广东工业大学学报 ›› 2007, Vol. 24 ›› Issue (2): 76-79.
摘要: 针对现今在非线性系统盲辨识研究中遇到的困难,提出了一种基于最小二乘支持向量机(LS-SVM)的非线性系统盲辨识方法.该方法直接对非线性系统输出进行过采样,运用LS-SVM非线性建模技术,并结合输入的分布特性,从而完成非线性系统的盲辨识.介绍了盲系统辨识问题的研究内容及过采样技术原理,对LS-SVM的盲系统辨识机理和算法步骤进行了阐述.仿真结果表明了该方法在解决非线性系统盲辨识问题上的切实可行性.
[1] J.A.K. Suykens,J. Vandewalle. Least Squares Support Vector Machine Classifiers[J] ,1999[2] Hua Y B,Abed-Meraim K,Wax M.Blind system identifi-cation using minimum noise subspace. IEEE Transac-tions on Signal Processing . 1997[3] Tan H-Z,Chow T W S.Blind and total identification ofARMA mode in higher order cumulants domain. IEEETransactions on Industrial Electronics . 1999[4] Tan H-Z,Chow T W S.Blind identification of quadraticnonlinear models using neural networks with higher ordercumulants. IEEE Transactions on Industrial Electron-ics . 2000[5] Abed-Meraim K,Qiu W Z,Hua Y B.Blind system identi-fication. Proceedings of IEEE . 1998[6] Bai E W,Fu M.A blind approach to Hammerstein modelidentification. IEEE Transactions on Signal Process-ing . 2002[7] Bai E W.A blind approach to the Hammerstein-Wienermodel identification. Automatica . 2002[8] ChowTW S,Tan H-Z.HOS-based nonparametric and par-ametric methodologies for machine fault detection. IEEE Transactions on Industrial Electronics . 2000[9] LUO H,LI Y.The application of blind channel identifica-tion techniques to prestack seismic deconvolution. Pro-ceedings of IEEE . 1998[10] Smola A J.Learning with kernels. . 1998 |
No related articles found! |
|