摘要： 提出一种新的网络流量预测方法。该方法根据网络流量历史值用灰色模型GM(1,1)进行预测，并用自适应过滤法对GM(1,1)预测时产生的残差进行修正，从而达到较高的预测精度。该方法综合了GM(1,1)预测所需原始数据少、方法简单等特点，具有较高的应用价值。实验结果表明，该方法是有效可行的。

关键词: 灰色模型, 预测, 残差, 自适应过滤

Abstract: A new network traffic prediction method is put forward. This method uses Grey Model(GM) called GM(1,1) to conduct prediction according to the historical data of network traffic, and also uses adaptive filtering method to modify the residual error producted by GM(1,1), which achieves a higher accuracy. Moreover, this method integrates some characters of GM(1,1), such as less data needed and easy for operation. It has higher value for application. Experimental results show this method is effective and feasible.

Key words: Grey Model(GM), prediction, residual error, adaptive filter

中图分类号: 

• TP316.8