Open Access
Add to BibSonomyAbstract
An optical chaos and hybrid wavelength division multiplexing/time division multiplexing (WDM/TDM) based large capacity quasi-distributed sensing network with real-time fiber fault monitoring is proposed. Chirped fiber Bragg grating (CFBG) intensity demodulation is adopted to improve the dynamic range of the measurements. Compared with the traditional sensing interrogation methods in time, radio frequency and optical wavelength domains, the measurand sensing and the precise locating of the proposed sensing network can be simultaneously interrogated by the relative amplitude change (RAC) and the time delay of the correlation peak in the cross-correlation spectrum. Assisted with the WDM/TDM technology, hundreds of sensing units could be potentially multiplexed in the multiple sensing fiber lines. Based on the proof-of-concept experiment for axial strain measurement with three sensing fiber lines, the strain sensitivity up to 0.14% RAC/με and the precise locating of the sensors are achieved. Significantly, real-time fiber fault monitoring in the three sensing fiber lines is also implemented with a spatial resolution of 2.8 cm.
© 2015 Optical Society of America
1. Introduction
Fiber Bragg grating (FBG) has attracted extensive attentions due to its intrinsic advantages of erosion resistance, immunity to electromagnetic interference, compactness, and excellent multiplexing capability [1]. In practical FBG based quasi-distributed sensing applications, plenty of wavelength-encoded FBGs should be distributed in the sensing network to realize multi-point measurements. However, directly monitoring wavelength shifts of all the FBGs is difficult to implement. Intensity demodulation incorporated with appropriate optical filter and novel interrogation scheme can be a good candidate. Meanwhile, compared to typical FBG, CFBG with wider bandwidth possesses higher reflection power, thus can improve the dynamic range of the measurements in grating intensity demodulation [2]. To comply with the development of large capacity, high precision, and long distance, various multiplexing schemes have been widely utilized in the sensing network, including wavelength division multiplexing (WDM) [3, 4], time division multiplexing (TDM) [5, 6], frequency division multiplexing (FDM) [7] or hybrid schemes [7–9]. In many TDM sensing networks [5, 6, 8–11], pulse generator is usually required for the locating of each sensor, resulting in a costly and complicated interrogation system. Moreover, almost all the sensing networks cannot provide real-time fiber fault monitoring, thus it may take a long time and lots of manpower to pinpoint the fiber fault.
Recently, optical chaos has been utilized in the applications of secure optical communication [12], regulation of laser’s coherent length [13], chaotic laser radar ranging [14], and fiber fault monitoring [15–17]. In our previous work [17], simultaneous and precise fiber fault locating in WDM-PON is implemented based on optical broadband chaos generated from the semiconductor optical amplifier (SOA) based ring cavity. Through applying cross-correlation algorithm to the reference signal and the Fresnel reflection signal of the fiber fault, precise locating of the fiber fault can be obtained. Up to now, applying optical chaos incorporated with hybrid WDM/TDM technology in sensing areas has not been reported.
In this paper, an optical chaos based large capacity quasi-distributed sensing network is proposed. CFBG intensity demodulation is adopted to improve the dynamic range of the measurements. Cross-correlation algorithm is applied to the chaotic reference signal and probe signal to interrogate the measurand sensing and the precise locating of each sensor. And hybrid WDM/TDM technology is employed to expand the multiplexing capability. Additionally, real-time fiber fault monitoring can be also implemented to improve its survivability in harsh environment. At last, a proof-of-concept experiment is conducted to demonstrate the proposed sensing network.
