基于网络流量数据的以太网异常状态监测研究  

Research on Ethernet Abnormal State Monitoring Based on Network Traffic Data

作　　者：吴韵怡 WU Yunyi(School of Computer Information Engineering, Guangzhou Huali Science and Technology Vocational College, Guangzhou 511325, China)

机构地区：[1]广州华立科技职业学院,计算机信息工程学院,广东广州511325

出　　处：《微型电脑应用》2022年第2期130-132,共3页Microcomputer Applications

摘　　要：以太网安全是当前人们关注的焦点,而以太网异常状态监测是其中最为关键的技术,当前以太网异常状态监测方法存在误差大,实时性差等缺陷。为了获得更优的以太网异常状态监测结果,提出基于网络流量数据的以太网异常状态监测方法。首先对以太网异常状态监测研究进展进行了调研,找到各种以太网异常状态监测方法存在的缺陷;然后采集以太网流量数据,将以太网异常状态监测看作是二分类问题,通过引入最小二乘支持向量机对以太网流量数据进行建模和分析,识别以太网异常状态,从而实现以太网异常状态监测,最后进行了以太网异常状态监测的仿真测试。测试结果表明,这种方法的以太网异常状态监测精度超过95%,远远高于当前其它方法的以太网异常状态监测精度,可以准确描述以太网状态变化特点。因此提出的方法对以太网异常状态监测效率高,可以对以太网异常状态进行实时辨识,可以有效保护以太网安全。Ethernet security is a focus of current people’s attention,and monitoring Ethernet abnormal state is the most critical technology.The current monitoring methods have large error and poor real-time performance.In order to obtain better monitoring results of Ethernet abnormal state,a method based on network traffic data is proposed.Firstly,the research progress of monitoring Ethernet abnormal state is studied,and the defects of various monitoring methods are found.Then,the Ethernet traffic data are collected,and monitoring the Ethernet abnormal state is regarded as a two classification problem.By introducing the least squares support vector machine to model and analyze the Ethernet traffic data,the Ethernet abnormal state can be identified.At last,the simulation test is carried out.The results show that the monitoring accuracy of this method is more than 95%,which is far higher than that of other methods.It can accurately describe the characteristics of Ethernet state change.The monitoring efficiency of this method is high.Real time identification of Ethernet abnormal state can effectively protect Ethernet security.

关 键 词：以太网 流量数据 模式识别技术 最小二乘支持向量机 监测效率 

分 类 号：TP393[自动化与计算机技术—计算机应用技术]