基于小波变换的网络异常检测研究  

STUDY OF THE NETWORK ABNORMAL DETECTION BASED ON THE WAVELET TRANSFORMS

作　　者：杨继鹏[1] 刘学诚[2] 

机构地区：[1]山东服装职业学院信息工程系,山东泰安271000 [2]泰山学院数学与系统科学学院,泰安271021

出　　处：《山东农业大学学报（自然科学版）》2012年第1期95-99,共5页Journal of Shandong Agricultural University：Natural Science Edition

摘　　要：由于普通的网络流量很难从中检测出异常,为了有效地分析网络流量,深入研究网络流量的性质,本文提出了一种基于小波的分解与重构思想,将网络流量通过小波变换分解成不同尺度下的逼近信号和细节信号,然后分别单支重构成低频序列和高频序列,根据低频序列和高频序列的特性,对异常网络流量进行检测。通过对真实网络流量的仿真实验,结果显示该方法能够比较简单且准确地检测出异常的网络流量。The abnormal network traffic is hardly to be datected from the common network traffic.In order to effectively analyze network traffic,and study deeply the characteristics of network traffic,this paper put forward a decomposition and reconstruction method based on the wavelet transformation and the network traffic will be decomposed into different scales of approximation signal and detail signal by wavelet transformation and the reconstruct them seperate from low frequency and high frequency sequence.The abnormal network traffic will be detected based on the characteristics of the low frequency and high frequency.Through the real network traffic simulation experiment,the result shows that this method is simple and accurate in detecting abnormal network traffic.

关 键 词：小波变换 网络流量 异常检测 

分 类 号：TP393.0[自动化与计算机技术—计算机应用技术]