Open Access
Add to BibSonomy
Abstract
In this letter, a distributed optical fiber hydrophone (DOFH) based on Φ-OTDR is demonstrated and tested in the field. The specially designed sensitized optical cable with sensitivity up to −146 dB rad/µPa/m is introduced, and an array signal processing model for DOFH is constructed to analyze the equivalence and specificity of the distributed array of acoustic sensors. In the field test, a 104-meter-long optical cable and a Φ-OTDR system based on heterodyne coherent detection (Het Φ-OTDR) is utilized, and underwater acoustic signal spatial spectrum estimation, beamforming and motion trajectory tracking with high accuracy can be realized. As far as we know, this is the first report on the field trial of DOFH based on Φ-OTDR. The DOFH has the potential to achieve an array range of tens of kilometers, with elements spaced up to the meter level and flexible configuration, which has a broad application prospect for marine acoustic detection.
© 2021 Optical Society of America under the terms of the OSA Open Access Publishing Agreement
1. Introduction
Marine acoustic monitoring is of great significance to national underwater military defense and marine mineral resources exploration [1,2]. Sonar (or hydrophone) is one of the most common techniques for marine acoustic monitoring, and has been developed unprecedentedly. Especially, due to the advantages, such as anti-electromagnetic interference, small volume, light weight, no electricity devices under water and so on, optical fiber hydrophone has become a research hotspot of the next generation hydrophone [3,4].
So far, the mostly used optical fiber hydrophone is based on interferometer structure, such as Michelson, Mach-Zehnder and Sagnac interferometer, which is a kind of point sensor with tens to a hundred meters of fiber wrapped around a mandrel to achieve the required responsivity [3,4]. The hydrophone needs to work in the form of an array, and the quasi-distributed hydrophone array is formed through time division multiplexing (TDM), wavelength division multiplexing (WDM) and space division multiplexing (SDM) technology, which makes the whole system structure complex, the array volume large and the cost expensive. The number of the elements in an array can only reach about one hundred [5,6]. In addition, the array element spacing is fixed, while a higher-frequency application needs a much smaller one. Therefore, the current interferometric optical fiber hydrophone cannot meet the trend of pursuing a larger array aperture and a more flexible frequency range in the field of marine acoustic detection.
Phase-sensitive optical time domain reflectometer (Φ-OTDR) is a new kind of distributed optical fiber acoustic sensor based on the interference effect of Rayleigh backscattering (RBS) in optical fiber [7–9]. In addition to the advantages of the optical fiber sensor, Φ-OTDR also has the unprecedented properties of long sensing range, high spatial resolution, and so on. State-of-the-art Φ-OTDR system can monitor vibrations with spatial resolution accurate to meter scale and sampling rate up to tens of kHz over distance of several kilometers [10]. Besides, the spatial resolution can be set flexibly on the interrogator as needed. It has been widely used in many important fields, including intrusion detection [11,12], pipeline security [13], as well as railway transportation [14,15]. Due to its unique advantages, Φ-OTDR has a promising application prospect in marine acoustic monitoring to realize large array scale with better flexibility and ease of implementation. It can provide an excellent solution for the improvement of array scale and the lightening of array. However, different from the application scenarios on land, the acoustic pressure sensitivity of conventional sensing optical cable in water is relatively low. The sensitivity of a single-mode fiber is about −212dB rad/µPa/m, and the steel tube protection structure of the fiber in the cable further reduces the sensitivity of the common communication cable. Besides, different from point sensors, the measurement unit of Φ-OTDR has a certain spatial aperture, i.e., gauge length, which may have an influence on the directivity and correlation of Φ-OTDR to acoustic field response. So far, there is no corresponding complete theory of array signal processing method for distributed optical fiber array.
In this letter, a distributed optical fiber hydrophone (DOFH) based on Φ-OTDR is demonstrated and tested in the field. A new type of sensitized optical cable is introduced. The sensitized optical cable consists of supporting mandrel, special optical fiber and cable sheath. The supporting mandrel of the optical cable is made of sound sensitive material, on which the bending resistant fiber is tightly wound. A layer of sound permeable tube is extruded for protection and waterproof and to reduce the loss of sound transmission. The sensitivity of the cable can reach about −146dB rad/µPa/m. An array signal processing model for DOFH is constructed, to analyze the equivalence and specificity of acoustic wave response using the Φ-OTDR as a distributed array of acoustic sensors. With the influence of the array element aperture, there will exist response inconsistency with the direction of acoustic source. For certain incident direction, the spatial correlation of Φ-OTDR will not change through the integral of acoustic wave within the array element aperture for far-field acoustic source.