2. Schematic diagram and operation principle
2.1 Schematic diagram
The schematic diagram of the large capacity quasi-distributed sensing network is depicted in Fig. 1, consisting of a central office (CO) and an optical sensing network (OSN). In the CO, a SOA (Model: S7FC1013S) based ring cavity provides broadband chaotic light, where the isolator and the polarization controller are used to ensure unidirectional propagation and appropriate polarization state, respectively. Note that the PC remains unchanged to provide a stable output of the chaotic light during the measurement. An erbium doped fiber amplifier (EDFA) acts as a pre-amplifier for sensing power boost, and an optical band pass filter (OBPF, Waveshaper, FINISAR) is used as a multi-channel filter in accordance with each sensing CFBG. A 99:1 optical coupler (OC) is adopted to divide the light into the reference light and the probe light. The reference light is directly detected without additional fiber line, and the probe light is launched into the OSN through a circulator. Besides, a set of demodulation device is employed to process the reference signal and the probe signal. It includes two avalanche photon detectors (APD, 1.1GHz), a real-time oscilloscope (OSC), and a data processing computing system. In the OSN, a large capacity CFBG sensors array is deployed through an optical power splitter. The number of the sensing fiber lines is m, determined by the branches of the splitter. And the number of the CFBGs deployed in one sensing fiber line is n. Meanwhile, m delay lines with different lengths are purposefully adopted in front end of each sensing fiber line. For a convenient introduction, each CFBG sensor is assigned a unique number of Sxy according to its physical address in the network as depicted in Fig. 1. The CFBGs deployed in one sensing fiber line possess different central wavelengths, while the CFBGs in different lines, but in the same relatively address, are identical. All the CFBGs possess strong reflectivity of ~20 dB to improve the signal to noise ratio (SNR), and the bandwidth is selected as a tradeoff result of the multiplexing capacity and the dynamic range of the measurements. The artificially arranged sensors array serves for the hybrid WDM/TDM technology to expand the multiplexing capacity.
The characters of the broadband chaotic light generated from the ring cavity are illustrated in Fig. 2. The optical 3-dB bandwidth of the broadband chaos is ~50 nm, covering the whole C-band. The output chaotic light possesses a random up-and-down waveform in time series and a good normalized auto-correlation property of sharp δ function shape.
Fig. 2 (a) Optical spectrum, (b) time series and (c) auto-correlation spectrum of the broadband chaos.
2.2 CFBG intensity demodulation
CFBG intensity demodulation is implemented by the programmed OBPF, which slices optical broadband chaos into multi-wavelength channels in consistent with the CFBG sensors. The transmission spectrum of a certain wavelength channel (red solid line) and the reflection spectrum of the corresponding CFBG (blue dash line) are simultaneously plotted in Fig. 3(a), the bandwidths of which are both ~0.8 nm (100 GHz). Wavelength shift of the CFBG can change the spectral overlap between each wavelength channel and corresponding CFBG sensor, resulting in power variation of the probe light. Take axial strain as an example, when the strain is applied on the CFBG, wavelength redshift occurs. Spectral overlaps with different strain can be observed in Fig. 3(b). And the optical power of the probe light as a function of strain is also presented in Fig. 3(c), possessing an excellent linearity of R2 ˃ 0.999 from 80 με to 880 με. It should be noted that although the chaotic signal possesses a random up-and-down waveform in time series, the average optical power of the chaos is stable. And the proposed CFBG intensity demodulation is just related to the average optical power of the chaos.
Fig. 3 (a) Transmission spectrum and reflection spectrum of the wavelength channel and the corresponding CFBG sensor. (b) Spectral overlaps with different strain. (c) Optical power of the probe light as a function of strain.
2.3 Cross-correlation interrogation method
As mentioned above, the reference signal and the probe signal are respectively acquired from the OSC via the APDs. To simultaneously interrogate the measurand sensing and the precise locating, the cross-correlation algorithm is adopted to process the reference and probe signals, which can be expressed as
is defined to be the reference light. denotes the light reflected by a certain CFBG, where is the corresponding time delay, and is the coefficient of the optical power variation determined by the wavelength shift. Therefore, the probe light can be defined as , namely the superposition of all the CFBGs reflection light. is the number of the CFBG sensors. Through data processing, the correlation peaks located at different positions in the correlation spectrum correspond to different CFBG sensors. The measurand sensing is obtained by monitoring the amplitude change of the peaks. And the precise locating of the sensors is calculated by , where and are the speed of light in vacuum and the refractive index of the fiber, respectively. Consequently, simultaneous interrogation of the measurand sensing and the precise locating is achieved. Meanwhile, fiber fault monitoring can be also implemented through processing the reference signal and the Fresnel reflection signal of the fiber fault based on the same principle.2.4 WDM/TDM multiplexing principle
To expand the multiplexing capacity, hybrid WDM/TDM technology is employed in this sensing network. The aforementioned OBPF slices the optical broadband chaos into multi-wavelength channels in accordance with the CFBG sensors, implementing CFBG intensity demodulation and WDM, simultaneously. Different-length delay lines utilized in the OSN skillfully realize TDM incorporated with the optical power splitter. Accordingly, the CFBG sensors can be discriminated due to their different positions of the corresponding correlation peaks in the cross-correlation spectrum.
