计算技术与自动化2025，Vol.44Issue(3)：164-171,8.DOI:10.16339/j.cnki.jsjsyzdh.202503029
基于攻击特征动态增量聚类的复合连续网络攻击安全检测方法
A Security Detection Method for Composite Continuous Network Attacks Based on Dynamic Incremental Clustering of Attack Features
摘要
Abstract
In composite continuous networks,the large residual value of information sharing is the main reason why the network system cannot accurately detect aggressive behavior.To improve the operational security of composite continuous networks and achieve precise detection of aggressive behavior,a dynamic incremental clustering based attack feature based composite continuous network attack security detection method is designed.Generate a distribution sequence of aggressive data,and complete the extraction of attack features in composite continuous networks by transforming the data sample for-mat.On this basis,dynamic hierarchical partitioning is implemented for composite continuous networks,and the dynamic in-cremental clustering dissimilarity of attack data is calculated in the clustering structure to achieve dynamic incremental cluste-ring processing of attack data for composite continuous networks oriented towards attack features.Set up necessary security attack detection nodes,combine the attack information collected within the nodes,perform security fusion of network attack data,and complete composite continuous network attack security detection.The test results show that this method can con-trol the information sharing residual within 0-1.32,and can detect accurately and efficiently aggressive behavior,which has outstanding application value in improving the operation security of network system.关键词
攻击特征/动态增量聚类/复合连续网络攻击/安全检测/分布序列/格式转化/动态层次/相异度Key words
attack features/dynamic incremental clustering/composite continuous network attacks/safety testing/dis-tribution sequence/format conversion/dynamic hierarchy/degree of dissimilarity分类
信息技术与安全科学引用本文复制引用
王远翔..基于攻击特征动态增量聚类的复合连续网络攻击安全检测方法[J].计算技术与自动化,2025,44(3):164-171,8.基金项目
湖南省自然科学基金资助项目(2019JJ58057) （2019JJ58057）
