Open Access
Add to BibSonomy
Abstract
This study proposes a liquid-crystal-based active wavelength filter for phase-sensitive optical time domain reflectometry to mitigate the amplified spontaneous emission (ASE) noise and accurately match the passband with the light source. The validity of the proposed system was verified using comparative experiments with conventional passive optical filters. The experiment showed an increase in signal-to-noise ratio (SNR) of up to 2.21 dB compared with passive filters. Additionally, the proposed system can effectively eliminate ASE noise, resulting in an SNR of 12.99 dB.
© 2022 Optica Publishing Group under the terms of the Optica Open Access Publishing Agreement
1. Introduction
Distributed acoustic sensing based on phase-sensitive optical time domain reflectometry (Φ-OTDR) has gained attention because of its high sensitivity, wide dynamic range, and other inherent advantages of optical fiber sensing over traditional point sensors [1–5] in various fields, such as border security [6], pipeline monitoring [7], and seismic monitoring [8].
The Φ-OTDR technology detects interference signals from Rayleigh backscattering to measure external sound or vibration signals applied at specific points of the sensing optical fiber. A highly coherent laser with a very narrow linewidth of several kilohertz (kHz) or less is used as a light source [9], and its output is pulse-modulated to localize the sensing position on the optical fiber. Additionally, pulsed light is amplified using an optical amplifier to provide a detection range of several tens of kilometers. However, amplifying a pulsed light using an optical amplifier inevitably results in amplified spontaneous emission (ASE) noise [10,11]. The ASE noise increases the background noise of pulsed light, resulting in degraded detection performance of Φ-OTDR that requires a narrow linewidth. Therefore, a narrow-band wavelength filter corresponding to the central wavelength of the light source is essential to eliminate ASE noise.
Passive filters, such as thin-film optical filters (TFF) [12,13] and athermal packaged fiber Bragg grating (FBG) [14], have been found to mitigate ASE noise in Φ-OTDR. Both filters effectively reduce the ASE noise with full width at half maximum values of about 0.8 nm (100 GHz) or less and low-temperature variability. However, due to the degradation for long-term operation, the passive filters have a disadvantage in that variation in transmission characteristics, such as reflectivity/transmissivity, center wavelength, deformation of the epoxy adhesive, and the effect of thermal coefficient may degrade the sensing signals. Another issue is that the center wavelength of the light source shifts, not only the passive filters. To address the above issue, an active wavelength filter should be used. Several schemes for active wavelength filters have been proposed, most notably the Fabry–Perot resonator [15–17], Mach–Zehnder interferometer [18], photonic bandgaps [19–21], and liquid crystal on silicon (LCoS) [22–24]. Unfortunately, some of Fabry–Perot structures based on mechanical movement have a nonlinearity issue [15–17]. Also, the interferometric structures are environmentally sensitive and may impact the performance of the sensing system [18]. The cholesteric liquid crystal-based photonic bandgaps cannot meet the requirements of the Φ-OTDR system, requiring narrow bandwidth [19–21]. Although the LCoS-based spatial light modulators offer a high spatial resolution, insensitiveness to polarization, and a high degree of flexibility, configuring the Φ-OTDR system using a two-dimensional LCoS structure is not cost effective [22–24].
This study proposes a cost effective liquid-crystal-based active wavelength filter (LC-AF) for the Φ-OTDR system to mitigate the ASE noise and accurately match the passband with the light source. The LC-AF is designed using a one-dimensional liquid crystal (LC) array structure without moving parts and has better cost performance than LCoS. The LC-AF can actively compensate for the transmitted wavelength by selectively supplying a voltage to the LC array even when the wavelength changes in the light source. Additionally, it can accurately balance the optical power of the sensing and reference signals by tuning the attenuation. Therefore, it is possible to prevent the degradation of the detection performance caused by wavelength mismatching. The validity of the proposed system was demonstrated through comparative experiments using conventional passive optical filters. In the experiment, the proposed system could effectively eliminate the ASE noise, resulting in an SNR of 12.99 dB, an improvement of 2.21 dB compared to the passive filters.