In the field test, a Φ-OTDR system based on heterodyne coherent detection (Het Φ-OTDR) is utilized [7], the 104-meter-long optical cable is deployed in the lake. Through array signal processing, underwater acoustic signal source signal orientation and motion trajectory tracking can be realized accurately. As far as we know, this is the first report on the field trial of DOFH based on Φ-OTDR. The results prove the feasibility for Φ-OTDR to be used as a DOFH to make up the deficiencies of the current hydrophone.
2. Φ-OTDR system
In the early stages, Φ-OTDR was based on direct intensity detection [11], which can be used for event detection, but quantitative measurements cannot be achieved because the intensity response to external disturbances is nonlinear. In recent years, the utilization of the optical phase of RBS has attracted much attention for its linear relationship with the vibration [7,8,16–18]. Het Φ-OTDR was the first phase-demodulated Φ-OTDR system scheme based on heterodyne coherent detection, and has been widely adopted due to its properties of relatively high signal-to-noise ratio (SNR) and flexibility in defining differential length in the digital domain [7]. The schematic of Het Φ-OTDR is shown in Fig. 1. The continuous-wave light from a narrow-linewidth (<3kHz) laser is split into probe light and local reference light by an fiber-optic coupler (OC). Then, an acousto-optic modulator (AOM) is used to chop the probe light into optical pulse with pulse width of 100ns, and generates a frequency shift of 160MHz. The probe pulses are amplified by an erbium-doped optical fiber amplifier (EDFA) and then injected into the fiber under test (FUT) through a circulator (CIR). The pulse repetition rate is 5kHz. RBS from FUT mixes with local reference light, and then will be detected by a balanced photodetector (BPD). The alternating current (AC) signal output by the BPD is collected by the data acquisition board (DAQ) with sampling rate of 500MS/s, and finally, the data is dealt with by a signal processor.
After coherent detection, the signal obtained by BPD is then expressed as
where ${E_B}$ is the amplitude of backscattered light, ${E_L}$ is the amplitude of local reference light, $\Delta \omega$ is the frequency of AOM. The transmission phase can be described as ${\phi _z} = 2\int_0^{u({t_z} - T)/2} {\beta (z )\textrm{ }} dz = 2\int_0^{u({t_z} - T)/2} {\textrm{ }({\overline \beta + \Delta \beta (z )} )} \textrm{ }dz$, $\overline \beta = 2\pi n/\lambda$ is the mean of transmission function, $T$ is the pulse temporal width and ${t_z}$ is the round-trip time. The coherent Rayleigh phase noise can be described as ${\phi _{Ray}}\textrm{ = }\Delta \beta (z )uT/2$, where u is the velocity of light in the fiber, $\Delta \beta = \omega \cdot dn/c + n \cdot d\omega /c$ is the variation of propagation function, $\Delta n = n(z) - \overline n$ is the difference in refractive index in space and $d\omega$ is the frequency shift in time domain caused by the laser source.To eliminate the influence of interference fading and to further reduce the noise floor and narrow the gap with interferometer hydrophone in noise floor, frequency diversity and aggregation scheme is introduced in the system [19,20]. As shown in Fig. 2, a phase modulator (PM) is added before the AOM, and the phase of probe beam is modulated by a sinusoidal signal of 20MHz to generate harmonic frequencies.
