Application of Improved Empirical Mode Decomposition Algorithm in Fiber Bragg Grating Perimeter Intrusion Behaviors Classification

Yong Chen; Wangyue An; Huanlin Liu; Zhiqiang Liu; Lixin Zhou

doi:10.3788/CJL201946.0304003

Chinese Journal of Lasers, Volume. 46, Issue 3, 0304003(2019)

Application of Improved Empirical Mode Decomposition Algorithm in Fiber Bragg Grating Perimeter Intrusion Behaviors Classification

Yong Chen^1、*, Wangyue An¹, Huanlin Liu², Zhiqiang Liu¹, and Lixin Zhou¹

Author Affiliations

¹ Key Laboratory of Internet of Things and Networking Control Under Ministry of Education, Chongqing University of Posts and Telecommunications, Chongqing 400065, China

² Key Laboratory of Fiber-Optic Communication Technology Under Ministry of Information Industry,Chongqing University of Posts and Telecommunications, Chongqing 400065, China

show less

Abstract Get PDF(in Chinese)

Figures & Tables(12)

Fig. 1. Schematic of matching wavelet calculation

Download full size

Fig. 2. Flow chart of improved EMD algorithm

Download full size

Fig. 3. Short time average zero-crossing rate distribution of intrusion signal. (a) Intrusion perturbation signal; (b) distribution of short time average zero-crossing rate

Download full size

Fig. 4. Photo of perimeter security system

Download full size

Fig. 5. Intrusion signal and short time average zero-crossing rate distribution before zeroing. (a) Intrusion signal; (b) short time average zero-crossing rate

Download full size

Fig. 6. Intrusion signal and short time average zero-crossing rate distribution after zeroing. (a) Intrusion signal; (b) short time average zero-crossing rate

Download full size

Fig. 7. Extracted intrusion signals. (a) Climbing; (b) colliding; (c) shearing; (d) nonintrusive; (e) touching

Download full size

Fig. 8. Identification results of climbing signal. (a) IMF components; (b) frequency domain distribution

Download full size

Fig. 9. Identification results of shearing signal. (a) IMF components; (b) frequency domain distribution

Download full size

Fig. 10. Identification results of nonintrusive signal. (a) IMF components; (b) frequency domain distribution

Download full size

Table 1. Identification results of five behaviors by three algorithms

View table

Table 1. Identification results of five behaviors by three algorithms

Behavior	Experimentnumber	Number of successful recognition			Recognition rate /%
Behavior	Experimentnumber	EMD	EEMD	Proposed algorithm	EMD	EEMD	Proposed algorithm
Climbing	100	78	83	96	78	83	96
Shearing	100	91	95	98	91	95	98
Colliding	100	85	90	96	85	90	96
Touching	100	80	86	97	80	86	97
Nonintrusive	100	94	100	100	94	100	100

Table 2. Denoising performance of three algorithms
View table
Table 2. Denoising performance of three algorithms
Noise level /dB Signal-to-noise ratio after denoising /dB
EMD EEMD Proposed algorithm
5 22.1683 23.2356 24.9389
10 26.6196 27.0813 28.1797
15 33.6422 34.1742 34.3243
20 35.3105 36.5819 37.7545
25 36.6726 38.7618 40.2439

Tools

Get Citation

Copy Citation Text

Yong Chen, Wangyue An, Huanlin Liu, Zhiqiang Liu, Lixin Zhou. Application of Improved Empirical Mode Decomposition Algorithm in Fiber Bragg Grating Perimeter Intrusion Behaviors Classification[J]. Chinese Journal of Lasers, 2019, 46(3): 0304003

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites