首页 | 本学科首页   官方微博 | 高级检索  
     检索      

混沌光学系统辨识的支持向量机方法
引用本文:叶美盈,汪晓东.混沌光学系统辨识的支持向量机方法[J].光学学报,2004,24(7):53-956.
作者姓名:叶美盈  汪晓东
作者单位:1. 浙江师范大学数理学院,金华,321004
2. 浙江师范大学信息科学与工程学院,金华,321004
基金项目:浙江省自然科学基金 ( 6 0 2 14 5 )资助课题
摘    要:将支持向量机用于混沌光学系统的辨识,以布拉格声光双稳混沌系统为例,通过计算机仿真实验,尝试了用最小二乘支持向量机进行混沌光学系统辨识的可行性,并将其与采用反向传播算法的前向神经网络辨识方法进行了比较。采用最小二乘支持向量机辨识的优点是其训练过程遵循结构风险最小化原则,不易发生过拟合现象;它通过解一组线性方程组可得到全局唯一的最优解;最小二乘支持向量机的拓扑结构在训练结束时自动获得而不需要预先确定。结果表明,本方法的辨识精度和速度均优于基于反向传播算法的前向神经网络,且对含噪混沌光学系统的辨识也同样适用,它可作为混沌光学系统辨识的有力工具。

关 键 词:非线性光学  混沌  辨识  支持向量机  光学双稳性
收稿时间:2003/8/21

Identification of Chaotic Optical System Based on Support Vector Machine
Ye Meiying Wang Xiaodong College of Mathematics and Physics,Zhejiang Normal University,Jinhua College of Information Science and Engineering,Zhejiang Normal University,Jinhua.Identification of Chaotic Optical System Based on Support Vector Machine[J].Acta Optica Sinica,2004,24(7):53-956.
Authors:Ye Meiying Wang Xiaodong College of Mathematics and Physics  Zhejiang Normal University  Jinhua College of Information Science and Engineering  Zhejiang Normal University  Jinhua
Institution:Ye Meiying 1 Wang Xiaodong 2 1 College of Mathematics and Physics,Zhejiang Normal University,Jinhua 321004 2 College of Information Science and Engineering,Zhejiang Normal University,Jinhua 321004
Abstract:A support vector machine based approach is presented for the identification of chaotic optical systems. The feasibility of this approach was demonstrated with the computer simulation through identifying a Bragg acoustooptic bistable chaotic system using a least squares support vector machine (LS SVM). The proposed identification method was compared with the feed forward neural network trained using back propagation algorithm for the system identification. The LS SVM possesses prominent advantages: over fitting is unlikely to occur by employing structural risk minimization criterion, the global optimal solution can be uniquely obtained owing to that its training is performed through the solution of a set of linear equations. Also, the LS SVM needs not determine its topology in advance, which can be automatically obtained when training process ends. Thus its identifying accuracy and speed were found to be better than that of a conventional feed forward neural network trained using back propagation algorithm. This method is robust with respect to noise, and it constitutes another powerful tool for the identification of chaotic optical systems.
Keywords:nonlinear optics  chaos  identification  support vector machines  optical bistability
本文献已被 CNKI 维普 万方数据 等数据库收录!
设为首页 | 免责声明 | 关于勤云 | 加入收藏

Copyright©北京勤云科技发展有限公司  京ICP备09084417号