An event recognition method for Phi-OTDR based on the Gaussian mixture model/hidden Markov models

Lilong Ma; Tuanwei Xu; Kaiheng Yang; Fang Li

doi:10.1117/12.2573624

10 October 2020 An event recognition method for Φ-OTDR based on the Gaussian mixture model/hidden Markov models

Lilong Ma, Tuanwei Xu, Kaiheng Yang, Fang Li

Proceedings Volume 11554, Advanced Sensor Systems and Applications X; 1155417 (2020) https://doi.org/10.1117/12.2573624
Event: SPIE/COS Photonics Asia, 2020, Online Only

Abstract

Fiber optic distributed acoustic sensing (DAS) based on phase-sensitive optical time-domain reflectometry (Φ-OTDR) technology has been widely used in safety monitoring areas including monitoring of oil/gas pipes, communication or power cable, perimeters and so on, however it suffers from the high nuisance alarm rate (NAR) due to the non stationarity characteristics of signal and the interference of external environment. In this paper, GMMs-HMMs is utilized to reduce nuisance alarm rate, we prove that short time signal unit of appropriate length can contain the main frequency domain characteristics of signal, GMMs-HMMs is efficient recognition method for frequency domain sequence of signal. the experience results show the average recognition accuracy rate is 88.89% for seven events.

Citation Download Citation

Lilong Ma, Tuanwei Xu, Kaiheng Yang, and Fang Li "An event recognition method for Φ-OTDR based on the Gaussian mixture model/hidden Markov models", Proc. SPIE 11554, Advanced Sensor Systems and Applications X, 1155417 (10 October 2020); https://doi.org/10.1117/12.2573624

ACCESS THE FULL ARTICLE

INSTITUTIONAL
Select your institution to access the SPIE Digital Library.

SELECT YOUR INSTITUTION

PERSONAL
Sign in with your SPIE account to access your personal subscriptions or to use specific features such as save to my library, sign up for alerts, save searches, etc.

PERSONAL SIGN IN

No SPIE Account? Create one

PURCHASE THIS CONTENT

SUBSCRIBE TO DIGITAL LIBRARY

50 downloads per 1-year subscription

Members: $195

Non-members: $335 ADD TO CART

25 downloads per 1 - year subscription

Members: $145

Non-members: $250 ADD TO CART

PURCHASE SINGLE ARTICLE

Includes PDF, HTML & Video, when available