The multiplexing capacity can be estimated as follows. Firstly, the number of the wavelength channels is determined by the WDM technology, which is related to the broadband chaos and the CFBG intensity demodulation. The 3-dB bandwidth of the broadband chaos is ~50 nm. In our experiment, the reflection bandwidth of the CFBG is ~0.8 nm. Based on the CFBG intensity demodulation method, normally at least another 0.8 nm protective bandwidth should be reserved to avoid the crosstalk between the adjacent wavelength channels. As a result, 32 wavelength channels could be achieved. Note that the bandwidth selection of the CFBG is a tradeoff result of the multiplexing capacity and the dynamic range of the measurements. Therefore, more wavelength channels could be achieved at the sacrifice of the dynamic range. Furthermore, the number of the time channels is determined by the number of the splitting branches, which is subjected to the optical power budget. As shown in the schematic diagram of the sensing network, an EDFA is utilized to boost the optical power, and two APDs are used for weak signal detection. The insertion loss of the different-length delay lines can be neglected due to the low loss of the single mode fiber. In consideration of a 1 × 16 splitter, about 24-dB insertion loss is introduced for each CFBG sensor. In our preliminary experiment, a variable optical attenuator (VOA) is utilized to simulate the insertion loss of the 1 × 16 splitter. Consequently, a distinct correlation peak can be observed in the correlation spectrum, which possesses enough SNR for sensing applications. Therefore 1 × 16 splitter could be adopted to guarantee sufficient optical power for each CFBG sensor. Namely, 16 time channels are achievable. In total, the number of the CFBG sensors multiplexed in the OSN could reach to 16 × 32 = 512.
3. Proof-of-concept experiment and discussions
To demonstrate the proposed sensing network, a proof-of-concept experiment is conducted with six CFBG sensors (S11, S12, S13, S21, S23, and S32) deployed in three sensing fiber lines (L1, L2 and L3). The schematic diagram of the OSN is plotted in Fig. 4(a). The central wavelengths of the CFBGs are 1549.2 nm for S11 and S21, 1550.8 nm for S12 and S32, 1552.3 nm for S13 and S23, respectively. All the CFBGs utilized in the experiment possess ~5-mm grating length, ~0.8-nm reflection bandwidth and ~20-dB reflectivity. The lengths of these three delay lines are 0, 50 m and 100 m. The time delay of the reflection light from each CFBG should be different to avoid the confusion of the sensors in the correlation spectrum, thus the lengths of L1 and L2 are artificially arranged to be less than the lengths of the delay lines of L2 and L3 in our experiment, respectively. As shown in Fig. 4(b), three wavelength channels of ~0.8 nm bandwidth is filtered by the OBPF in consistent with the CFBG sensors of different central wavelengths. And the 0.8-nm protective bandwidth is adopted to avoid the crosstalk between the adjacent wavelength channels. Moreover, an additional channel (testing channel) of 2.4-nm (300-GHz) bandwidth is reserved to provide sufficient optical power for fiber fault monitoring. The radio frequency (RF) spectrum and the auto-correlation spectrum of the multi-channel chaos are depicted in Fig. 4(c) and 4(d), respectively. Axial strain as a measurand is selected to demonstrate the sensing applications.
Fig. 4 (a) OSN of the proof-of-concept experiment. (b) Optical spectrum of the multi-channel chaos (including three wavelength channels and a testing channel). (c) RF spectrum and (d) auto-correlation spectrum of the multi-channel chaos.
The cross-correlation spectra free of strain (blue solid line) and with strain applied (red solid line) are simultaneously plotted in Fig. 5(a). When the strain is applied, the reflected optical power of the sensor decreases due to the change of the spectral overlap. As a result, strain sensing can be achieved by monitoring the amplitude change of each correlation peak. Furthermore, all the CFBG sensors can be discriminated due to their different positions of the peaks in the correlation spectra. The spatial resolution of the locating is determined by the full-width at half-maximum (FWHM) of the correlation peak according to 3-dB criterion. The FWHM is related to the bandwidth of the chaotic signal. The bandwidth of the spectrum-limited chaos exceeds the 1.1-GHz response bandwidth of the APD. It implies that the chaotic signal is low pass filtered by the APD. And the resolution is range-independent [15]. As a result, 2.8 cm spatial resolution is obtained through estimating the FWHM as shown in Fig. 5(b).
