Application on Network Traffic Prediction Based on Least Squares Support Vector Machine

Abstract:

The network traffic is one of the important metrics for describing network behaviors, it plays an important role in network design, network protocol and traffic project implementation. In order to solve some problems in network traffic prediction, according to actual data for network- monitoring traffic, an approach to network traffic prediction is presented based on least squares support vector machine (LS-SVM), it mainly includes selecting for sample data of network traffic, normalization processing of data, network traffic model trained by LS-SVM and network traffic prediction, etc. Actual application results indicate that the method of network traffic prediction has high accuracy and good feasibility.

• Periodical: Applied Mechanics and Materials (Volumes 20-23)
• Pages: 364-369
• DOI: https://doi.org/10.4028/www.scientific.net/AMM.20-23.364
• Online since: January 2010
• Authors: Yu Zhuo Ren, Ke Wen Xia, Yan Wang, Jun Shi
• Keywords: Least Squares Support Vector Machines (Ls-SVM), Network Traffic, Traffic Prediction

References

[1] GENI: Global environment for network innovations, http: /www. geni. net.

[2] PlanetLab: An open platform for developing, deploying and accessing planetary-scale services. http: /www. planet-lab. org.

[3] Liu Donglin and Shuai Dianyun: Analysis on Network Flow Time Sequences and Extraction of Nonlinear Characteristic Quantities[J]. Acta Electronica Sinica, Vol. 31(2003), pp.56-60.

[4] Li Fangfang, Zhao Yingkai and Yan xi: Matlab Toolbox and Design of Least Squares Support Vector Machine Based on MATLAB [J]. Journal of Computer Applications, Vol. 26(2006), pp.358-360.

[5] Vapnik V: The nature of statistical learning theory[M]. New York: Spring-Verlag, Vol. 15(1995), pp.85-102.

[6] Wang Zhi, Cao Yang, Xu Ning, and Luo Kun: Design of Intelligent Network Analysis and Forecast Support System[J]. Journal of Wuhan University(Natural Science Edition), Vol. 5(2001), pp.75-78.

[7] Liu Yan, Han Liangxiu, Yang Jun and Gao Chuanshan: On the Relationship Between Self-similarity in TCP Traffic and Network Performance [J]. Journal of Chinese Computer Systems, Vol. 13(2004), pp.62-80.

[8] Wang Yasha, Wang Guangxing: Study of a Data Sample Policy in Network Performance Management [J]. Journal of Computer Research and Development, Vol. 39(2002), pp.1031-1037.