The optical field after PM is expressed as
Then, the signal obtained by BPD can be re-written as
The backscattered traces of harmonics at different frequencies have low correlation, interference fading won’t occur at the same time, and can be eliminated by optimum selection, direct averaging, or weighted averaging method [20]. Besides, the sensing response for different harmonics is highly coherent, whereas the additive measurement noise, including shot noise, thermal noise, and so on, is Gaussian random variables and irrelevant. Then, aggregation results of several harmonics will help reduce the noise floor. Here, vectors are rebuilt in the process of I/Q demodulation [7], and rotated-vector-sum method [21]is used to realize the aggregation of 5 harmonics. Data recorded in the laboratory using a coil of fiber placed inside an isolation chamber are shown in Fig. 3. An 625Hz sinusoidal signal is applied on PZT at position 930m. It’s obvious that phase errors induced by fading points are eliminated after frequency aggregation, as shown in Fig. 3(a). And according to the power spectral density (PSD) of differential phase in Fig. 3(b), the noise floor of aggregation results is about 7dB lower than that of a single harmonic, which is of great importance for the detection of smaller acoustic signals.
Fig. 3. (a) The differential phase distribution along the distance; (b) The PSD at the disturbance location.
3. Sensitized optical cable
Due to its low sound pressure sensitivity, ordinary single-mode optical cable cannot meet the requirements of underwater acoustic detection. It is necessary to take measures to enhance the sensitivity of the sensing optical cable. Then, a new type of sensitized optical cable is designed, which consists of supporting mandrel, special bending resistant fiber and cable sheath, shown as Fig. 4. The supporting mandrel of the optical cable is made of sound sensitive material, on which the bending resistant fiber is tightly wound. When a region on the cable is excited by underwater sound, the internal support mandrel is subjected to radial deformation, and the length of the winding fiber changes accordingly, resulting in the phase modulation of Rayleigh scattered light in the region. Since the Young's modulus of supporting mandrel material is much smaller than that of optical fiber, the sensitivity of sound pressure in this method can be greatly improved. In order to protect the wound fiber and ensure the acoustic coupling efficiency, the outermost layer of the cable is extruded with a layer of polyurethane cable sheath. The ultimate diameter of the cable is only about 12.5mm, the length of fiber evenly wound on each meter of cable is about 7.5m, and the picture of sensitized optical cable is shown in Fig. 5.
Assume that the length of the fiber wound on cable of length ${L_c}$ is L, the phase change of the round-trip transmission light caused by the change of fiber length can be expressed as $\Delta \varphi \textrm{ = 2}\beta \cdot \Delta L$, and the acoustic pressure sensitivity of the cable can be obtained as follows:
To quantify the sensitivity of the cable, measurement is carried out in the frequency range of 20–500Hz. The diagram and picture of the experiment is shown in Fig. 7. The 5-m-long cable is wound in a loop and placed at the same position as the piezoelectric (PZT) hydrophone in the standing wave tube. The PZT hydrophone that is the calibrated standard hydrophone is utilized as the reference for the calibration of sensitized optical cable. A standard sound source is placed at the bottom of standing wave tube, and is driven by a signal generator. The acoustic pressure signal acted on the PZT hydrophone is consistent with that on the sensitized optical cable. According to the phase change value demodulated by Φ-OTDR system and the sound pressure detected by PZT hydrophone, the sensitivity of the sensitized optical cable can be obtained, and the experimental results are shown in Fig. 8. Obviously, within the range of 20Hz to 500Hz, the frequency response consistency is good with a standard deviation of 0.88dB, and the average acoustic pressure sensitivity is as high as −146dB rad/µPa/m, which is basically consistent with the theoretical value and can fully meet the requirements of underwater acoustic detection. Besides, we put the cable in a water pressure tank with the hydrostatic pressure of no less than 3MPa, and apply lateral pressure on the cable with about 1000N/100mm, neither of which will change the loss characteristics of the cable, and the structure of the cable is intact and its performance has not deteriorated.
4. Array signal processing model for DOFH
The Φ-OTDR transforms an optical fiber into a distributed array of acoustic sensors, and considering the difference between the equivalent sensing array and point measurement array, a distributed sensing array model is essential to be redefined, including sensing element, array element spacing, array aperture, etc. In our previous work, a distributed array model in the air is introduced [24,25], but no directional response properties are analyzed and the structure of sensitized optical cable should be considered here. As shown in Fig. 9, the detected signal of each sensing channel of Φ-OTDR is obtained by spatial phase difference between two adjacent positions ${x_{i,0}}$ and ${x_{i,1}}$. The differential fiber length $\Delta L$ is gauge length, the corresponding spatial scale $\Delta x = {x_{i,1}} - {x_{i,0}}$ along the cable axis is termed as array element aperture. Array element spacing is defined as the distance between initial positions of two adjacent sensing channels, $d = {x_{i,0}} - {x_{i - 1,0}}$. Besides, it is worth mentioning that there may be overlap between two adjacent array elements as long as there is no overlap between the pulse widths of the front and rear differential points in the two elements on basis of consideration of spatial correlation [26]. Finally, array aperture is defined as the distance from the beginning of the first element to the end of the final element, $L = {x_{N,1}} - {x_{1,0}}$. N is the number of array elements.
