DocumentCode :
1375
Title :
Configurable Filter-Based Endpoint Detection in DMZI Vibration System
Author :
Xiangdong Huang ; Jia Yu ; Kun Liu ; Tiegen Liu ; Qinnan Chen
Author_Institution :
Sch. of Electron. Inf. Eng., Tianjin Univ., Tianjin, China
Volume :
26
Issue :
19
fYear :
2014
fDate :
Oct.1, 1 2014
Firstpage :
1956
Lastpage :
1959
Abstract :
In order to accurately determine the vibration starting point of a disturbance frame in the dual Mach-Zehnder interferometry vibration system, this letter proposes a novel endpoint detection method based on convolution window filtering. Through analyzing the fast Fourier transform spectrum of the undisturbed noise, we can acquire the cutoff frequency parameter to configure one high-pass filter to remove the noise, which mainly consists of low-frequency components. Then, the time instant corresponding to the vibration starting point can be detected through applying a thresholding search to the output of the high-pass filter. Experiments show that the proposed detection method outperforms the wavelet-based method. Hence, the proposed detection method has wide application values for its high precision, flexibility, and reliability.
Keywords :
Fourier transform optics; Mach-Zehnder interferometers; convolution; fast Fourier transforms; high-pass filters; optical filters; optical noise; optical sensors; optical windows; photodetectors; vibrations; DMZI vibration system; configurable filter-based endpoint detection; convolution window filtering; cutoff frequency parameter; dual Mach-Zehnder interferometry vibration system; fast Fourier transform spectrum analysis; high-pass filter; low-frequency components; noise removal; thresholding search; Convolution; Finite impulse response filters; Noise; Optical fiber filters; Optical fiber sensors; Optical fibers; Vibrations; Endpoint detection; configurable high-pass filter; convolution window; thresholding search;
fLanguage :
English
Journal_Title :
Photonics Technology Letters, IEEE
Publisher :
ieee
ISSN :
1041-1135
Type :
jour
DOI :
10.1109/LPT.2014.2343274
Filename :
6867293
Link To Document :
بازگشت