Fig. 5 (a) Correlation spectra free of strain (blue solid line) and with strain applied (red solid strain). (b) FWHM of the correlation spectrum with a spatial resolution of 2.8 cm.
As mentioned above, optical power of the probe light as a function of strain possesses a linear response, implying a linear amplitude response of the correlation peak as well. Strain measurement results of S11, S12, S21, and S31 are shown in Fig. 6, respectively. The RAC is utilized to estimate the strain measurement sensitivity. The sensitivity up to 0.14% RAC/με and excellent linearity of R2 ˃ 0.99 are achieved with a measurement step of 80 με. Meanwhile, we also obtain a RAC fluctuation of the CFBG sensor free of strain in our experiment, which is less than 0.013 RAC. It implies that the minimum detectable quantity is 0.013 RAC in the correlation spectrum. Therefore, the strain resolution is about 9.3 με in our experiment. It is noted that the dead zone, slightly discrepant sensitivity and dynamic range of the measurement primarily derive from the mismatch between the sensors and the multi-wavelength channels. Moreover, in order to increase the dynamic range of the measurement, the CFBG with wider reflection bandwidth should be inscribed, whereupon wider protective bandwidth should be reserved to avoid the crosstalk between the adjacent wavelength channels. Consequently, the multiplexing capacity decreases.
At last, the experiment of real-time fiber fault monitoring in the OSN is implemented. As shown in Fig. 7, three fiber breakpoints of B1, B2, and B3 are respectively introduced between S11 and S12, S21 and S23, and in front of S31. In the case of B1, the peaks of S12 and S13 disappear and a peak of the breakpoint appears in the correlation spectrum, which can be distinguished due to its different position from the sensors’. Similarly, the breakpoints of B2 and B3 can be searched out as well. Meanwhile, simultaneous three breakpoints monitoring in different sensing lines is also elucidated here. Through the peak searching algorithm, precise locating of the breakpoints can be pinpointed with a spatial resolution of 2.8 cm.
It is noted that the fiber fault occurring in one sensing line will not impede the operation in other sensing lines; and the precise pinpoint of the fiber breakpoints can be implemented without extra manpower and waiting time. Consequently, the operational reliability and survivability is promoted, and it is significant for a quasi-distributed fiber sensing network with large coverage areas, intricate fiber lines and plenty of sensors.
4. Conclusions
Large capacity quasi-distributed sensing network based on optical chaos and hybrid WDM/TDM is proposed and proof-of-concept demonstrated. Through applying cross-correlation algorithm to the chaotic reference signal and probe signal, the measurand sensing and the precise locating of each sensor can be simultaneously interrogated in the cross-correlation spectrum. Additionally, assisted with hybrid WDM/TDM technology, the multiplexing capacity could be potentially expanded to 512 with the artificially arranged OSN. A proof-of-concept experiment for the axial strain measurement with six CFBG sensors multiplexed in three sensing fiber lines is implemented. The sensitivity up to 0.14% RAC/με with excellent linearity is obtained. Meanwhile, real-time fiber fault monitoring with a high spatial resolution of 2.8 cm promotes its survivability in harsh environment. Furthermore, our sensing network has promising and widespread sensing applications of not only axial strain but also temperature, liquid level, acceleration and so on.
Acknowledgments
This work is supported by sub-Project of the Major Program of the National Natural Science Foundation of China (No. 61290315), the National Natural Science Foundation of China (No. 61275004), and the Fundamental Research Funds for the Central Universities, HUST: 2014TS042.