2. Principle
Figure 1 shows a schematic of an LC-AF comprising a polarization unit, a grating, and an LC array. A polarization unit, made up of a collimating lens, a polarization retarder, a half-wave plate, and an imaging lens, separates randomly polarized light signals into two orthogonal polarizations. Light is split into perpendicular and horizontal polarizations to the optical axis as it passes through the polarization retarder composed of yttrium orthovanadate via the collimating lens. Additionally, the half-wave plate converts the horizontal polarization to perpendicular polarization. The output lights are then directed into the grating element, which spreads the spectrum angularly to the imaging lens. Subsequently, the dispersed lights are incident on the LC array. The LC array is composed of an array of 1 × 384 nematic LC cells with a zero-twisted configuration, a polarizer, and a reflective mirror with a reflectivity of 80%. The horizontal positions of the cells along the LC array can be matched to the wavelength of reflected light in the range of 1527.60 to 1565.50 nm. The intensity of reflected light for each wavelength can be modulated by the electrically controlled birefringence (ECB) effect of each individual cell as shown in Fig. 2. The ECB effect is based on the changes in the LC molecules along the direction of the electric field according to the applied voltage [25,26]. In the on-state with applied voltage, the electric field induces a tilt of the LC molecules, causing a change in birefringence of the cell. On the other hand, in the off-state without applied voltage, LC cell behaves like a half-wave plate, and thus light is blocked. Accordingly, the LC-AF can implement various forms of spectral modulation by selectively applying a voltage to the LC array.
Fig. 1. Schematic of an active wavelength filter based on LC array. CL: collimating lens, PR: polarization retarder, HWP: half-wave plate, L: imaging lens, LC: liquid crystal.
Fig. 2. Schematic of the LC array as the electrically controlled birefringence cells. V: voltage, E: electric field.
The lights reflected from the LC array are directed to the optical fiber output port through the same path as the incident path. The intensity of light transmitting LC-AF is given as follows:
In this study, the LC-AF was used as an optical bandpass filter, which is an essential component for eliminating ASE noise in the Φ-OTDR system. Since ASE noise increases the background noise of amplified light and degrades the detection performance, a narrow optical bandpass filter is required. The proposed LC-AF could be used as a narrow bandpass filter that transmits a wavelength by selecting one LC from the LC array. Additionally, the LC-AF can be compensated for by selectively applying a voltage to the LC array corresponding to the wavelength changes. Accordingly, the wavelength transmitted through LC-AF can actively compensate even for the wavelength changes of the light source.
3. Experimental setup
Figure 3 shows the experimental setup of the proposed Φ-OTDR with LC-AF. A two-channel LC-AF with a spectral range of 1527.60 to 1565.50 nm, 3-dB bandwidth of 0.075 to 38 nm, attenuation range of 0 to 20 dB, and insertion loss of 3.5 dB was used in the experiment. An external-cavity laser (Rio, ORION) with a center wavelength of 1552.25 nm, linewidth of 5.3 kHz, and output power of 13.4 mW was used as the light source with polarization-maintaining (PM) output. An optical PM coupler with a 90:10 ratio was used to divide the laser output into 90% probe light and 10% reference light. The 90% light (i.e., probe light) was modulated by a semiconductor optical amplifier (Thorlabs, BOA1004P) to generate the pulsed probe wave with a pulse width of 100 ns (allowing for 10 m spatial resolution) and a repetition rate of 1 kHz. Furthermore, an acoustic-optic frequency shifter (G&H, T-M080-0.4C2J-3-F2S) introduced an 80 MHz optical frequency shift into the probe pulse. The optical power of the probe pulse was amplified by an erbium-doped fiber amplifier (Taclink, WZEDFA). At this time, ASE noise was generated and removed by the first channel of LC-AF, which selects the 247th LC corresponding to the 1552.240 nm with a 3-dB bandwidth of 0.075 nm. The probe pulse was then launched into the sensing fiber via an optical circulator and generated Rayleigh backscattering signals at each point along the sensing fiber. Additionally, the backscattered light was detected by a balanced photodetector (BPD, Luna, PBPD001) after it was combined with the reference light at the 50:50 optical PM coupler. Before being combined with the probe light, the reference light passes through the second channel of LC-AF to match the optical power of the probe and reference signals for the interference of Rayleigh backscattering light. The two outputs of the BPD were sampled by the field-programmable gate array-based signal processing board with a 16-bit 100 MHz sampling rate. The AC term of the BPD output (${\boldsymbol{E}_{\boldsymbol{BPD}}}(\boldsymbol{t} )$) in a radio frequency band, known as a beat signal, can be expressed as follows:
Fig. 3. Experimental setup of the proposed Φ-OTDR with LC-AF. ECL: external-cavity laser; SOA: semiconductor optical amplifier; AFS: acousto-optic frequency shifter; EDFA: erbium-doped fiber amplifier; LC-AF: liquid-crystal-based active filter; RB: Rayleigh backscattering; BPD: balanced photodetector.
