Premium
Wavelet packet denoising robust regression applied to estimation of equivalent circuit parameters for thickness‐shear‐mode acoustic wave sensor
Author(s) -
Tan HuWei
Publication year - 1999
Publication title -
journal of chemometrics
Language(s) - English
Resource type - Journals
SCImago Journal Rank - 0.47
H-Index - 92
eISSN - 1099-128X
pISSN - 0886-9383
DOI - 10.1002/(sici)1099-128x(199911/12)13:6<543::aid-cem561>3.0.co;2-v
Subject(s) - robust regression , mathematics , robustness (evolution) , algorithm , linear regression , statistics , biochemistry , chemistry , gene
An Erratum has been published for this article in Journal of Chemometrics 14(1) 2000, 47. Fluctuation in characteristic parameters of the equivalent circuit for a thickness‐shear‐mode (TSM) acoustic wave sensor is a troublesome problem encountered in their practical applications. This fluctuation is due to interference from normal and non‐normal noise of the impedance analyser. In this paper, a novel robust non‐linear fitting method, namely complex least trimmed squares regression via wavelet packet denoising (WPD‐CLTSR), is proposed and utilized for robust parameter estimation to alleviate this fluctuation. The generalized simulated annealing (GSA) algorithm is adopted as an optimization procedure in the process of WPD‐CLTSR to guarantee convergence to the global optimum. The results for both simulated and experimental data sets show a significant improvement in the fluctuation compared with a conventional least squares method and a robust regression method, i.e. the ordinary complex least squares regression (OCLSR) method and the complex least trimmed squares regression (CLTSR) method. Therefore the WPD‐CLTSR method provides a safe alternative to the other widely used methods, no matter whether the noise is normal or non‐normal. Copyright © 1999 John Wiley & Sons, Ltd.