Similarity Estimation of Sensor Signals using Continuous Wavelet Correlation Analysis

Author

Sasaoka, Hideichi ; Matsuura, Yuhou

Author_Institution

Res. & Dev. Headquarters, Yamatake Corp., Kanagawa

fYear

2006

Firstpage

3290

Lastpage

3292

Abstract

Based on the wavelet transform, a method to estimate similarities among sensor signals is proposed. In this method, the time-frequency information which is obtained through the continuous wavelet transform is analyzed using the correlation analysis. It is possible to detect the frequency band which is not influenced by noise. This method is applied to chemical process signals. As the result, one signal was found to be highly related to product property. An inappropriate tuning of control loops was also found easily with the support of this method

Keywords

correlation methods; signal processing; time-frequency analysis; wavelet transforms; chemical process signals; continuous wavelet correlation analysis; control loops; frequency band detection; sensor signals; similarity estimation; time-frequency information; wavelet transform; Chemical processes; Chemical sensors; Continuous wavelet transforms; Information analysis; Signal analysis; Signal processing; Time frequency analysis; Tuning; Wavelet analysis; Wavelet transforms; Continuous Wavelet Transform; Correlation Analysis; Similarity Estimation;

fLanguage

English

Publisher

ieee

Conference_Titel

SICE-ICASE, 2006. International Joint Conference

Conference_Location

Busan

Print_ISBN

89-950038-4-7

Electronic_ISBN

89-950038-5-5

Type

conf

DOI

10.1109/SICE.2006.315008

Filename

4108325