模型参数联合求解的网络流量混沌预测

发布时间：2018-03-01 17:20

本文关键词： 网络流量混沌理论最小二乘支持向量机径向基核函数参数优化　出处：《微电子学与计算机》2016年02期 　论文类型：期刊论文

【摘要】：流量预测是网络管理的基础,为了提高网络流量的预测精度,考虑模型参数之间的相互影响,提出一种模型参数联合求解的网络流量混沌预测模型.首先收集网络流量历史数据,采用混沌理论对历史数据进行重构,并确定模型参数范围,然后运用遗传算法模拟自然界的"适者生存、优胜劣汰"机制对模型参数进行联合求解,根据最优个体得到模型的最合理参数.最后根据最合理参数对网络流量训练样本进行学习,建立网络流量预测模型,并采用仿真实验对模型的性能进行对比分析.结果表明,此模型通过混沌理论可以有效挖掘历史数据中的网络流量变化特点,通过参数联合求解建立了精度高的网络流量预测模型,为非线性网络流量预测提供了一种新的建模思路.
[Abstract]:Traffic forecasting is the basis of network management. In order to improve the accuracy of network traffic prediction, the interaction between model parameters is considered. This paper presents a chaotic forecasting model of network traffic which is solved jointly with model parameters. Firstly, the historical data of network traffic are collected, and the historical data are reconstructed by using chaos theory, and the parameters range of the model is determined. Then the genetic algorithm is used to simulate the natural "survival of the fittest, survival of the fittest" mechanism to solve the model parameters. According to the optimal individual, the most reasonable parameters of the model are obtained. Finally, the network traffic training samples are studied according to the most reasonable parameters, the network traffic prediction model is established, and the performance of the model is compared and analyzed by simulation experiments. The results show that, This model can effectively mine the characteristics of network traffic change in historical data by chaos theory, and set up a high precision network traffic prediction model through the joint solution of parameters, which provides a new modeling idea for nonlinear network traffic prediction.
【作者单位】：肇庆学院教育技术与计算机中心;
【基金】：广东省科技项目(2012B061700063)
【分类号】：TP393.06

【参考文献】