Fiber-optic Vibration Signal Recognition Based on BLCD Decomposition and ACO-DBN Network

Table 1. Experimental instruments and related parameters
View table
View in Article
Table 1. Experimental instruments and related parameters
Linewidth of a laser Power Sensor module Sampling rate of the collector Collected number
2 kHz 25 mW Standard single mode fiber $5 \times 10^{6}$ sample/s 100 000

Table 2. Comparison of results of different decomposition methods
View table
View in Article
Table 2. Comparison of results of different decomposition methods
Decomposition method Correlation coefficient Root mean square error
ISC₁ ISC₂ ISC₃ ISC₁ ISC₂ ISC₃
LCD 0.751 5 0.513 3 0.509 5 0.041 7 0.041 0 0.046 4
BLCD 0.763 0 0.577 5 0.522 2 0.040 1 0.040 4 0.044 7

Table 3. Comparison of the effects of two filtering ways

View table

View in Article

Table 3. Comparison of the effects of two filtering ways

Filtering method	Signal	Signal to noise/ dB	Root mean quare error
Add larger ISC	Rain	21.59	0.002 3
	Climb	7.372	0.019 5
	Knock	6.955	0.032 5
	Wind	8.132	0.028 3
IF	Rain	27.09 3	0.001 2
	Climb	12.11 6	0.011 3
	Knock	17.29 8	0.009 9
	Wind	9.732	0.007 4

Table 4. Recognition results before and after feature dimensionality reduction
View table
View in Article
Table 4. Recognition results before and after feature dimensionality reduction
Feature Rain Climb Knock Wind Recognition rate Recognition time
Direct $Φ$ 40% 100% 90% 46.67% 69.17% 2.64 s
Fisher 93.33% 100% 100% 83.33% 94.17% 0.83 s

Table 5. Comparison of recognition results of different methods

View table

View in Article

Table 5. Comparison of recognition results of different methods

Method	Rain	Climb	Knock	Wind	Recognition rate	Recognition time
BP	76.67%(23/30)	96.67%(29/30)	17.86%(5/30)	100%(30/30)	73.48%	1.499 s
ACO-BP	90%(27/30)	86.67%(26/30)	80%(24/30)	93.33%(28/30)	87.50%	1.362 s
DBN	76.67%(23/30)	100%(30/30)	93.33%(28/30)	83.33%(25/30)	83.33%	1.523 s
ACO-DBN	93.33%(28/30)	100%(30/30)	93.33%(28/30)	96.67%(29/30)	95.83%	0.715 s

Table 6. Comparison of multi-point vibration identification results

View table

View in Article

Table 6. Comparison of multi-point vibration identification results

Method	knock at one point	knock at two points	Recognition rate	Recognition time
BP	63.33%(19/30)	50%(15/30)	56.67%	1.214 s
ACO-BP	80%(24/30)	73.33%(22/30)	76.67%	1.327 s
DBN	90%(27/30)	86.67%(26/30)	88.34%	1.163 s
ACO-DBN	96.67%(29/30)	90%(27/30)	93.33%	0.815 s

Tools

Get Citation

Copy Citation Text

Yuzhao MA, Ruisong WANG, Xinglong XIONG. Fiber-optic Vibration Signal Recognition Based on BLCD Decomposition and ACO-DBN Network[J]. Acta Photonica Sinica, 2021, 50(2): 44

Download Citation

EndNote(RIS)BibTex Plain Text

Set citation alerts for article

Save article for my favorites