From Eq. (4), the amplitude (${A_{probe}}{A_{ref}}$) and phase ($\phi (t )$) of the Rayleigh backscattering trace can be obtained from the following:
A single-mode optical fiber of about 4,100 m in length was used as the sensing fiber in the experimental setup. A cylindrical fiber stretcher based on a lead zirconate titanate (PZT), which was wound with a 10-m length optical fiber, provides the external vibration at about 2,100-m location of the sensing fiber. A sinusoidal signal with an amplitude of 3 Vpp, an offset voltage of 1.5 V, and a frequency of 100 Hz was applied to the PZT fiber stretcher.
Figure 4 shows an example of a measured Rayleigh backscattering trace of the proposed Φ-OTDR under deterioration where the central wavelength of the light source or filter shifts. Figure 4 (a) shows the Rayleigh backscattering trace in a normal state, while Fig. 4 (b)–(d) show the deteriorated states. As illustrated in Fig. 4 (b) and (c), where the central wavelengths of the laser source and passive filter are changed due to the degradation, the amplitudes are decreased due to the mismatch of center wavelengths between the laser source and the filter. However, the proposed Φ-OTDR system can compensate for the transmitted wavelength of the LC-AF through a feedback loop to avoid performance degradation due to wavelength shifts of a light source, as shown in Fig. 4 (d). The LC-AF is controlled for each LC cell through serial communication with the signal processing board. When the power of the Rayleigh backscattering trace decreases below the threshold due to the wavelength change, the LC-AF performs scanning for 1 to 384 LC cells. The start/end cell for performing scanning in the LC-AF can be set within a range of 1 to 384 cells. The signal processing board detects the maximum value from the Rayleigh backscattering trace measured through the scanning and compensates for the transmission wavelength by reconfiguring the on/off-states of the LC array.
Fig. 4. Example of Rayleigh backscattering trace (a) in the normal state, under degradation, (b) at the light source, (c) at the passive filter, and (d) with LC-AF.
4. Results
Comparative experiments were conducted without filters and with TFF, FBG, and LC-AF to confirm the performance of the proposed Φ-OTDR system. The TFF has a channel spacing of 100 GHz (0.8 nm). The FBG has a reflectivity of 99.08% and a temperature sensitivity of 0.67 pm/°C. Figure 5 shows the spectral outputs of various filters, which were measured using a broadband light source (Luxpert, LXASE) and an optical spectrum analyzer (Yokogawa, AQ6319) with a wavelength resolution of 0.01 nm. The TFF showed a low insertion loss of 0.88 dBm at 1,552.25 nm and a 3-dB bandwidth of 0.69 nm. For the FBG, the insertion loss was 2.47 dBm, and the 3-dB bandwidth was 0.252 nm. However, since the FBG was used with an optical circulator, the insertion loss of 1.98 dBm using the optical circulator should be considered. The LC-AF showed an insertion loss of 4.14 dBm and a 3-dB bandwidth of 0.075 nm. Among the filters, the LC-AF showed the narrowest 3-dB bandwidth but had the largest insertion loss. Since LC-AF was mainly used in reconfigurable optical add/drop multiplexers of optical networks requiring rough insertion losses, optical configurations were not optimized to reduce the insertion loss. Considering the reflectivity of LC array and losses from other optical components of LC-AF, insertion loss is expected to be reduced to at least 2.0 dB, suggesting a potential improvement in its performance.
