网络与信息安全学报2025，Vol.11Issue(6)：120-130,11.DOI:10.11959/j.issn.2096-109x.2025069
基于动态聚类和信誉评估的抗攻击联邦框架
Anti-attack federated framework based on dynamic clustering and reputation evaluation
摘要
Abstract
The practical application of federated learning in scenarios with heterogeneous data faces a serious chal-lenge,which is that poisoning attacks by malicious clients can degrade the global model.A robust federated learn-ing framework DCFL(dynamic clustering federated learning)was proposed based on dynamic clustering and dual-dimensional reputation evaluation.Firstly,DCFL employed the HDBSCAN algorithm for adaptive client clustering without predefined cluster numbers,and redistributed noise points based on similarity measurements,dynamically grouping clients with similar data distribution characteristics,preventing misclassification of heterogeneous but be-nign clients as malicious.Secondly,within clusters,both model update similarity and loss consistency were exam-ined comprehensively as complementary indicators to evaluate client update quality.Finally,a dynamic threshold adjustment strategy was designed,while reputation-weighted intra-cluster aggregation effectively enhances cluster personalization.Experiments validated DCFL's performance advantages across four different data distribution and security environments.Particularly in high-heterogeneity attack scenarios,DCFL improved testing accuracy by ap-proximately 9 percentage points compared to traditional clustered federated learning algorithms,demonstrating its robustness and adaptability in complex federated environments.关键词
联邦学习/动态聚类/数据异构性/恶意攻击/信誉评估Key words
federated learning/dynamic clustering/data heterogeneity/malicious attack/reputation evaluation分类
信息技术与安全科学引用本文复制引用
ZHANG Weiyi,LUO Qirui,ZHANG Lin..基于动态聚类和信誉评估的抗攻击联邦框架[J].网络与信息安全学报,2025,11(6):120-130,11.基金项目
国家自然科学基金(62372247,61872194) （62372247,61872194）
南京邮电大学校级自然科学基金(NY222142) The National Natural Science Foundation of China(62372247,61872194),School-Level Natural Science Foundation of Nanjing University of Posts and Telecommunications(NY222142) （NY222142）
