Citation: |
|
[1] | Juarez J C, Maier E W, Choi K N, et al. Distributed fiber-optic intrusion sensor system[J]. J Light Technol, 2005, 23(6): 2081-2087. doi: 10.1109/JLT.2005.849924 |
[2] | Pan Z Q, Liang K Z, Ye Q, et al. Phase-sensitive OTDR system based on digital coherent detection[J]. Proc SPIE, 2011, 8311: 83110S. doi: 10.1117/12.905657 |
[3] | Juarez J C, Taylor H F. Field test of a distributed fiber-optic intrusion sensor system for long perimeters[J]. Appl Opt, 2007, 46(11): 1968-1971. doi: 10.1364/AO.46.001968 |
[4] | Lindsey N J, Martin E R, Dreger D S, et al. Fiber‐optic network observations of earthquake wavefields[J]. Geophys Res Lett, 2017, 44(23): 11792-11799. doi: 10.1002/2017GL075722 |
[5] | Cedilnik G, Hunt R, Lees G. Advances in train and rail monitoring with DAS[C]//Proceedings of the 26th International Conference on Optical Fiber Sensors, 2018: ThE35. |
[6] | Wu H J, Chen J P, Liu X R, et al. One-dimensional CNN-based intelligent recognition of vibrations in pipeline monitoring with DAS[J]. J Light Technol, 2019, 37(17): 4359-4366. doi: 10.1109/JLT.2019.2923839 |
[7] | Johannessen K, Drakeley B, Farhadiroushan M. Distributed acoustic sensing-a new way of listening to your well/reservoir[C]//SPE Intelligent Energy International, Utrecht, the Netherlands, 2012: 149602. |
[8] | Bao X Y, Zhou D P, Baker C, et al. Recent development in the distributed fiber optic acoustic and ultrasonic detection[J]. J Light Technol, 2017, 35(16): 3256-3267. doi: 10.1109/JLT.2016.2612060 |
[9] | Muanenda Y. Recent advances in distributed acoustic sensing based on phase-sensitive optical time domain reflectometry[J]. J Sens, 2018, 2018: 3897873. |
[10] | Adeel M, Shang C, Zhu K, et al. Nuisance alarm reduction: using a correlation based algorithm above differential signals in direct detected phase-OTDR systems[J]. Opt Express, 2019, 27(5): 7685-7698. doi: 10.1364/OE.27.007685 |
[11] | 饶云江, 吴敏, 冉曾令, 等. 基于准分布式FBG传感器的光纤入侵报警系统[J]. 传感技术学报, 2007, 20(5): 998-1002. doi: 10.3969/j.issn.1004-1699.2007.05.011 Rao Y J, Wu M, Ran Z L, et al. A fiber-optic intrusion alarm system based on quasi-distributed FBG sensors[J]. Chin J Sens Actuators, 2007, 20(5): 998-1002. doi: 10.3969/j.issn.1004-1699.2007.05.011 |
[12] | Mahmoud S S, Katsifolis J. Elimination of rain-induced nuisance alarms in distributed fiber optic perimeter intrusion detection systems[J]. Proc SPIE, 2009, 7316: 731604. doi: 10.1117/12.818096 |
[13] | 吴红艳, 贾波, 卞庞. 光纤周界安防系统端点检测技术的研究[J]. 仪器仪表学报, 2013, 34(4): 743-748. doi: 10.3969/j.issn.0254-3087.2013.04.004 Wu H Y, Jia B, Bian P. Study on endpoint detection technology based on fiber perimeter security system[J]. Chin J Sci Instrum, 2013, 34(4): 743-748. doi: 10.3969/j.issn.0254-3087.2013.04.004 |
[14] | 王思远, 娄淑琴, 梁生, 等. M-Z干涉仪型光纤分布式扰动传感系统模式识别方法[J]. 红外与激光工程, 2014, 43(8): 2613-2618. doi: 10.3969/j.issn.1007-2276.2014.08.036 Wang S Y, Lou S Q, Liang S, et al. Pattern recognition method of fiber distributed disturbance sensing system based on M-Z interferometer[J]. Infrared Laser Eng, 2014, 43(8): 2613-2618. doi: 10.3969/j.issn.1007-2276.2014.08.036 |
[15] | 刘琨, 何畅, 刘铁根, 等. 一种用于光纤周界安防系统的端点检测方法[J]. 光电子·激光, 2014, 25(11): 2136-2140. Liu K, He C, Liu T G, et al. An endpoint detection method for fiber perimeter security system[J]. J Opto Laser, 2014, 25(11): 2136-2140. |
