首页 | 本学科首页   官方微博 | 高级检索  
     检索      

基于经验模态分解的局域网络入侵检测算法
引用本文:李响.基于经验模态分解的局域网络入侵检测算法[J].西南师范大学学报(自然科学版),2016,41(8).
作者姓名:李响
作者单位:重庆电子工程职业学院,重庆,401331
摘    要:通过对局域网络入侵的准确检测可以保障网络安全,由于局域网网络入侵信号具有瞬时频率特性,采用传统的时频分析方法难以实现有效检测,出现检测不准确的问题.为此提出基于经验模态分解的局域网络入侵检测算法,分析网络攻击的防护原理和DOS攻击对网络的危害,对信号处理方法进行检测方法设计.卡尔曼滤波方法对DOS入侵信号进行前置滤波,去除入侵信号的EMD虚假分量,采用小波阈值去噪方法进行信号提纯,采用经验模特分解方法,使得DOS入侵信号特征与干扰组成成分最佳匹配,提取HHT频谱实现对入侵信号的准确检测.仿真结果表明,采用该检测方法进行局域网入侵检测,精度较高,抗干扰性强,检测性能优于传统算法.

关 键 词:局域网  入侵检测  网络安全  经验模态分解

Local Network Intrusion Detection Algorithm Based on Empirical Mode Decomposition
LI Xiang.Local Network Intrusion Detection Algorithm Based on Empirical Mode Decomposition[J].Journal of Southwest China Normal University(Natural Science),2016,41(8).
Authors:LI Xiang
Abstract:Based on local area network,intrusion of accurate detection can guarantee the network security, and due to the LAN network intrusion signal instantaneous frequency characteristics,the traditional time-frequency analysis method is difficult to achieve effective detection,detection inaccurate problem.Based on empirical mode decomposition of local area network intrusion detection algorithm.Analysis has been done of the protection principle of the network attack and DOS attacks the harm of the network.Detection method based on signal processing method is designed,using the kalman filtering method to front DOS in-vading signal filtering,and to remove the intrusion signal EMD false component;wavelet threshold denois-ing method has been adopted to improve the signal purification,using empirical model decomposition meth-od,and to makes the DOS composition best match intrusion signal characteristics and interference,extrac-tion of HHT spectrum to achieve accurate detection of intrusion signal.The simulation results show that the detection method of network intrusion detection,high precision,strong anti-interference,and detec-tion performance is better than traditional algorithm.
Keywords:
本文献已被 CNKI 等数据库收录!
点击此处可从《西南师范大学学报(自然科学版)》浏览原始摘要信息
点击此处可从《西南师范大学学报(自然科学版)》下载免费的PDF全文
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号