Figure 6 shows the measured Rayleigh backscattering trace along the sensing fiber obtained without filter and with TFF, FBG, and LC-AF when the external vibration is located close to 2,100 m. Figure 6 (a) and (b) represent the original and normalized Rayleigh backscattering traces, respectively. The Rayleigh backscattering traces were measured up to a distance of 4,100 m, which is the same length as the sensing fiber used in the experiment. The peak signals at 3,098 and 4,081 m in Fig. 6 (a) and (b) are due to Fresnel reflection at the splicing point and end of the sensing fiber, respectively. In Fig. 6 (a), the mean values of the intensity without filter and with TFF, FBG, and LC-AF were 8,516, 7,935, 5,992, and 4,172 arbitrary units, respectively. The intensity of the Rayleigh backscattering trace measured by the LC-AF was reduced by about 48.9% compared with those without filters, which is related to the insertion loss characteristics of the filters, as shown in Fig. 5. The moving differential method [29] was used to locate the external vibration applied by the PZT fiber stretcher. Figure 6 (c) and (d) show the localization results. To compare the SNR, which is the ratio between the signal and background noise peak intensities ($20 \times {\log _{10}}({V_{peak\; signal}}/{V_{peak\; noise}})$), the original data were normalized and shown in Fig. 6 (d). When the filters, LC-AF, FBG, and TFF, were used to eliminate the ASE noise, the applied external vibration signal could be identified at 2107 m; otherwise, noise peak signals appeared at the 772 and 4,079 m, as shown in Fig. 6 (c). In Fig. 6 (d), the SNRs without filter and with TFF, FBG, and LC-AF was 1.51, 8.29, 10.78, and 12.99 dB, respectively. This implies that the proposed Φ-OTDR with LC-AF can ultimately improve the SNR even though it undergoes relatively higher insertion loss.
Fig. 6. Rayleigh backscattering trace comparison with different filter types. (a) and (b) show the original and normalized signals, respectively, whereas (c) shows the localization of external vibration and (d) its normalized signals.
To achieve a high SNR in the Φ-OTDR system, it is important to minimize the fluctuations in the Rayleigh backscattering trace. Figure 7 shows the 100 consecutive superimposed traces of the Rayleigh backscattering original signal obtained without filter and with TFF, FBG, and LC-AF to investigate the fluctuations of the Rayleigh backscattering trace. No external vibration was applied, and a range of 2,000 to 2,400 m was selected to indicate the variation of the Rayleigh backscattering trace. To compare the variance of Rayleigh backscattering traces, the mean and standard deviation were determined at each point of the sensing fiber, and the relative standard deviation was calculated by dividing the standard deviation by the mean. Figure 7 shows the relative standard deviation values for each case below the Rayleigh backscattering trace. The relative standard deviation without filter, TFF, FBG, and LC-AF mean values were 0.12076, 0.0808, 0.06677, and 0.05893, respectively. Among the comparisons, the Rayleigh backscattering trace fluctuation was the least in the LC-AF with the narrowest 3-dB bandwidth. This implies that the effect of ASE noise in the Φ-OTDR system can be efficiently removed as the 3-dB bandwidth of the filter becomes narrower.
Fig. 7. Consecutive superimposed Rayleigh backscattering traces and its relative standard deviation with different filter types. (a) without filter, (b) TFF, (c) FBG, and (d) LC-AF.
5. Conclusions
In this study, the LC-AF was used to reduce the temporal degradation of detection performance caused by the mismatching of center wavelengths between the laser source and the filter in the Φ-OTDR system. LC-AF can actively compensate for the transmitted wavelength even when the wavelength changes in the light source through a feedback loop. The validity of the proposed system was determined through comparative experiments using conventional optical passive filters, TFF and FBG. According to the experimental results, LC-AF with a narrow 3-dB bandwidth effectively eliminated the ASE noise, resulting in an SNR of 12.99 dB and an SNR improvement of 2.31 dB compared to the passive filters. This technique would be effective in constructing a robust Φ-OTDR system that can be long-term operated in a harsh environment with a high chance of wavelength shifts and fluctuations in either the light source or its filter.