Different from point measurement arrays, array element spacing $d$, array element number $N$ and array aperture $L$ of Φ-OTDR tend to be more flexible, because these parameters can be set up in the digital domain, and several different array arrangements can be conveniently achieved in one measurement. Meanwhile, some limitations need to be considered in the array arrangement. Especially, the influence of the integral of acoustic signals within the array element aperture should be analyzed. Take the far-field acoustic source signal which is the most common in underwater acoustic applications as an example [27], a far-field acoustic source propagating at angle $\alpha $ to the axis of cable expressed as $\varepsilon (x,t) = exp (j\omega t - jkx\cos \alpha )$ is imposed on the array element. The detected signal of i-th equivalent array element with array element aperture $\Delta x$ is expressed as,
Fig. 10. The response directivity of distributed equivalent array elements with different acoustic frequencies and array element apertures. (a) 375 Hz and 1 m; (b) 375 Hz and 2 m; (c) 375 Hz and 3 m; (d) 625 Hz and 1 m; (e) 625 Hz and 2 m; (b) 625 Hz and 3 m.
According to Eq. (4), when the incident direction is fixed, the amplitude response of each array element is basically the same, as the angle for each array element remains unchanged in the far field condition [25], and the phase of detected signal is linearly related to the array element position, and the spatial correlation will not change through the integral of acoustic wave, which can satisfy the demand of array signal processing as long as the amplitude response is not zero at the incident direction.
5. Results of field test
The field test is conducted in a reservoir in Zhejiang Province, and the setup of the field test is shown in Fig. 11. The Φ-OTDR system is placed in the platform by the lake, and a section of sensitized optical cable with a length of 104m is placed in the lake. The sensitized optical cable is bound to a floating ball and a 6kg weight block for every 15m and suspended in a water depth of 7m. By observing and adjusting the floating ball position to ensure that the cable is straight. A section of guiding optical fiber of approximately tens of meters in length is used to connect the Φ-OTDR system to the sensitized optical cable in the lake. There are about 200m redundant fibers at the end of the sensitized optical cable. In the test, the electric ship tows the sound source system powered by UPS to several different positions respectively and anchors, and by adjusting the sound source to produce sound signals of different frequencies for beamforming test. Moreover, the position of the electric ship can be calibrated by telescope so as to compare with the experimental results.
As shown in Fig. 12, in the test, the acoustic source generates signals of 375Hz and 625Hz at three positions respectively, and the relative positions of the sensitized optical cable and the sound source were determined through the two-point positioning of the ranging telescope. Even if the ship is anchored, it would drift slowly across the water, so we corrected its position before each test.
Fig. 12. The location map of acoustic source and sensitized optical cable. The black line segment is the optical cable used for spatial spectrum estimation :(a) 375 Hz;(b) 625 Hz.
The actual azimuth angle of acoustic source is known, and according to the analysis and simulation in section 4, the acoustic source is actually located in the direction with higher amplitude response coefficient. The acoustic wavelength for 375 Hz and 625 Hz signal are 4 m and 2.4 m separately. The array element aperture is taken to be 3.5 m, and the corresponding fiber differential length is about 26 m, according to the ratio of the winding fiber length to the cable 7.5. Then, the array element aperture is about 3.5 m. For 375 Hz acoustic signal, array element spacing is set as 1.34 m, and pulse width is 100ns, and the number of array elements is 20. For 625 Hz acoustic signal, array element spacing is set as 0.67 m, and pulse width is 50ns, and the number of array elements is 40. The segment of sensitized optical cable used for spatial spectrum estimation is counted after 50 m of the cable proximal end, which is shown in Fig. 12 as the black line. After building the distributed array model, spatial spectrum estimation is conducted and the classical multiple signal classification (MUSIC) method [29] is utilized, and the estimated azimuth of sound source is located at the peak of spatial spectrum curve.
