期刊文献+

一种新的系统盲反向辨识方法

A new method of blind inverse identification of system
下载PDF
导出
摘要  基于系统输出信号过采样原理的系统反向辨识是一种新的盲辨识方法,这种方法不仅适用于最小相位系统,而且也适用于非最小相位系统。通过输出信号的过采样,可以获取未知输入信号和系统更多的信息。文中首先提出了这种解决盲辨识问题的算法,然后利用辨识出的系统模型恢复了源输入信号。这种方法在控制工程领域具有较高的实用价值。 Based on output oversampling of system inverse identification, is a new method of system identification, it can be applied in on not only minimum phase system but also nonminimum phase system. By oversampling on output signal, the information about unknown input signal and system can be gotten. This paper concern blind inverse problem arithmetic firstly, then restore the source signal by using system model which has been identified. This method has high practicality in control engineering field.
出处 《陕西工学院学报》 2003年第3期13-17,21,共6页 Journal of Shaanxi Institute of Technology
基金 国家自然科学基金资助项目(66974030)。
关键词 过采样 系统辨识 盲反向 over-sampling system identification blind inverse
  • 相关文献

参考文献6

  • 1Deng Z, H Zhang, S Liu, L Zhou. Optimal and self-tuning white noise estimators with applications to deconvolution and filtering problems[ J]. Automatica. 1996,32(2) : 199-216.
  • 2Mendel M. Tutorial on high-order statistics(spectra) in signal processing and system theory[J]. Theoretical results and applications. Proc IEEE 1991,79(3) :277-305.
  • 3Godard D N. Self-recovering equalization and cartier tracking in two-dimensional data communication systems[J]. IEEE Trans Commun 1980,COM-28 : 1867-1875.
  • 4Hua Y. Fast maximum likelihood for blind identification of multiple fir channels [ J ]. IEEE Trans Signal Processing, 1996,44 ( 3 ) :661-672.
  • 5Gurelli M I, C L Nikias. Evam : An eigenvector-based algorithm for multichahannel blind deconvolution of input colored signals [ J ]. IEEE Trans Signal Processing, 1995,43 ( 1 ) : 134-149.
  • 6Tong L, G Xu, T Kailath. Blind identification and equalization based on second-order statistics: A time domain approach[J]. IEEE Trans Inform Theory,1994,4012) :340-349.

相关作者

内容加载中请稍等...

相关机构

内容加载中请稍等...

相关主题

内容加载中请稍等...

浏览历史

内容加载中请稍等...
;
使用帮助 返回顶部