

CLC Number:

TP393.09

Juan CHEN, Yujie CHEN, Zongling WU, Di TIAN, Jie ZHONG. User-centric satellite edge computing architecture for task offloading optimization[J]. Journal of Computer Applications, 2026, 46(8): 2524-2532.

陈娟, 陈玉杰, 吴宗玲, 田谛, 钟杰. 用户为中心的卫星边缘计算架构优化任务卸载[J]. 《计算机应用》唯一官方网站, 2026, 46(8): 2524-2532.

Figures/Tables 10

Tab. 1 Symbol summary

符号	描述
$N$	UE数
$M$	AP数
$C$	AP集群大小
$ρ$	AP天线数
$n$	一个UE或任务
$D n$	任务 $n$ 数据量
$τ n$	处理任务 $n$ 截止时间
$x n (1)$	任务 $n$ 卸载至本地或服务器二进制决策
$x n (2)$	任务 $n$ 卸载至SEC或云服务器二进制决策
$T n$	处理任务 $n$ 的任务完成总延迟，单位s
$E n$	处理任务 $n$ 的总能耗，单位J
$p n$	UE $n$ 的传输功率
$c n$	处理每比特任务 $n$ 的CPU周期数
$f n$	UE $n$ 的计算资源能力
$f e$	SEC服务器的计算资源能力
$f c$	云服务器的计算资源能力
$K$	SEC服务器通道数
$D e$	SEC服务器最大资源容量
$l s$	UE与卫星之间的距离，单位km
$l c$	卫星与云中心之间的距离，单位km

Tab. 1 Symbol summary

符号	描述
$N$	UE数
$M$	AP数
$C$	AP集群大小
$ρ$	AP天线数
$n$	一个UE或任务
$D n$	任务 $n$ 数据量
$τ n$	处理任务 $n$ 截止时间
$x n (1)$	任务 $n$ 卸载至本地或服务器二进制决策
$x n (2)$	任务 $n$ 卸载至SEC或云服务器二进制决策
$T n$	处理任务 $n$ 的任务完成总延迟，单位s
$E n$	处理任务 $n$ 的总能耗，单位J
$p n$	UE $n$ 的传输功率
$c n$	处理每比特任务 $n$ 的CPU周期数
$f n$	UE $n$ 的计算资源能力
$f e$	SEC服务器的计算资源能力
$f c$	云服务器的计算资源能力
$K$	SEC服务器通道数
$D e$	SEC服务器最大资源容量
$l s$	UE与卫星之间的距离，单位km
$l c$	卫星与云中心之间的距离，单位km

Fig. 1 Architecture of UCSEC system

Fig. 2 Architecture of MSPPO_GMA algorithm

Tab. 2 Experimental parameter setting

参数	值	参数	值
$N$	50	$δ$	［0，1］
$M$	50	$d 0$	10 m
$C$	5	$d 1$	50 m
$B a$	20 MHz	$p n p$	24 dBm
$B f$	40 MHz	$h m$	15 m
$f n$	［0.4，1.0］ GHz	$h n$	1.5 m
$G m$	［5，14］ dB	$f$	1.9 GHz
$D n$	［1，50］ MB	$N a$	-174 dBm/Hz
$c n$	［300，737.5］ cycles	$ρ$	16
$p n$	［1，24］ dBm	$σ s$	8 dB
$p m f$	30 dBm	$K$	10
$f e$	4 GHz	$μ m n$	0.5
$f c$	10 GHz	$κ$	$5 × 10 - 27$
$D e$	400 MB	$α$	0.5
$b n$	40 MJ	$μ$	0.99
$τ n$	［0.1，1］ s	$β$	0.5
$l s$	1 000 km^［20］	$i$	8
$l c$	2 000 km^［20］	$B$	64

Tab. 2 Experimental parameter setting

参数	值	参数	值
$N$	50	$δ$	［0，1］
$M$	50	$d 0$	10 m
$C$	5	$d 1$	50 m
$B a$	20 MHz	$p n p$	24 dBm
$B f$	40 MHz	$h m$	15 m
$f n$	［0.4，1.0］ GHz	$h n$	1.5 m
$G m$	［5，14］ dB	$f$	1.9 GHz
$D n$	［1，50］ MB	$N a$	-174 dBm/Hz
$c n$	［300，737.5］ cycles	$ρ$	16
$p n$	［1，24］ dBm	$σ s$	8 dB
$p m f$	30 dBm	$K$	10
$f e$	4 GHz	$μ m n$	0.5
$f c$	10 GHz	$κ$	$5 × 10 - 27$
$D e$	400 MB	$α$	0.5
$b n$	40 MJ	$μ$	0.99
$τ n$	［0.1，1］ s	$β$	0.5
$l s$	1 000 km^［20］	$i$	8
$l c$	2 000 km^［20］	$B$	64

Fig. 3 Comparison of reward convergence with different learning rates

Fig. 4 Average task completion latency and energy consumption with different weighting factors

Fig. 5 Comparison of reward convergence of different algorithms

Fig. 6 Performance comparison of various algorithms under different UE numbers

Fig. 7 Performance comparison of various algorithms under different task data sizes

