CAJ | 학술논문

根据网络流量在大时间尺度上的自相似性,以及在小时间尺度上异常流量、Lipschitz正则性与小波变换模极大值三者之间的关系,提出基于小波分析的网络流量异常检测方法.设计了采用该方法检测网络流量异常的模型,解决了方法实现过程中小波选择、模极大值曲线衰减判断、Hurst指数与Lipschitz指数求解等一些关键问题.实验表明,提出的方法能够较好的发现网络流量异常事件并定位异常发生时刻.
근거망락류량재대시간척도상적자상사성,이급재소시간척도상이상류량、Lipschitz정칙성여소파변환모겁대치삼자지간적관계,제출기우소파분석적망락류량이상검측방법.설계료채용해방법검측망락류량이상적모형,해결료방법실현과정중소파선택、모겁대치곡선쇠감판단、Hurst지수여Lipschitz지수구해등일사관건문제.실험표명,제출적방법능구교호적발현망락류량이상사건병정위이상발생시각.
According to the self-similarity of network traffic in large scale, and the relationship among network traffic anomaly in small scale, Lipschitz regularity and wavelet transform modulus maxima, one anomaly detection based on wavelet analysis was pro-posed. Key issues of the choice of wavelet , judgment of the decay of modulus maxima curve, calculation of Hurst and Lipschitz were resolved, and software model for the method was designed. The experimental results show that the proposed method can find out network traffic anomaly in time, and locate the traffic anomaly time well.