Optics and Photonics Journal, 2013, 3, 122-125
doi:10.4236/opj.2013.32B030 Published Online June 2013 (http://www.scirp.org/journal/opj)
Analysis of the Interference Signal of the Distributed
Optical Fiber Sensing Based on DSP*
Tingting Du1, Xiong Wan1,2#, Zhimin Zhang1, Huming Zhang1, Qing Zhang1
1Key Laboratory of Nondestructive Testing(Ministry of Education), Nanchang Hangkong University, Nanchang, Jiangxi, China
2Key Laboratory of Spatial Active Opto-electronic Techniques, Shanghai Institute of Technical Physics,
Chinese Academy of Sciences, Shanghai, China
Email: #nchundt114@126.com
Received 2013
In this paper, the author analyzes characteristics and extracting method of interference signal of the distributed optical
fiber sensing. In the distributed optical fiber sensing, realizing alarm and positioning function only through the cross-
correlation operation will increase the load of the system, can make misinformation rate of the system be improved
greatly. Therefore, before the localization algorithm, adding a interference signal feature recognition is very necessary,
can reduce unnecessary operation loss and reduce the load of the system, also reducing the number of the false posi-
Keywords: The Optical Fiber Sensing; Mach-zehnder Interferometer; Energy; Interference Signal; Signal Recognition
1. Introduction
With the development of optical fiber communication
technology, the optical fiber sensing technology draws a
great attention, and is widely used in the earthquake
monitoring[1,6,7], temperature alarm [2] and becomes
one of the most important technology in information so-
ciety, and has huge market potential. Optical fiber sensor
uses fiber itself sensitivity characteristics work. When
the light from the laser light source spreads in the fiber ,if
external factors such as stress, temperature, electric field
and magnetic field change, the characteristic parameters
such as the light wave amplitude, phase, wavelength and
polarization state will change[3].Therefore, the optical
fiber sensing technology draws a great attention, be re-
searched and applied in almost every field .
This paper uses the sensor that is based on mach-
zehnder interferometer [5,8]. The biggest advantage of
the interferometric sensor is that interference detecting
sensitivity is extremely high. In mach-zehnder interfer-
ometer fiber-optic sensor, alarm and positioning are two
main functions of the distributed optical fiber sensing
system. The pretreatment of the interference signal is of
great significance to improve the alarm accuracy and the
positioning accuracy. Sorting the interference signal that
is gathered by Digital Signal Processor (DSP), establishing
corresponding mode, filtering useless information and noise,
selecting and extracting needed information, can greatly
improve the sensor system alarm and positioning accu-
racy. Based on the short time analysis method, the paper
discusses and analyzes short-term energy of identifica-
tion method.
2. Experimental System
2.1. Experimental System Structure
The experiment system schematic diagram as shown in
Figure 1, uses DFB laser, its wave length is 1550 nm,
the line width is 2MHZ. The light that the DFB laser
produces is divide d into two beam of light by the coupler
1. One of the two beam of light throug h the fiber coupler
3 goes into MZI, spreading along the clockwise in the
interferometer, occurring interference in the coupler 4.
Finally, half of light power of the coupler 2 is received
by photoelectric detector PIN1.The other of the two
beam of light spreads along the counterclockwise in the
interferometer, occurring interference in the coupler 3.
Finally, half of light power of the coupler 3 is received
by photoelectric detector PIN2.
*This work was su pported by Chinese Natural Science Foundation (grant
61271397 ), Jiangxi provincial Natural Scien ce Foundation (grant 20122
BAB202009), Jiangxi provincial education department Science and
Technology Foundation (grant GJJ12408) and preferentially funded by
the “Hundred Talents Plan” of the Chin ese Academy of Scie nces.
#Corresponding author.
When vibration or pressure signal applies to optical
fiber interference arm, geometric size and refractive in-
dex parameters of fiber will change, resulting in light
wave phase change. Through photoelectric detector, the
Copyright © 2013 SciRes. OPJ
T. T. DU ET AL. 123
Figure 1. Experimental of the interference signal analysis.
change of intensity of interference light will turn into a
change of electrical signal. Through detecting and recog-
nizing these electrical signals, we can realize alarm of the
invasion signal. The experiment uses Digital Signal Proc-
essor (DSP) process signal. Because DSP has strong
graphic display ability, not only can show the original
signal graphics, also can display FFT spectrum diagram
of signal. Because DSP2812 in a period can complete
32-bit * 32-bit multip licative additiv e operation. It makes
the processing speed of algorithm in DSP's chip faster.
Therefore, we use DSP do operation for interference
2.2. System Positioning Principle
In Figure 1, supposing the optical fiber length is L me-
ters between coupler 3 and coup ler 4, If in a moment this
position that has x meters from coupler 3 happens dis-
turbance. The disturbance signal will respectively spread
along the clockwise and anti-clockwise direction. The
time that the light spreading along Clockwise direction
and arriving to coupler 2 takes is 1. The time that the
light spreading along counterclockwise direction and
arriving to coupler 3 takes is 2. By measuring the time
difference, we can get vibration point position so as to
realize positioning[4].