Tab. 3 Performance under different ablation algorithms

算法	平均任务完成延迟/s	平均能耗/J	平均成本
MAPPO	1.805	1.002	2.088
MSPPO_MA	1.783	0.840	1.953
MSPPO_GRU	1.803	0.883	1.954
MSPPO_GMA	1.739	0.712	1.652

References 25

[1]	施建锋，陈忻阳，李宝龙. 面向物联网的云边端协同计算中任务卸载与资源分配算法研究［J］. 电子与信息学报， 2025， 47（2）： 458-469.
	Shi Jianfeng， Chen Xinyang， Li Baolong. Research on task offloading and resource allocation algorithm in cloud-edge-end collaborative computing for internet of things［J］. Journal of Electronics and Information Technology， 2025， 47（2）： 458-469.
[2]	Waqar N， Hassan S A， Mahmood A， et al. Computation offloading and resource allocation in MEC-enabled integrated aerial terrestrial vehicular networks： a reinforcement learning approach［J］. IEEE Transactions on Intelligent Transportation Systems， 2022， 23（11）： 21478-21491.
[3]	Ammar H A， Adve R， Shahbazpanahi S， et al. User-centric cell-free massive MIMO networks： a survey of opportunities， challenges and solutions［J］. IEEE Communications Surveys and Tutorials， 2022， 24（1）： 611-652.
[4]	Ammar H A， Adve R， Shahbazpanahi S， et al. Handoffs in user-centric cell-free MIMO networks： a POMDP framework［J］. IEEE Transactions on Wireless Communications， 2024， 23（8）： 10319-10335.
[5]	宋传龙，张倩武，何健，等. 基于MADDPG算法的星地协同边缘计算任务卸载方法［J］. 系统工程与电子技术， 2026， 48（1）： 350-360.
	Song Chuanlong， Zhang Qianwu， He Jian， et al. Edge computing task offloading method of satellite-ground collaborative based on MADDPG algorithm［J］. Systems Engineering and Electronics， 2026， 48（1）： 350-360.
[6]	Tang Q， Fei Z， Li B， et al. Stochastic computation offloading for LEO satellite edge computing networks： a learning-based approach［J］. IEEE Internet of Things Journal， 2024， 11（4）： 5638-5652.
[7]	Zhang R， Zhao B. Task offloading and resource allocation for cloud-edge collaboration in low earth orbit satellite networks［C］// ICCT 2023. Piscataway： IEEE， 2023： 764-769.
[8]	Zhang H， Su R， Zhu Y， et al. User-centric cell-free massive MIMO system for indoor industrial networks［J］. IEEE Transactions on Communications， 2022， 70（11）： 7644-7655.
[9]	Luo J， Fan J， Zhang J. Distributed hybrid precoding for edge computing-assisted cell-free massive MIMO systems with local CSI［J］. IEEE Internet of Things Journal， 2024， 11（6）： 10880-10892.
[10]	Fang H， Jia Y， Wang Y， et al. Matching game based task offloading and resource allocation algorithm for satellite edge computing networks［C］// ISNCC 2022. Piscataway： IEEE， 2022： 1-5.
[11]	Wang H， Han J， Cao S， et al. Computation offloading strategy of multi-satellite cooperative tasks based on genetic algorithm in satellite edge computing［C］// SAGC 2021. Piscataway： IEEE， 2021： 22-28.
[12]	赵璐，赵宇川，黄安旭，等. 卫星边缘计算网络中基于深度强化学习的依赖性任务卸载方法［J］. 小型微型计算机系统， 2025， 46（11）： 2747-2754.
	Zhao Lu， Zhao Yuchuan， Huang Anxu， et al. Dependent task offloading approach via deep reinforcement learning in satellite edge computing networks［J］. Journal of Chinese Computer Systems， 2025， 46（11）： 2747-2754.
[13]	Chen G， Qi S， Shen F， et al. Information-aware driven dynamic LEO-RAN slicing algorithm joint with communication， computing， and caching［J］. IEEE Journal on Selected Areas in Communications， 2024， 42（5）： 1044-1062.
[14]	Qin Z， Yao H， Mai T， et al. Multi-agent reinforcement learning aided computation offloading in aerial computing for the internet-of-things［J］. IEEE Transactions on Services Computing， 2023， 16（3）： 1976-1986.
[15]	Gao X， Liu R， Kaushik A， et al. Dynamic resource allocation for virtual network function placement in satellite edge clouds［J］. IEEE Transactions on Network Science and Engineering， 2022， 9（4）： 2252-2265.
[16]	Ngo H Q， Ashikhmin A， Yang H， et al. Cell-free massive MIMO versus small cells［J］. IEEE Transactions on Wireless Communications， 2017， 16（3）： 1834-1850.
[17]	Buzzi S， D’Andrea C， Zappone A， et al. User-centric 5G cellular networks： resource allocation and comparison with the cell-free massive MIMO approach［J］. IEEE Transactions on Wireless Communications， 2020， 19（2）： 1250-1264.
[18]	Qin L， Lu H， Chen Y， et al. Toward decentralized task offloading and resource allocation in user-centric MEC［J］. IEEE Transactions on Mobile Computing， 2024， 23（12）： 11807-11823.
[19]	Guenach M， Gorji A A， BourdouxO A. Joint power control and access point scheduling in fronthaul-constrained uplink cell-free massive MIMO systems［J］. IEEE Transactions on Communications， 2021， 69（4）： 2709-2722.
[20]	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.
[21]	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.
[22]	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.
[23]	Kang H， Chang X， Mišić J， et al. Cooperative UAV resource allocation and task offloading in hierarchical aerial computing systems： a MAPPO-based approach［J］. IEEE Internet of Things Journal， 2023， 10（12）： 10497-10509.
[24]	Qin P， Li H， Fu Y， et al. Learning-based NOMA-enabled queue-aware task offloading and AAV 3D trajectory planning for SAGIN［J］. IEEE Transactions on Vehicular Technology， 2025， 74（8）： 12364-12375.
[25]	Prabhashana S C， van Huynh D， Singh K， et al. Machine learning-based resource allocation in 6G integrated space and terrestrial networks-aided intelligent autonomous transportation［J］. IEEE Transactions on Intelligent Transportation Systems， 2025， 26（10）： 17750-17762.

