Abstract: The chaos related theories are used to analyze the real network traffic about its chaos prediction attributes and the prediction-based IP QoS model and scheme are firstly given. The OPNET-based simulating result shows that the QoS performaces in heavy-load environment are all improved remarkably, comparing with the traditional static QoS configuring and measure-based dynamic QoS setting methods.

Key words: Chaos-based predictionr, DiffServ, IP QoS, QoS model

摘要： 利用混沌相关理论对具有综合业务应用的IP网流量进行了混沌特性与预测分析，提出了基于混沌预测的IP QoS实现模型与算法。OPNET的网络仿真实验表明，与传统的静态QoS参数配置方法和基于测量的QoS设置方法比较，该模型及方法在网络重负载环境下使各业务的QoS性能指标得到有效改善。

关键词: 混沌预测, DiffSer, IP QoS, QoS模型

CLC Number: 

• TP393.01