References and links
1. A. D. Kersey, M. A. Davis, H. J. Patrick, M. LeBlanc, K. Koo, C. Askins, M. Putnam, and E. J. Friebele, “Fiber grating sensors,” J. Lightwave Technol. 15(8), 1442–1463 (1997). [CrossRef]
2. A. Sanderson, S. Ogin, A. Crocombe, M. Gower, and R. Lee, “Use of a surface-mounted chirped fibre Bragg grating sensor to monitor delamination growth in a double-cantilever beam test,” Compos. Sci. Technol. 72(10), 1121–1126 (2012). [CrossRef]
3. Q. Sun, D. Liu, L. Xia, J. Wang, H. Liu, and P. Shum, “Experimental demonstration of multipoint temperature warning sensor using a multichannel matched fiber Bragg grating,” IEEE Photon. Technol. Lett. 20(11), 933–935 (2008). [CrossRef]
4. Y. Zhang, X. Xie, and H. Xu, “Distributed temperature sensor system based on weak reflection fiber gratings combined with WDM and OTDR,” Opto-Electron. Eng. 39(8), 69–74 (2012).
5. M. Zhang, Q. Sun, Z. Wang, X. Li, H. Liu, and D. Liu, “A large capacity sensing network with identical weak fiber Bragg gratings multiplexing,” Opt. Commun. 285(13-14), 3082–3087 (2012). [CrossRef]
6. C. Hu, H. Wen, and W. Bai, “A novel interrogation system for large scale sensing network with identical ultra-weak fiber Bragg gratings,” J. Lightwave Technol. 32(7), 1406–1411 (2014). [CrossRef]
7. X. Li, Q. Sun, D. Liu, R. Liang, J. Zhang, J. Wo, P. P. Shum, and D. Liu, “Simultaneous wavelength and frequency encoded microstructure based quasi-distributed temperature sensor,” Opt. Express 20(11), 12076–12084 (2012). [CrossRef] [PubMed]
8. Z. Luo, H. Wen, H. Guo, and M. Yang, “A time- and wavelength-division multiplexing sensor network with ultra-weak fiber Bragg gratings,” Opt. Express 21(19), 22799–22807 (2013). [CrossRef] [PubMed]
9. X. Li, Q. Sun, J. Wo, M. Zhang, and D. Liu, “Hybrid TDM/WDM-based fiber-optic sensor network for perimeter intrusion detection,” J. Lightwave Technol. 30(8), 1113–1120 (2012). [CrossRef]
10. Y. Dai, Y. Liu, J. Leng, G. Deng, and A. Asundi, “A novel time-division multiplexing fiber Bragg grating sensor interrogator for structural health monitoring,” Opt. Lasers Eng. 47(10), 1028–1033 (2009). [CrossRef]
11. W. Wang, J. Gong, B. Dong, D. Y. Wang, T. J. Shillig, and A. Wang, “A large serial time-division multiplexed fiber Bragg grating sensor network,” J. Lightwave Technol. 30(17), 2751–2756 (2012). [CrossRef]
12. A. Argyris, D. Syvridis, L. Larger, V. Annovazzi-Lodi, P. Colet, I. Fischer, J. García-Ojalvo, C. R. Mirasso, L. Pesquera, and K. A. Shore, “Chaos-based communications at high bit rates using commercial fibre-optic links,” Nature 438(7066), 343–346 (2005). [CrossRef] [PubMed]
13. M. Peil, I. Fischer, W. Elsäßer, S. Bakić, N. Damaschke, C. Tropea, S. Stry, and J. Sacher, “Rainbow refractometry with a tailored incoherent semiconductor laser source,” Appl. Phys. Lett. 89(9), 091106 (2006). [CrossRef]
14. M. Zhang, Y. Ji, Y. Zhang, Y. Wu, H. Xu, and W. Xu, “Remote radar based on chaos generation and radio over fiber,” IEEE Photon. J. 6, 1–12 (2014). [CrossRef]
15. Y. Wang, B. Wang, and A. Wang, “Chaotic correlation optical time domain reflectometer utilizing laser diode,” IEEE Photon. Technol. Lett. 20(19), 1636–1638 (2008). [CrossRef]
16. A. Wang, N. Wang, Y. Yang, B. Wang, M. Zhang, and Y. Wang, “Precise fault location in WDM-PON by utilizing wavelength tunable chaotic laser,” J. Lightwave Technol. 30(21), 3420–3426 (2012). [CrossRef]
17. L. Xia, D. Huang, J. Xu, and D. Liu, “Simultaneous and precise fault locating in WDM-PON by the generation of optical wideband chaos,” Opt. Lett. 38(19), 3762–3764 (2013). [CrossRef] [PubMed]
