A parameter estimation method for model analysis

DOI10.1007/BF02896486MaRDI QIDQ854450zbMATH OpenFDO

Authors Sunjoo Kwon, Jae Heon Yun, Seyoung Oh

Publication date 4 December 2006

Published in Journal of Applied Mathematics and Computing (Search for Journal in Brave)

Full work available at URL https://doi.org/10.1007/BF02896486

zbMATH Keywords

algorithmnumerical examplesOverdetermined systemspseudoinversessignal processingstructured nonlinear total least norm (SNTLN) method

Mathematics Subject Classification ID

Point estimation (62F10) Numerical solutions to overdetermined systems, pseudoinverses (65F20) Signal theory (characterization, reconstruction, filtering, etc.) (94A12)

Review text

The paper is devoted to the development of an efficient hybrid algorithm of Structured Nonlinear Total Least Norm (SNTLN) method, and a scheme of separation by parts of parameters. A regularization process for SNTLN method is introduced. The process of parameter estimation separation scheme is formulated and a new algorithm is derived. The proposed algorithm is developed by means of applying both of parameter separation scheme and the regularized structured nonlinear total least squares method. A major benefit of the parameter separation scheme is the reduction of the residual norm steadily to the accurate approximate. It is also not sensitive to the initial choice of the parameters. This scheme guarantees locally convergence of parameters. Numerical simulations on signal processing models are described to illustrate the obtained results and to show that the suggested algorithm is often superior to SNTLN method in the reduction of the residual.

Recommendations

• Efficient parameters estimation method for the separable nonlinear least squares problem
• Exponential modeling with unknown model order using structured nonlinear total least norm
• Structured Total Least Norm for Nonlinear Problems
• scientific article; zbMATH DE number 2185756
• scientific article; zbMATH DE number 465411

Cites work

• A method for structured linear total least norm on blind deconvolution problem
• Blind Deconvolution Using a Regularized Structured Total Least Norm Algorithm
• Choosing regularization parameters in iterative methods for ill-posed problems
• scientific article; zbMATH DE number 2104353 (Why is no real title available?)
• Regularization by Truncated Total Least Squares
• Separable Nonlinear Least Squares with Multiple Right-Hand Sides
• Structured Total Least Norm for Nonlinear Problems
• The Differentiation of Pseudo-Inverses and Nonlinear Least Squares Problems Whose Variables Separate
• Tikhonov Regularization and Total Least Squares
• Total Least Norm Formulation and Solution for Structured Problems

Cited in

(5)

• Exponential modeling with unknown model order using structured nonlinear total least norm
• scientific article; zbMATH DE number 2185756 (Why is no real title available?)
• EFFECTIVE TECHNIQUES FOR ESTIMATING THE PARAMETERS OF GRUBBS MODEL WITH APPLICATIONS
• A method for model identification and parameter estimation
• Parameter resolution of estimation methods for linear regression models

This page was built for publication: A parameter estimation method for model analysis

Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q854450)