基于改进差分算法的网络异常攻击流量入侵识别方法  

Intrusion identification method of network abnormal attack traffic based on improved differential algorithm

作　　者：赖振辉 甘汉波 LAI Zhen-hui;GAN Han-bo(Cenxi Secondary Vocational School,Wuzhou 543200,Guangxi Zhuang Autonomous Region,China)

机构地区：[1]岑溪市中等专业学校,广西梧州543200

出　　处：《信息技术》2025年第3期122-127,共6页Information Technology

基　　金：2021年度广西职业教育教学改革研究项目(GXZZJG-2021A073)。

摘　　要：目前方法对异常攻击流量的入侵很难精准识别,文中提出基于改进差分算法的网络异常攻击流量入侵识别方法。根据网络流量数据属性与攻击特征对应关系,直观呈现网络动态变化。构建字符串长度识别模型、二级域名可信性过滤模型、IP离散性验证模型以及异常入侵数据阻断模型,利用改进差分进化适应度函数对正常访问流量与异常攻击流量进行数据划分,由此实现入侵识别。经实验验证,所提识别算法各项评估指标均高于其他算法,准确率提高10%左右,具有较强的入侵识别能力。It is difficult to accurately identify the intrusion of abnormal attack traffic by current methods.An intrusion identification method of network abnormal attack traffic based on improved differential algorithm is proposed.Based on the relationship between network traffic data attributes and attack characteristics,the dynamic changes of the network are visually displayed.The string length recognition model,the second-level domain name credibility filtering model,the IP discreteness verification model and the abnormal intrusion data blocking model are constructed.The improved differential evolution fitness function is used to divide the normal access traffic and the abnormal attack traffic,so as to achieve intrusion identification.The experiment results show that the proposed recognition algorithm has higher evaluation indexes than other algorithms,and the accuracy rate is increased by about 10%,so it has strong ability of intrusion identification.

关 键 词：改进差分算法 网络攻击 持续性变量 异常入侵流量 入侵识别 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]