摘要
本文提出一种基于小波变换与自回归模型的网络流量预测方法,将流量数据构成的原始序列进行小波分解,并将分解得到的近似部分和各细节部分分别单支重构到原级别上;对各个重构后的序列建立自回归模型,由所拟合的模型分别进行预测;结合各个重构后序列的预测结果,可以得到对原始序列的预测结果。实验结果表明,这种方法比传统的几种网络流量预测方法具有更高的预测准确度。
Network traffic prediction based on wavelet transform and autoregressive model is proposed. The original discrete series consisting of network traffic data is decomposed into approximate series and several detail series. The result of single branch reconstruction of each decomposed series is more unitary than the original series in frequency, and it can be built traffic model with autoregressive model. The prediction of the original series can be obtained by the synthesis of each reconstructed series prediction result. As shown in a set of experiments, the novel method is of higher accuracy in comparison with the traditional ones.
出处
《计算机科学》
CSCD
北大核心
2007年第7期47-49,54,共4页
Computer Science
基金
国家自然科学基金资助项目(60273021)
关键词
流量预测
小波变换
MALLAT算法
自回归模型
Traffic prediction, Wavelet transform, Mallat algorithm, Autoregressive model