Future studies will focus on field application tests using the proposed Φ-OTDR system, such as detecting leaks in underground water pipelines and measuring resonant frequencies in high-voltage direct-current modular multilevel converter systems, requiring high SNR within a sensing range of 10 km. Furthermore, long-range measurements will be performed through the improvement of the insertion loss of the LC-AF.
Funding
Korea Institute of Energy Technology Evaluation and Planning; Ministry of Trade, Industry and Energy (No. 2017931010060); Korea Water Resources Corporation (20-A-T-002).
Disclosures
The authors declare no conflicts of interest.
Data availability
Data underlying the results presented in this paper are not publicly available at this time but may be obtained from the authors upon reasonable request.
References
1. Z. He and Q. Liu, “Optical fiber distributed acoustic sensors: A review,” J. Lightwave Technol. 39(12), 3671–3686 (2021). [CrossRef]
2. Y. Wang, H. Yuan, X. Liu, Q. Bai, H. Zhang, Y. Gao, and B. Jin, “A comprehensive study of optical fiber acoustic sensing,” IEEE Access 7, 85821–85837 (2019). [CrossRef]
3. H. M. Ogden, M. J. Murray, J. B. Murray, C. Kirkendall, and B. Redding, “Frequency multiplexed coherent φ-OTDR,” Sci. Rep. 11(1), 17921 (2021). [CrossRef]
4. N. Rouibah, L. Barazane, M. Benghanem, and A. Mellit, “IoT-based low-cost prototype for online monitoring of maximum output power of domestic photovoltaic systems,” ETRI J. 43(3), 459–470 (2021). [CrossRef]
5. J. D. Jeong, J. Han, I. W. Lee, and J. W. Chong, “Implementation of cost-effective wireless photovoltaic monitoring module at panel level,” ETRI J. 40(5), 664–676 (2018). [CrossRef]
6. F. Peng, H. Wu, X. H. Jia, Y. J. Rao, Z. N. Wang, and Z. P. Peng, “Ultra-long high-sensitivity phi-OTDR for high spatial resolution intrusion detection of pipelines,” Opt. Express 22(11), 13804–13810 (2014). [CrossRef]
7. J. Zuo, Y. Zhang, H. Xu, X. Zhu, Z. Zhao, X. Wei, and X. Wang, “Pipeline leak detection technology based on distributed optical fiber acoustic sensing system,” IEEE Access 8, 30789–30796 (2020). [CrossRef]
8. M. R. Fernández-Ruiz, H. F. Martins, E. F. Williams, C. Becerril, R. Magalhães, L. Costa, S. Martin-Lopez, Z. Jia, Z. Zhan, and M. González-Herráez, “Seismic monitoring with distributed acoustic sensing from the near-surface to the deep oceans,” J. Lightwave Technol. 40(5), 1453–1463 (2022). [CrossRef]
9. J. Li, Z. Zhang, J. Gan, Z. Zhang, X. Heng, K. Zhou, H. Zhao, S. Xu, and Z. Yang, “Influence of laser linewidth on phase-OTDR system based on heterodyne detection,” J. Lightwave Technol. 37(11), 2641–2647 (2019). [CrossRef]
10. J. Huang, “Impact of ASE noise in WDM systems,” Optik 122(15), 1376–1380 (2011). [CrossRef]
11. P. Muniz-Cánovas, Y. O. Barmenkov, A. V. Kir’yanov, J. L. Cruz, and M. V. Andrés, “ASE narrow-band noise pulsing in erbium-doped fiber amplifier and its effect on self-phase modulation,” Opt. Express 27(6), 8520–8528 (2019). [CrossRef]
12. Y. Muanenda, S. Faralli, C. J. Oton, and F. Di Pasquale, “Dynamic phase extraction in a modulated double-pulse ϕ-OTDR sensor using a stable homodyne demodulation in direct detection,” Opt. Express 26(2), 687–701 (2018). [CrossRef]
