| 基于奇异值分解的混沌时间序列Volterra预测 |
| |
| 引用本文： | 陆振波,蔡志明,姜可宇.基于奇异值分解的混沌时间序列Volterra预测[J].武汉理工大学学报(交通科学与工程版),2007,31(4):672-675. |
| |
| 作者姓名： | 陆振波 蔡志明 姜可宇 |
| |
| 作者单位： | 海军工程大学电子工程学院,武汉,430033 |
| |
| 摘 要： | 提出一种用于混沌时间序列预测的奇异值分解Volterra滤波器.在Volterra滤波系数计算过程中,采用奇异值分解的方法得到线性方程组的最小二乘解.4种混沌序列的预测实验表明:该滤波器对混沌流的预测性能远优于NLMS自适应Volterra滤波器,前者的一步预测相对误差比后者小3~4个数量级.
|
| 关 键 词： | 混沌 预测 奇异值分解 Volterra滤波器 |
| 修稿时间： | 2007-02-08 |
Prediction of Chaotic Time Series Using Singular Value Decomposition Volterra Filter |
| |
| Lu Zhenbo,Cai Zhiming,Jiang Keyu.Prediction of Chaotic Time Series Using Singular Value Decomposition Volterra Filter[J].journal of wuhan university of technology(transportation science&engineering),2007,31(4):672-675. |
| |
| Authors: | Lu Zhenbo Cai Zhiming Jiang Keyu |
| |
| Institution: | College of Electronic Engineering, Naval of University Engineering, Wuhan 430033 |
| |
| Abstract: | A new singular value decomposition(SVD) Volterra filter is proposed to predict chaotic times series.The least square Volterra filter coefficients are received by using singular value decomposition in solving linear equations.Predictive experiments of four kinds of chaotic time series are carried out,and the results show that SVD Volterra filter has much better prediction performance for chaotic flow than normalized least mean square(NLMS) adaptive Volterra filter. |
| |
| Keywords: | chaos prediction singular value decomposition Volterra filter |
| 本文献已被 CNKI 维普 万方数据 等数据库收录！ |
