融合提升小波降噪和LSSVM的网络流量在线预测  

Online prediction of network traffic by integrating lifting wavelet de-noising and LSSVM

作　　者：李明迅[1] 孟相如[1] 袁荣坤[1] 温祥西[1] 陈新富[1] 

机构地区：[1]空军工程大学电讯工程学院,西安710077

出　　处：《计算机应用》2012年第2期340-342,346,共4页journal of Computer Applications

基　　金：陕西省自然科学基金资助项目(SJ08F14;2009JQ8008)

摘　　要：针对网络流量数据被噪声污染而无法进行准确建模与预测的问题,将提升小波降噪(LWD)技术和在线最小二乘支持向量机(LSSVM)相结合,提出了一种网络流量的集成式在线预测方法。该方法首先对采集的流量数据进行降噪,然后采用相空间重构理论计算流量的时延、嵌入维数,据此确定训练样本并建立在线预测模型,对网络流量数据进行预测。实验结果表明,该方法能有效滤除流量噪声,实现在线预测,提高预测精度。Concerning the problem that the network traffic data has been pollfited by noise so that accurate modeling and predicting cannot be achieved, an integrated network traffic online predicting method based on lifting wavelet de-noising and online Least Squares Support Vector Machines （LSSVM） was proposed. First, the Lifting Wavelet De-noising （LWD） was used to pre-process network traffic data, then the phase space reconstruction theory was introduced to calculate the delay time and embedded dimension. On this basis, the training samples were formed and the online LSSVM prediction model was constructed to predict the network traffic. The experimental results show that this prediction model can eliminate the noise effectively and predict the network traffic.

关 键 词：网络流量预测 提升小波降噪 最小二乘支持向量机 在线算法 

分 类 号：TP393.06[自动化与计算机技术—计算机应用技术]