Figure 13 shows the result of spatial spectrum estimation when the acoustic source is at different positions with frequency of 375 Hz and 625 Hz. For 375 Hz acoustic signal, the results are shown in Fig. 13(a)–(c) with different positions 1–3. Similarly, the results of 625 Hz signal are shown in Fig. 13(d)–(f) with different positions 1–3 respectively. This result verifies that the DOFH system under test can generate sufficient array gain and effectively detect directional underwater targets. In terms of positioning accuracy, the comparison between the actual azimuth angle obtained by two-point positioning method with ranging telescope and the estimated azimuth angle through MUSIC method is given in Table 1. The accuracy of the estimated azimuth angle is high, and the error with angle measured by the telescope is within 10°.
Fig. 13. Spatial spectrum estimation results. (a)–(c): 375 Hz with positions 1–3; (d)–(f):625 Hz with positions 1–3.
Table 1. The comparison between the actual azimuth angle and the estimated azimuth angle
After the completion of space spectrum estimation, beamforming technology, also known as spatial filtering technology, can be used to realize directional enhancement of acoustic signals, and can significantly suppress the random noise introduced by the system and the environment. Then, the SNR of the system can be greatly improved. The process of beamforming is shown in Fig. 14. Take the 375 Hz signal at position 3 as an example, the estimated azimuth angle ${\theta _d}$ is 121° , the number of array elements M is 20, and array element spacing $d$ is 1.34 m. For uniform linear array, the weight coefficient can be obtained by ${W_i} = {e^{ - j({i - 1} )\beta ({{\theta_d}} )}}\textrm{/}M$, where ${\beta _d} = 2\pi d\cos ({{\theta_d}} )/\lambda$ and $\lambda$ is wavelength of the underwater acoustic signal. Then, the results of beamforming can be expressed as $Y\textrm{ = }\sum\limits_{i = 1}^M {{W_i}{S_i}(t )}$. The power spectral density (PSD) of differential phase with and without beamforming is shown in Fig. 15, and it’s obvious that the noise level after beamforming is suppressed by about 13 dB, while the acoustic signal of 375 Hz remains stable, and the SNR will be improved accordingly, which is consistent with the theory.
When the electric ship returns, the sound source remains audible until the UPS power is exhausted, during which time the system continues to collect acoustic data and calculate the estimated azimuth angle. Then, by drawing the azimuth curve with time, the motion trajectory tracking of the sound source can be clearly seen, which is shown in Fig. 16.
6. Discussion
In the field test, the azimuth angle estimation error may be partly induced by the measurement error of ranging telescope. In addition, the optical cable is not completely straight underwater because of the influence of gravity and the optical fiber may not be wound strictly evenly on the supporting mandrel, both of which will results in a deviation between the ideal azimuth angle and the actual one. In the subsequent tests, the influence of the above factors can be eliminated as far as possible by optimizing the optical cable layout and manufacturing process.
In the manuscript, the feasibility of the DOFH is verified by using Het Φ-OTDR and a 104-m-long sensitized optical cable. The optical cable in the scheme can be manufactured automatically, which is beneficial to the implementation of large-scale hydrophone array in the later stage. The noise level of Φ-OTDR is much higher than that of interferometer hydrophone, which limits the application of detection of weak acoustic signals. Frequency diversity and aggregation scheme is proven effective to eliminate the influence of interference fading and reduce the noise floor simultaneously, which is of great significance to narrow the inherent gap with interferometer hydrophone in noise floor. 5 frequency harmonics are utilized for diversity and aggregation, the noise floor is reduced by 7dB in the preliminary results. The current system noise floor is still much higher than that of the traditional interferometric hydrophone. According to the relation between diversity scale and noise reduction [20], measures should be taken to further realize diversity and aggregation with large scale. What’s more, the sensing distance and detection bandwidth in Φ-OTDR are mutually restricted. Under the condition of long sensing distance, division multiplexing or other methods are needed to further optimize the detection bandwidth of the system.
