面向天基信息网络的智能模型分布式训练技术

栗渊钧; 杨德伟; 李佳宁; 冯笑

doi:10.11959/j.issn.2096-8930.2025004

您当前的位置：

首页 >

文章列表页 >

面向天基信息网络的智能模型分布式训练技术

研究 | 更新时间：2025-04-10

- 面向天基信息网络的智能模型分布式训练技术
- Distributed Training Techniques for Intelligent Model in Space-Based Information Networks
- 天地一体化信息网络 2025年6卷第1期页码：24-34
- 作者机构：
  
  北京理工大学信息与电子学院，北京 100081
  2.中国电子科技集团公司第十五研究所，北京 100083
- 作者简介：
  
  [ "栗渊钧（1999- ），男，北京理工大学信息与电子学院硕士生，主要研究方向为边缘智能计算。" ]
  [ "杨德伟（1979- ），男，博士，北京理工大学信息与电子学院副教授，主要研究方向为卫星通信、边缘智能计算。" ]
  [ "李佳宁（2000- ），女，北京理工大学信息与电子学院硕士生，主要研究方向为边缘智能计算。" ]
  [ "冯笑（1989- ），男，硕士，中国电子科技集团公司第十五研究所高级工程师，主要研究方向为计算机体系架构、边缘智能计算。" ]
- 基金信息：
  
  国家重点研发计划资助项目(2022YFB2902703)
- DOI：10.11959/j.issn.2096-8930.2025004
  中图分类号： TP393
- 收稿：2024-11-15，
  
  修回：2025-02-10，
  
  纸质出版：2025-03-20
- 稿件说明：
移动端阅览
栗渊钧,杨德伟,李佳宁等.面向天基信息网络的智能模型分布式训练技术[J].天地一体化信息网络,2025,06(01):24-34.

LI Yuanjun,YANG Dewei,LI Jianing,et al.Distributed Training Techniques for Intelligent Model in Space-Based Information Networks[J].Space-Integrated-Ground Information Networks,2025,06(01):24-34.
栗渊钧,杨德伟,李佳宁等.面向天基信息网络的智能模型分布式训练技术[J].天地一体化信息网络,2025,06(01):24-34. DOI： 10.11959/j.issn.2096-8930.2025004.

LI Yuanjun,YANG Dewei,LI Jianing,et al.Distributed Training Techniques for Intelligent Model in Space-Based Information Networks[J].Space-Integrated-Ground Information Networks,2025,06(01):24-34. DOI： 10.11959/j.issn.2096-8930.2025004.

摘要

针对天基信息网络中智能模型的分布式训练存在数据分布异构、模型陈旧以及隐私安全等问题，提出基于区块链的智能模型联邦学习架构和安全高效训练方法，引入差分隐私噪声机制和参数评估方法，有效应对隐私泄露、中毒攻击和单点故障威胁；采用基于时延最小的模型聚合方法，通过轨道内外的模型广播及区块广播过程，加速模型训练。仿真结果表明，所提方法能使不同结构的智能模型快速收敛，缩短训练时间，并有效应对安全隐私威胁。

Abstract

In addressing the issues of data distribution heterogeneity

outdated models

and data privacy and security in distributed training of intelligent models

a federated learning architecture of intelligent models was designed based on blockchain technology and applied to space-based information networks. A secure and efficient training method for intelligent models was proposed based on this architecture

where a differential privacy noise mechanism

the blockchain technology and a parameter evaluation method were introduced to effectively deal with privacy leakage

poisoning attacks and single-point failure threats. Meanwhile

using a model aggregation method based on the minimized delay

the model training was accelerated via the processes of intra-orbit and inter-orbit model broadcasting and block broadcasting. The simulation results indicated that the proposed method enables intelligent models of different structures to converge rapidly

shorten the model training time

and effectively deal with security and privacy threats.

关键词

Keywords

references

CENTENARO M , COSTA C E , GRANELLI F , et al . A survey on technologies, standards and open challenges in satellite IoT [J ] . IEEE Communications Surveys & Tutorials , 2021 , 23 ( 3 ): 1693 - 1720 .

ZHANG J X , ZHANG X , WANG P , et al . Double-edge intelligent integrated satellite terrestrial networks [J ] . China Communications , 2020 , 17 ( 9 ): 128 - 146 .

KONE VCNÝ J , BRENDAN MCMAHAN H , YU F X , et al . Federated learning: strategies for improving communication efficiency [J ] . ArXiv e-Prints , 2016 : arXiv: 1610.05492 .