3. Experimental Installment
3.1. Experiment
Outputed signal of the optical fiber sensor in addition to
containing vibration interference signal, also contains
noise interference signal. In order to avoid noise signal,
to improve the effectiveness of the subsequent testing, it
is necessary to identify disturbance signal and noise sig-
nal. This experiment mainly analyses the characteristics
of two kinds of actual signal that are disturbance signal
and noise signal.
In the experiment, we use the short-term energy analy-
sis method. Short-term energy analysis is based on frame
and window for signal. As such, the basic concept for
processing interference signal is that divide discrete in-
terference signal into a certain length unit to process, also
is "short" processing method.
If the discrete sequence that analog signal outputed by
the sensor through AD change is ( ),0,1,2...
ii m, the
short-term energy calculation formula is
()( ()())()()
 
 
The represents sliding window function, n = 0,
Particularly, when the window function is rectangular
window function, the short-term energy calcula-
tion form ul a is ()Sn
() ().
Snx i
 
The original signal diagram is shown in Figures 2-3.
Vibration interference signal diagram and noise signal
diagram have obvious difference. The short-term energy
of signal chart is shown in Figures 4-7. Vibration inter-
ference signal of the short-term energy diagram and
noise signal of the short-term energy diagram also have
obvious difference.
3.2. Experimental Analysis
The experiment system of the AD sampling rate is 100
KHZ, every frame sampling point is 800. Every frame
signal of the short time average energy is shown in Table
Figure 2. Vibration interference s ign al original diagram.
Copyright © 2013 SciRes. OPJ
Figure 3. Noise interference s ignal original diagram.
Figure 4. Vibration interference signal of the short-term
energy diagram.
Figure 5. Vibration interference signal of the short-term
energy diagram.
From the above, we can be seen that we can use the
short time average energy distinguish between the vibra-
tion signal and noise signa l. When the short time average
energy is smaller than a certain threshold (for example
12), we judge it as vibration signal. When the short time
average energy is greater than a certain threshold, we
judge it as noise signal.
Figure 6. Noise interference signal of the short-term energy
Figure 7. Noise interference signal of the short-term energy
Table 1. Short time average energy.
Signal types Frame numbe r Short time average
vibration signal 800100XH0 0.5027
vibration signal 800100XH1 3.0623
vibration signal 800100XH2 0.4807
vibration signal 800100XH3 9.2760
vibration signal 800100XH4 6.6546
Noise signal 800100X0 17.9425
Noise signal 800100X1 16.8052
Noise signal 800100X2 16.4081
4. Conclusions
This experiment completes the signal of characteristic
identification by using DSP. The signal feature recogni-
tion is an important part of signal pretreatment of the
distributed optical fiber sensing. This experiment ana-
lyzes signal characteristics in time domain. The technol-
ogy filters undesired signal, greatly reducing the unnec-
Copyright © 2013 SciRes. OPJ
Copyright © 2013 SciRes. OPJ
essary operation loss, decreasing the system load, and at
the same time, improving the alarm positioning accuracy.
[1] D. Lgardner and S. Lgarrett, “Fiber Optic Seismic Sen-
sor,” Fiber Optic and Laser sensor V, Vol. 1987, No. 838,
pp. 271-278
[2] C. X. Ao, K. X. Jin, T. Li, et al., “Mach-Zehnder Optical
Fiber Temperature sensor research,” Journals of Shanxi
Technology University, Vol. 22, No. 6, 2004, pp.104-107.
[3] Y. He and D. M. Liu, “Optical Fiber Sensor Phase De-
modulation Technology Based on Labview,” 2006, Vol.
19, No.1, pp. 94-97.
[4] J. Tang, “Fiber Optic Interferometer Phase Detection
Research Based on Mach-ZeHnder,” Sensing Technology,
Vol. 6, No. 2, 2000, pp. 96-100.
[5] E. UDD, “Fiber Optical Sensors: An Introduction for
Engineers and Scientists,” John Wiley & Sons, Canad,
[6] J. P. Dakin, “Distributed Anti-Stokes Ratio Thermome-
try” 3rd International Conference on Optical Fibre Sen-
sors, San Diego,1985.
[7] T. Ohmachi and H. Matsumot, “Sea Water Pressure In-
cluded by Seismic Ground Motions and Tsunami,” Pro-
ceeding of International Tsunami Symposium, 2001, pp.
[8] A. A. Chtcherbakow, P. L.Swart and S. J. Spammer “A
Fiber Optic Disturbance Location Sensor Using Modified
Sagnac and Mach-Zehnder Interferometers,” Proceeding
of 12th Optical Fiber Sensor, 1997, Vol. 16, pp. 516-519.