›› 2014, Vol. 27 ›› Issue (5): 175-.

• 论文 • 上一篇    下一篇

基于PGM-NMF的网络流量异常检测研究

王晓鸽   

  1. (兰州交通大学 电子与信息工程学院,甘肃 兰州 730070)
  • 出版日期:2014-05-15 发布日期:2014-05-14
  • 作者简介:王晓鸽(1987—),女,硕士研究生。研究方向:数据挖掘,人工智能,网络安全。E-mail:929476235@qq.com

Research on Network Traffic Anomaly Detection Based on PGM-NMF

WANG Xiaoge   

  1. (School of Electronic and Information Engineering,Lanzhou Jiaotong University,Lanzhou 730070,China)
  • Online:2014-05-15 Published:2014-05-14

摘要:

通过对网络流量数据进行采样,小波空间变化过滤噪声,构建了基于信息熵的网络流量矩阵,使用PGM-NMF算法对网络流量矩阵进行分解,构建的基于非负子空间方法的残余矩阵,应用Q 图实现网络流量的异常检测。理论分析及实验结果表明,与PCA方法相比,PGM-NMF算法在网络流量的异常检测中具有较好检测性能。

关键词: PGA NMF算法, 网络流量矩阵, 残余矩阵, 异常检测模型, Q

Abstract:

A network traffic matrix is established based on information entropy by sampling the network traffic data and filtering noise by spatial variability of wavelet.the PGM-NMF algorithm is used for network traffic matrix decomposition to build residual matrix based on nonnegative subspace method.The network traffic anomaly detection is achieved by Q plot.Theoretical and experimental results show that PGM-NMF has a better detection performance than PCA in network traffic anomaly detection.

Key words: PGM NMF algorithms;network traffic matrix;residual matrix;anomaly intrusion detection model;Q plot

中图分类号: 

  • TP301