基于混沌特征的网络延迟预测模型

Network Delay Prediction Model Based on Chaotic Characteristic

结合非线性时间序列分析方法与CAIDA组织授权的真实Internet海量样本数据,计算了网络延迟演化序列的混沌特征量,结果表明演化序列具有混沌特征.在此基础上,对混沌系统中典型的Logistic模型加以改进,提出了一种基于Logistic模型的以正余弦函数作为指数衰减因子的模型,以描述网络延迟的演化态势.使用微粒群算法根据实际数据,分别从算法收敛性、模型的拟合准确度及预测准确度等方面对备选模型参数选优.实验结果表明最终优选模型在结构选择上比较合理,能够准确反映网络延迟的变化情况.

Together with the nonlinear time series analysis and the giant data samples authorized by CAIDA,the chaotic character of network delay time series indicates that its evolvement process existed chaotic identity.Then,we proposed a revised Logistic model with sine and cosine functions to describe the evolvement state of network delay.Moreover,particle swarm optimization（PSO） algorithm is adopted for the parameters estimation of the revised model,which is evaluated from the perspective of convergence, fitting accuracy and forecast accuracy. The result reviews that the structure of the optimized model is reasonable,and reflects the movement of network delay accurately.