7. Conclusion
This work demonstrates a distributed optical fiber hydrophone (DOFH) based on Φ-OTDR and a field test is conducted. A new type of sensitized optical cable is introduced to make up for the low sensitivity of ordinary cable, and an array signal processing model for DOFH is constructed to analyze the equivalence and specificity of distributed array of acoustic sensors, and to guide the further array signal processing. Het Φ-OTDR with frequency diversity and aggregation is utilized to eliminate interference fading and reduce the noise level simultaneously. On basis of the optical fiber cable, underwater acoustic signal orientation, beamforming and motion trajectory tracking are realized, which indicate that Φ-OTDR is a feasible way for DOFH and has the potential to achieve an array range of tens of kilometers, with elements spaced up to the meter level and flexible configuration, which has a broad application prospect for marine acoustic detection. Hence, future work will be focused on optimization of mass production process of sensitized optical cable, the further reduction of Φ-OTDR noise level, and improvement in detection bandwidth and spatial resolution and commercialization of this technology.
Funding
National Key Research and Development Program of China (2017YFB0405501); Science and Technology Commission of Shanghai Municipality (19YF1453400); National Natural Science Foundation of China (61535014, 61675216, 61905260, 61905262); the Development council of the Chinese Academy of Sciences (KFJ-STS-QYZD-084).
Disclosures
The authors declare that there are no conflicts of interest related to this article.
References
1. W. H. Munk, “Acoustic Monitoring of Global Ocean Warming: The Very First Encounters with Concern About Marine Mammals and Sound,” Aquat Mamm 41(4), 524–528 (2015). [CrossRef]
2. A. Toky, R. P. Singh, and S. Das, “Localization schemes for Underwater Acoustic Sensor Networks - A Review,” Computer Science Review 37, 100241 (2020). [CrossRef]
3. A. Dandridge, “Fiber Optic Interferometric Sensors at Sea,” Opt. Photonics News 30(6), 34–41 (2019). [CrossRef]
4. C. K. Kirkendall and A. Dandridge, “Overview of high performance fibre-optic sensing,” J. Phys. D: Appl. Phys. 37(18), R197–R216 (2004). [CrossRef]
5. V. S. Lavrov, M. Y. Plotnikov, S. M. Aksarin, M. E. Efimov, V. A. Shulepov, A. V. Kulikov, and A. U. Kireenkov, “Experimental investigation of the thin fiber-optic hydrophone array based on fiber Bragg gratings,” Opt. Fiber Technol. 34, 47–51 (2017). [CrossRef]
6. Y. Yang, T. W. Xu, S. W. Feng, J. F. Huang, and F. Li, Optical Fiber Hydrophone Based on Distributed Acoustic Sensing[M]//ZHANG J, FU S, ZHUGE Q, TANG M, GUO T. Fiber Optic Sensing and Optical Communication. Spie-Int Soc Optical Engineering; Bellingham2018.
7. Z. Pan, K. Liang, Q. Ye, H. Cai, R. Qu, and Z. Fang, “Phase-sensitive OTDR system based on digital coherent detection,” Proc SPIE 8311, 83110S (2011). [CrossRef]
8. A. Masoudi, M. Belal, and T. P. Newson, “A distributed optical fibre dynamic strain sensor based on phase-OTDR,” Meas. Sci. Technol. 24(8), 085204 (2013). [CrossRef]
9. D. Chen, Q. Liu, and Z. He, “Phase-detection distributed fiber-optic vibration sensor without fading-noise based on time-gated digital OFDR,” Opt. Express 25(7), 8315 (2017). [CrossRef]
10. A. Masoudi and T. P. Newson, “Contributed Review: Distributed optical fibre dynamic strain sensing,” Rev. Sci. Instrum. 87(1), 011501 (2016). [CrossRef]
11. J. C. Juarez, E. W. Maier, K. N. Choi, and H. F. Taylor, “Distributed fiber-optic intrusion sensor system,” J. Lightwave Technol. 23(6), 2081–2087 (2005). [CrossRef]
