作者投稿和查稿 主编审稿 专家审稿 编委审稿 远程编辑

计算机工程 ›› 2011, Vol. 37 ›› Issue (9): 187-189. doi: 10.3969/j.issn.1000-3428.2011.09.065

• 人工智能及识别技术 • 上一篇    下一篇

基于Gamma小波模型的网络流量预测

孙 勇,白光伟,赵 露   

  1. (南京工业大学计算机科学与技术系,南京 210009)
  • 出版日期:2011-05-05 发布日期:2011-05-12
  • 作者简介:孙 勇(1985-),男,硕士研究生,主研方向:网络通信流分析与建模;白光伟,教授、博士;赵 露,硕士研究生
  • 基金资助:
    国家自然科学基金资助项目(60673185);教育部留学回国人员科研启动基金资助项目(教外司留[2007]1108号);江苏省“青蓝工程”中青年学术带头人培养对象基金资助项目(苏教师[2007]2号)

Network Traffic Prediction Based on Gamma Wavelet Model

SUN Yong, BAI Guang-wei, ZHAO Lu   

  1. (Dept. of Computer Science and Technology, Nanjing University of Technology, Nanjing 210009, China)
  • Online:2011-05-05 Published:2011-05-12

摘要: 网络流量的精确预测是实现动态流量管理及控制的前提,由此提出一种基于Gamma小波模型的预测方法。将原始数据分解为高频信号和低频信号,采用Gamma小波模型对低频信号进行建模并获取服从Gamma分布的序列,分别对刚获取的序列以及高频信号采用加权一阶局域法进行预测,重构小波以合成数据。通过实验和数学分析的方法,证实该预测模型能够进行网络流量的短期预测。

关键词: 网络流量预测, 流量管理及控制, Gamma小波模型, 局域预测, 短期预测

Abstract: Dynamic traffic management and control are based on the accurate prediction of network traffic. This paper proposes a prediction method based on Gamma wavelet model. Original data series are decomposed by Haar wavelet into a low frequency signal and several high frequency signals. It models the low frequency signal and obtains the new series which obey Gamma distribution. It predicts the new series and the high frequency signals based on the weighted first order local prediction. The respective prediction series are synthesized to get the final prediction data. Experimental results and analytic studies show that the model does perform well in the short-term network traffic prediction.

Key words: network traffic prediction, traffic management and control, Gamma wavelet model, local prediction, short-term prediction

中图分类号: