论文部分内容阅读
大规模网络中的流量行为体现为一个相当复杂的非线性系统 ,目前国内外对它的研究还没有成熟的方法 .文章考虑网络流量非线性的特点 ,通过不同的数学模型将流量时间序列分解成趋势成分、周期成分、突变成分和随机成分 .根据分解 ,利用相应的数学工具分别建模四个相对简单的子成分以仿真复杂流量 .使用分解模型分析CER NET主干网络和NSFNET主干网络的长期流量行为 ,并将分析结果同传统的ARIMA季节模型比较 .通过比较仿真自相关函数和预报误差 ,发现分解模型在描述流量宏观行为时具有简单和高精度的优点 .
The traffic behavior in large-scale network is a rather complicated nonlinear system, so there is no mature method to study it at home and abroad.Based on the nonlinear characteristics of network traffic, the paper decomposes the traffic time series into Trend component, periodic component, mutation component and stochastic component.According to the decomposition, four relatively simple subcomponents are modeled respectively to simulate complex flow with the corresponding mathematical tools.The decomposition model is used to analyze the long-term flow of CERNET backbone network and NSFNET backbone network And compared the results with the traditional ARIMA seasonal model.Compared with the simulation autocorrelation function and forecast error, it is found that the decomposition model has the advantages of simplicity and precision in describing the macro behavior of traffic.