12. J. C. Juarez and H. F. Taylor, “Field test of a distributed fiber-optic intrusion sensor system for long perimeters,” Appl. Opt. 46(11), 1968 (2007). [CrossRef]
13. 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), 13804–10 (2014). [CrossRef]
14. F. Peng, N. Duan, Y. J. Rao, and J. Li, “Real-Time Position and Speed Monitoring of Trains Using Phase-Sensitive OTDR,” IEEE Photonics Technol. Lett. 26(20), 2055–2057 (2014). [CrossRef]
15. Z. Y. Wang, H. R. Zheng, L. C. Li, J. J. Liang, X. Wang, B. Lu, Q. Ye, R. H. Qu, and H. W. Cai, “Practical multi-class event classification approach for distributed vibration sensing using deep dual path network,” Opt. Express 27(17), 23682–92 (2019). [CrossRef]
16. G. Fang, T. Xu, S. Feng, and F. Li, “Phase-Sensitive Optical Time Domain Reflectometer Based on Phase-Generated Carrier Algorithm,” J. Lightwave Technol. 33(13), 2811–2816 (2015). [CrossRef]
17. S. Zhou, F. Hao, and Z. M. Zeng, “Phase demodulation method in phase-sensitive OTDR without coherent detection,” Opt. Express 25(5), 4831 (2017). [CrossRef]
18. Z. N. Wang, L. Zhang, S. Wang, N. T. Xue, F. Peng, M. Q. Fan, W. Sun, X. Y. Qian, J. R. Rao, and Y. J. Rao, “Coherent Phi-OTDR based on I/Q demodulation and homodyne detection,” Opt. Express 24(2), 853 (2016). [CrossRef]
19. J. Zhou, Z. Pan, Q. Ye, H. Cai, R. Qu, and Z. Fang, “Characteristics and Explanations of Interference Fading of phi-OTDR With a Multi-Frequency Source,” J. Lightwave Technol. 31(17), 2947–2954 (2013). [CrossRef]
20. A. H. Hartog, L. B. Liokumovich, N. A. Ushakov, O. I. Kotov, T. Dean, T. Cuny, A. Constantinou, and F. V. Englich, “The use of multi-frequency acquisition to significantly improve the quality of fibre-optic-distributed vibration sensing,” Geophys. Prospect. 66(S1), 192–202 (2018). [CrossRef]
21. D. Chen, Q. Liu, and Z. He, “High-fidelity distributed fiber-optic acoustic sensor with fading noise suppressed and sub-meter spatial resolution,” Opt. Express 26(13), 16138–46 (2018). [CrossRef]
22. Q. S. Li, Research on the Characteristics of the Interferometric Fiber-Optic Hydrophone with a Concentric Mandrel Including an Enhanced Sensitivity Layer [D]. National University of Defense Technology; Changsha, Hunan, China, 2005.
23. Z. Minghua, M. Siqi, X. Yao, and Y. Chaowei, “Settlement Calculation of Geogrid-encased Stone Columns Based on Thick-walled Cylinder Model,” J. Hunan Univ. Nat. Sci. 46(5), 101 (2019). [CrossRef]
24. J. Liang, Z. Wang, B. Lu, X. Wang, L. Li, Q. Ye, R. Qu, and H. Cai, “Distributed acoustic sensing for 2D and 3D acoustic source localization,” Opt. Lett. 44(7), 1690 (2019). [CrossRef]
25. Z. Wang, J. Yang, J. Gu, B. Lu, L. Ye, K. Ying, Q. Ye, R. Qu, and H. Cai, “Multi-source aliasing suppression for distributed fiber acoustic sensing with directionally coherent enhancement technology,” Opt. Lett. 45(20), 5672 (2020). [CrossRef]
26. L. Marcon, A. Galtarossa, and L. Palmieri, “High-frequency high-resolution distributed acoustic sensing by optical frequency domain reflectometry,” Opt. Express 27(10), 13923–33 (2019). [CrossRef]
27. S. Doclo and M. Moonen, “Design of far-field and near-field broadband beamformers using eigenfilters,” Signal Process. 83(12), 2641–2673 (2003). [CrossRef]
28. T. G. Giallorenzi, J. A. Bucaro, A. Dandridge, G. H. Sigel, J. H. Cole, S. C. Rashleigh, and R. G. Priest, “Optical fiber sensor technology,” IEEE J. Quantum Electron. 18(4), 626–665 (1982). [CrossRef]
29. R. O. Schmidt, “Multiple emitter location and signal parameter estimation,” IEEE Trans. Antennas Propag. 34(3), 276–280 (1986). [CrossRef]
