面向攻击路径的网络安全分析  

Network security analysis for attack paths

作　　者：杨龙[1] YANG Long(BAOJI University of arts and sciences,Baoji,Shaanxi 721013,China)

机构地区：[1]宝鸡文理学院,陕西宝鸡721013

出　　处：《自动化与仪器仪表》2022年第7期170-174,共5页Automation & Instrumentation

基　　金：《基于数据挖据技术的校园网环境下大学生访问网络不良信息的行为分析与对策研究》(ZK2017071)。

摘　　要：为提高网络的安全性,提出一种基于改进蚁群算法的攻击路径威胁分析方法。在该方法中,首先利用Text CNN生成网络攻击图,然后在定义代价定量计算公式的基础上,利用改进蚁群算法对攻击路径进行寻优,并计算最小攻击代价和潜在攻击目标价值,最终得出网络的受威胁的程度。结果表明,采用Text CNN分类器对CEV漏洞的分类精确率和召回率达80.3%和79.6%,攻击图生成方法完整性达98%,且生成时间较短;在不同场景下的对比发现,采用提出的蚁群改进的攻击路径威胁度量方法在场景a中的评分更高,说明场景a网络受到的网络攻击威胁更大。通过以上研究,为网络安全威胁的定量分析提供了一种参考和借鉴。In order to improve the security of network,an attack path threat analysis method based on improved ant colony algorithm is proposed.In this method,firstly,text CNN is used to generate the network attack graph,then on the basis of defining the cost quantitative calculation formula,the improved ant colony algorithm is used to optimize the attack path,calculate the minimum attack cost and the value of potential attack targets,and finally the degree of threat of the network is get.The results show that the classification accuracy and recall rate of CEV vulnerabilities by text CNN classifier are 80.3% and 79.6%,the integrity of attack graph generation method is 98%,and the generation time is short;The comparison in different scenarios shows that the proposed ant colony improved attack path threat measurement method has a higher score in scenario a,indicating that the network in scenario a is more threatened by network attacks.Through the above research,it provides a reference for the quantitative analysis of network security threats.

关 键 词：攻击路径 网络安全 Text CNN 蚁群算法 信息素 

分 类 号：TP392[自动化与计算机技术—计算机应用技术]