13. H. Wu, C. Shang, K. Zhu, and C. Lu, “Vibration detection in distributed acoustic sensor with threshold-based technique: a statistical view and analysis,” J. Lightwave Technol. 39(12), 4082–4093 (2021). [CrossRef]
14. H. He, L. Shao, H. Li, W. Pan, B. Luo, X. Zou, and L. Yan, “SNR enhancement in phase-sensitive OTDR with adaptive 2-D bilateral filtering algorithm,” IEEE Photonics J. 9(6), 1–10 (2017). [CrossRef]
15. C. Williams, N. Hong, M. Julian, S. Borg, and H. J. Kim, “Tunable mid-wave infrared Fabry-Perot bandpass filters using phase-change GeSbTe,” Opt. Express 28(7), 10583–10594 (2020). [CrossRef]
16. J. S. Milne, J. M. Dell, A. J. Keating, and L. Faraone, “Widely tunable MEMS-based Fabry–Perot filter,” J. Microelectromech. Syst. 18(4), 905–913 (2009). [CrossRef]
17. H. Kim and M. Song, “A fiber laser spectrometer demodulation of fiber Bragg grating sensors for measurement linearity enhancement,” J. Opt. Soc. Korea 17(4), 312–316 (2013). [CrossRef]
18. J. Mora, B. Ortega, A. Díez, J. L. Cruz, M. V. Andrés, J. Capmany, and D. Pastor, “Photonic microwave tunable single-bandpass filter based on a Mach-Zehnder interferometer,” J. Lightwave Technol. 24(7), 2500–2509 (2006). [CrossRef]
19. Y. Huang and S. Zhang, “Optical filter with tunable wavelength and bandwidth based on cholesteric liquid crystals,” Opt. Lett. 36(23), 4563–4565 (2011). [CrossRef]
20. C. Sun and J. Lu, “Optical filters based on cholesteric, blue and sphere mesophases,” Polymers 14(22), 4898 (2022). [CrossRef]
21. A. M. Labeeb, Y. A. Ward, and M. Fikry, “Thermal control of tunable photonic optical bandgaps in different cholesteric liquid crystals mixtures,” J. Mol. Liq. 340, 117179 (2021). [CrossRef]
22. Z. Zhang, Z. You, and D. Chu, “Fundamentals of phase-only liquid crystal on silicon (LCOS) devices,” Light: Sci. Appl. 3(10), e213 (2014). [CrossRef]
23. M. Wang, L. Zong, L. Mao, A. Marquez, Y. Ye, H. Zhao, and F. Vaquero Caballero, “LCoS SLM study and its application in wavelength selective switch,” Photonics 4(4), 22 (2017). [CrossRef]
24. J. Liu and J. Wang, “Demonstration of polarization-insensitive spatial light modulation using a single polarization-sensitive spatial light modulator,” Sci. Rep. 5(1), 1–7 (2015). [CrossRef]
25. R. A. Soref and M. J. Rafuse, “Electrically controlled birefringence of thin nematic films,” J. Appl. Phys. 43(5), 2029–2037 (1972). [CrossRef]
26. G. Labrunie and J. Robert, “Transient behavior of the electrically controlled birefringence in a nematic liquid crystal,” J. Appl. Phys. 44(11), 4869–4874 (1973). [CrossRef]
27. M. Ren, P. Lu, L. Chen, and X. Bao, “Theoretical and experimental analysis of $\varPhi$-OTDR based on polarization diversity detection,” IEEE Photonics Technol. Lett. 28(6), 697–700 (2016). [CrossRef]
28. H. Gabai and A. Eyal, “On the sensitivity of distributed acoustic sensing,” Opt. Lett. 41(24), 5648–5651 (2016). [CrossRef]
29. Y. Lu, T. Zhu, L. Chen, and X. Bao, “Distributed vibration sensor based on coherent detection of phase-OTDR,” J. Lightwave Technol. 28(22), 3243–3249 (2010). [CrossRef]
