Acta Optica Sinica, Volume. 42, Issue 7, 0706005(2022)
Variational Mode Decomposition and Permutation Entropy Method for Denoising of Distributed Optical Fiber Vibration Sensing System
Figures & Tables(17)
Fig. 1. Waveform diagram of vibration simulation signal. (a) Time-domain; (b) frequency-domain
Fig. 3. Results of EMD-CC denoising method. (a) Signal after denoising; (b) error
Fig. 5. Results of CEEMD-CC denoising method. (a) Signal after denoising; (b) error
Fig. 7. Results of VMD-PE denoising method. (a) Signal after denoising; (b) error
Fig. 9. Denoising result of touching signal. (a) Modes of VMD-PE; (b) detail comparison of three denoising methods
Fig. 10. Denoising results of wheel rolling signal. (a) Modes of VMD-PE; (b) detail comparison of three denoising methods
Fig. 11. Denoising results of rain signal. (a) Modes of VMD-PE; (b) detail comparison of three denoising methods
Table 1. Permutation entropy of 8 kinds of signals
View tableTable 1. Permutation entropy of 8 kinds of signals
Signal x1(t) x2(t) x3(t) x4(t) x5(t) x6(t) x7(t) x8(t) Hp 0.4610 0.1753 0.2002 0.1427 0.0060 0.2273 0.6163 0.9689
Table 2. Correlation coefficient between each IMF of EMD and x(t)
View tableTable 2. Correlation coefficient between each IMF of EMD and x(t)
IMF IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 Correlation coefficient 0.2886 0.5992 0.4933 0.6414 0.3104 0.0040
Table 3. Correlation coefficient between each IMF of CEEMD and x(t)
View tableTable 3. Correlation coefficient between each IMF of CEEMD and x(t)
IMF IMF1 IMF2 IMF3 IMF4 IMF5 IMF6 Correlation coefficient 0.2756 0.5387 0.5423 0.6635 0.6445 0.0565
Table 4. Permutation entropy value of IMF1 under different K values
View tableTable 4. Permutation entropy value of IMF1 under different K values
K value 1 2 3 4 Hp of IMF1 0.4375 0.4566 0.4529 0.7664
Table 5. Indicators of three denoising methods
View tableTable 5. Indicators of three denoising methods
Method EMS IO RSN /dB Computing time/s Number of IMFs EMD-CC 1.0511 0.1298 9.1571 0.0263 6 CEEMD-CC 0.4068 0.0617 13.2791 7.8412 6 VMD-PE 0.1694 0.0030 17.0834 0.3331 3
Table 6. Denoising results of three kinds of actual vibration signals under three methods
View tableTable 6. Denoising results of three kinds of actual vibration signals under three methods
Method Net-touching Wheel rolling Raining RDNSN /dB Computing time /s Recognition accuracy /% RDNSN /dB Computing time /s Recognition accuracy /% RDNSN /dB Computing time /s Recognition accuracy /% Before denoising - - 78.3 - - 80.4 - - 76.6 EMD-CC 35.2241 0.5471 83.4 33.0175 0.5501 86.7 30.9846 0.3792 85.8 CEEMD-CC 35.1737 8.6542 90.5 32.9347 7.1748 93.3 30.1071 8.9113 92.4 VMD-PE 32.5358 1.4432 99.3 30.5546 1.6320 99.8 29.3435 1.2349 99.6
Tools
Get Citation
Copy Citation Text
Miao Yu, Yaolu Zhang, Yutong He, Mingyang Sun, Qian Kong, Zhifeng Zheng. Variational Mode Decomposition and Permutation Entropy Method for Denoising of Distributed Optical Fiber Vibration Sensing System[J]. Acta Optica Sinica, 2022, 42(7): 0706005
Paper Information
Category: Fiber Optics and Optical Communications
Received: Jul. 16, 2021
Accepted: Oct. 8, 2021
Published Online: Mar. 28, 2022
The Author Email: Yu Miao (yumiao@zsc.edu.cn)
© Copyright 2018-2021 | Chinese Laser Press.
All Rights Reserved 沪ICP备15018463号-20