MCMAHAN B , MOORE E , RAMAGE D , et al . Communication-efficient learning of deep networks from decentralized data [C ] // Proceedings of the 20th International Conference on Artificial Intelligence and Statistics . [ S.l. : s.n. ] , 2017 : 1273 - 1282 .

CHEN H , XIAO M , PANG Z B . Satellite-based computing networks with federated learning [J ] . IEEE Wireless Communications , 2022 , 29 ( 1 ): 78 - 84 .

SALIM S , MOUSTAFA N , HASSANIAN M , et al . Deep-federated-learning-based threat detection model for extreme satellite communications [J ] . IEEE Internet of Things Journal , 2024 , 11 ( 3 ): 3853 - 3867 .

MOUSTAFA N , KHAN I A , HASSANIN M , et al . DFSat: deep federated learning for identifying cyber threats in IoT-based satellite networks [J ] . IEEE Transactions on Industrial Informatics , 2022 ( 99 ): 1 - 8 .

LI K , ZHOU H C , TU Z , et al . Distributed network intrusion detection system in satellite-terrestrial integrated networks using federated learning [J ] . IEEE Access , 2020 , 8 : 214852 - 214865 .

RAZMI N , MATTHIESEN B , DEKORSY A , et al . Ground-assisted federated learning in LEO satellite constellations [J ] . IEEE Wireless Communications Letters , 2022 , 11 ( 4 ): 717 - 721 .

RAZMI N , MATTHIESEN B , DEKORSY A , et al . On-board federated learning for dense LEO constellations [C ] // Proceedings of the ICC 2022 - IEEE International Conference on Communications . Piscataway : IEEE Press , 2022 : 4715 - 4720 .

RAZMI N , MATTHIESEN B , DEKORSY A , et al . Scheduling for ground-assisted federated learning in LEO satellite constellations [C ] // Proceedings of the 2022 30th European Signal Processing Conference (EUSIPCO) . Piscataway : IEEE Press , 2022 : 1102 - 1106 .

FREDRIKSON M , JHA S , RISTENPART T . Model inversion attacks that exploit confidence information and basic countermeasures [C ] // Proceedings of the 22nd ACM SIGSAC Conference on Computer and Communications Security . New York : ACM , 2015 : 1322 - 1333 .

YANG Z Q , ZHANG J Y , CHANG E C , et al . Neural network inversion in adversarial setting via background knowledge alignment [C ] // Proceedings of the 2019 ACM SIGSAC Conference on Computer and Communications Security . New York : ACM , 2019 : 225 - 240 .

FREDRIKSON M , LANTZ E , JHA S , et al . Privacy in pharmacogenetics: an end-to-end case study of personalized warfarin dosing [J ] . Proceedings of the USENIX Security Symposium UNIX Security Symposium , 2014 : 17 - 32 .

LU Y L , HUANG X H , ZHANG K , et al . Blockchain and federated learning for 5G beyond [J ] . IEEE Network , 2021 , 35 ( 1 ): 219 - 225 .

POKHREL S R . Blockchain brings trust to collaborative drones and LEO satellites: an intelligent decentralized learning in the space [J ] . IEEE Sensors Journal , 2021 , 21 ( 22 ): 25331 - 25339 .

QU Y Y , UDDIN M P , GAN C Q , et al . Blockchain-enabled federated learning: a survey [J ] . ACM Computing Surveys , 2022 , 55 ( 4 ): 1 - 35 .

NGUYEN D C , DING M , PHAM Q V , et al . Federated learning meets blockchain in edge computing: opportunities and challenges [J ] . IEEE Internet of Things Journal , 2021 , 8 ( 16 ): 12806 - 12825 .

DENG L . The MNIST database of handwritten digit images for machine learning research best of the web [J ] . IEEE Signal Processing Magazine , 2012 , 29 ( 6 ): 141 - 142 .

LECUN Y , BOTTOU L , BENGIO Y , et al . Gradient-based learning applied to document recognition [J ] . Proceedings of the IEEE , 1998 , 86 ( 11 ): 2278 - 2324 .

KRIZHEVSKY A , SUTSKEVER I , HINTON G E . ImageNet classification with deep convolutional neural networks [J ] . Communications of the ACM , 2017 , 60 ( 6 ): 84 - 90 .

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

基于移动边缘计算的空天地一体化网络架构

天基信息光网络波长路由关键技术