User-centric satellite edge computing architecture for task offloading optimization

用户为中心的卫星边缘计算架构优化任务卸载

RichHTML

PDF

Knowledge

Abstract

Cite this article

share this article

Figures/Tables 10

References 25

Related Articles 15

Recommended Articles

Metrics

[1]	Yunle WANG, Xiang FENG, Huiqun YU. VLMDs-Privacy： privacy-enhanced strategy for cooperative decision-making in socially-aware multi-agent systems [J]. Journal of Computer Applications, 2026, 46(5): 1518-1525.
[2]	Xiayu WU, Hong ZHANG. Review of evolution and changes in crowd evacuation calculation methods [J]. Journal of Computer Applications, 2026, 46(4): 1309-1322.
[3]	Shuai SU, Chenglin LIU. Observer-based leader-following consensus of heterogeneous descriptor multi-agent system with disturbance [J]. Journal of Computer Applications, 2026, 46(4): 1211-1217.
[4]	Tianyu XUE, Aiping LI, Liguo DUAN. Vehicular edge computing scheme with task offloading and resource optimization [J]. Journal of Computer Applications, 2025, 45(6): 1766-1775.
[5]	Hongwei FAN, Woping XU. Resource allocation for relay in intelligent reflecting surface assisted wireless powered communication networks [J]. Journal of Computer Applications, 2025, 45(5): 1619-1624.
[6]	Pengcheng XU, Lei HE, Chuan LI, Weiqi QIAN, Tun ZHAO. Deep symbolic regression method based on Transformer [J]. Journal of Computer Applications, 2025, 45(5): 1455-1463.
[7]	Yufei XIANG, Zhengwei NI. Edge federation dynamic analysis for hierarchical federated learning based on evolutionary game [J]. Journal of Computer Applications, 2025, 45(4): 1077-1085.
[8]	Huahua WANG, Liang HUANG, Jiajie CHEN, Jiening FANG. Dynamic allocation algorithm for multi-beam subcarriers of low orbit satellites based on deep reinforcement learning [J]. Journal of Computer Applications, 2025, 45(2): 571-577.
[9]	Jing WANG, Xuming FANG. Intelligent joint power and channel allocation algorithm for Wi-Fi7 multi-link integrated communication and sensing [J]. Journal of Computer Applications, 2025, 45(2): 563-570.
[10]	Jun ZENG, Yinghua TONG, Defang WANG. Anomaly detection method based on cumulative probability fluctuation and automated clustering [J]. Journal of Computer Applications, 2025, 45(12): 3864-3871.
[11]	Lin WEI, Shihao ZHANG, Mengyang HE. Workflow task optimization and energy-efficient offloading method for computing power network [J]. Journal of Computer Applications, 2025, 45(12): 3916-3924.
[12]	Lin WEI, Jinyang LI, Yajie WANG, Mengyang HE. Highly reliable matching method based on multi-dimensional resource measurement and rescheduling in computing power network [J]. Journal of Computer Applications, 2025, 45(11): 3632-3641.
[13]	Shuai ZHOU, Hao FU, Wei LIU. Spatial-temporal Transformer-based hybrid return implicit Q-learning for crowd navigation [J]. Journal of Computer Applications, 2025, 45(11): 3666-3673.
[14]	Yanpeng ZHANG, Yuqian ZHAO, Fan ZHANG, Tenghai QIU, Gui GUI, Lingli YU. Capacitated vehicle routing problem solving method based on improved MAML and GVAE [J]. Journal of Computer Applications, 2025, 45(11): 3642-3648.
[15]	Jian SUN, Wei ZHANG, Baoquan MA, Zhuiwei WU, Xiaohuan YANG, Tao WU. MATOS： UAV swarm assisted moving-aware adaptive-parallel computing task offloading system [J]. Journal of Computer Applications, 2025, 45(10): 3259-3269.