天地一体化信息网络2025，Vol.6Issue(1)：24-34,11.DOI:10.11959/j.issn.2096-8930.2025004
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
摘要
Abstract
In addressing the issues of data distribution heterogeneity,outdated models,and data privacy and security in distributed train-ing of intelligent models,a federated learning architecture of intelligent models was designed based on blockchain technology and ap-plied to space-based information networks.A secure and efficient training method for intelligent models was proposed based on this ar-chitecture,where a differential privacy noise mechanism,the blockchain technology and a parameter evaluation method were intro-duced 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 broad-casting and block broadcasting.The simulation results indicated that the proposed method enables intelligent models of different struc-tures to converge rapidly,shorten the model training time,and effectively deal with security and privacy threats.关键词
天基信息网络/联邦学习/智能模型/区块链技术Key words
space-based information network/federated learning/intelligent model/blockchain technology分类
信息技术与安全科学引用本文复制引用
栗渊钧,杨德伟,李佳宁,冯笑..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,2025,6(1):24-34,11.基金项目
国家重点研发计划资助项目(No.2022YFB2902703)National Key Research and Development Program of China(No.2022YFB2902703) （No.2022YFB2902703）
