|NeSC Bibliographic Database|
A novel visual discriminator on network traffic pattern
Han,L van Hemert,J.I
Appeared in: The Second International Conference on Advanced Engineering Computing and Applications in Sciences (ADVCOMP 2008) Valencia Spain
Publisher: IEEE Computer Society Press
Field of Science: e-Science
Abstract: The wavelet transform has been shown to be a powerful tool for characterising network traffic. However, the resulting decomposition of a wavelet transform typically forms a high-dimension space. This is obviously problematic on compact representations, visualizations, and modelling approaches that are based on these high-dimensional data. In this study, we show how data projection techniques can represent the high-dimensional wavelet decomposition in a low dimensional space to facilitate visual analysis. A low-dimensional representation can significantly reduce the model complexity. Hence, features in the data can be presented with a small number of parameters. We demonstrate these projections in the context of network traffic pattern analysis. The experimental results show that the proposed method can effectively discriminate between different application flows, such as FTP and P2P.
|Last Updated: 22 Jun 12 11:02|