[16] | 朱程辉, 瞿永中, 王建平. 基于时频特征的光纤周界振动信号识别[J]. 光电工程, 2014, 41(1): 16-22. doi: 10.3969/j.issn.1003-501X.2014.01.004 Zhu C H, Qu Y Z, Wang J P. The vibration signal recognition of optical fiber perimeter based on time-frequency features[J]. Opto-Electron Eng, 2014, 41(1): 16-22. doi: 10.3969/j.issn.1003-501X.2014.01.004 |
[17] | 王建平, 郝钊, 朱程辉. 基于相空间重构的光纤周界信号识别算法研究[J]. 合肥工业大学学报(自然科学版), 2017, 40(5): 643-648. doi: 10.3969/j.issn.1003-5060.2017.05.014 Wang J P, Hao Z, Zhu C H. Research on vibration signal recognition of optical fiber perimeter based on phase space reconstruction[J]. J Hefei Univ Technol (Nat Sci), 2017, 40(5): 643-648. doi: 10.3969/j.issn.1003-5060.2017.05.014 |
[18] | 刘琨, 翁凌锋, 江俊峰, 等. 基于过零率的光纤周界安防系统入侵事件高效识别[J]. 光学学报, 2019, 39(11): 1106002. Liu K, Weng L F, Jiang J F, et al. Zero-crossing rate based efficient identification of intrusion events in fiber perimeter security systems[J]. Acta Opt Sin, 2019, 39(11): 1106002. |
[19] | 王照勇, 潘政清, 叶青, 等. 用于光纤围栏入侵告警的频谱分析快速模式识别[J]. 中国激光, 2015, 42(4): 0405010. Wang Z Y, Pan Z Q, Ye Q, et al. Fast pattern recognition based on frequency spectrum analysis used for intrusion alarming in optical fiber fence[J]. Chin J Lasers, 2015, 42(4): 0405010. |
[20] | Cao C, Fan X Y, Liu Q W, et al. Practical pattern recognition system for distributed optical fiber intrusion monitoring system based on phase-sensitive coherent OTDR[C]//Asia Communications and Photonics Conference 2015, 2015: ASu2A. 145. |
[21] | 黄翔东, 张皓杰, 刘琨, 等. 基于综合特征的光纤周界安防系统高效入侵事件识别[J]. 物理学报, 2017, 66(12): 124206. doi: 10.7498/aps.66.124206 Huang X D, Zhang H J, Liu K, et al. High-efficiency intrusion recognition by using synthesized features in optical fiber perimeter security system[J]. Acta Phys Sin, 2017, 66(12): 124206. doi: 10.7498/aps.66.124206 |
[22] | 邹东伯, 刘海, 赵亮, 等. 分布式光纤振动传感信号识别的研究[J]. 激光技术, 2016, 40(1): 86-89. Zou D B, Liu H, Zhao L, et al. Research of signal recognition of distributed optical fiber vibration sensors[J]. Laser Technol, 2016, 40(1): 86-89. |
[23] | 帅师, 王翦, 吴红艳, 等. 一种分布式光纤传感系统的信号识别方法[J]. 复旦学报(自然科学版), 2018, 57(5): 611-618. Shuai S, Wang J, Wu H Y, et al. A signal recognition method for distributed optical fiber sensor system[J]. J Fudan Univ (Nat Sci), 2018, 57(5): 611-618. |
[24] | Tejedor J, Martins H F, Piote D, et al. Toward prevention of pipeline integrity threats using a smart fiber-optic surveillance system[J]. J Light Technol, 2016, 34(19): 4445-4453. doi: 10.1109/JLT.2016.2542981 |
[25] | Tejedor J, Macias-Guarasa J, Martins H F, et al. A novel fiber optic based surveillance system for prevention of pipeline integrity threats[J]. Sensors, 2017, 17(2): 355. doi: 10.3390/s17020355 |
[26] | 李志辰, 刘琨, 江俊峰, 等. 光纤周界安防系统的高准确度事件识别方法[J]. 红外与激光工程, 2018, 47(9): 0922002. Li Z C, Liu K, Jiang J F, et al. A high-accuracy event discrimination method in optical fiber perimeter security system[J]. Infrared Laser Eng, 2018, 47(9): 0922002. |
[27] | 陈沛超, 游赐天, 丁攀峰. 光纤周界防区入侵事件的模式识别研究[J]. 中国激光, 2019, 46(10): 1006001. Chen P C, You C T, Ding P F. Pattern recognition of intrusion events in perimeter defense areas of optical fiber[J]. Chin J Lasers, 2019, 46(10): 1006001. |
[28] | Huang N E, Shen Z, Long S R, et al. The empirical mode decomposition and the Hilbert spectrum for nonlinear and non-stationary time series analysis[J]. Proc Math Phys Eng Sci, 1998, 454(1971): 903-995. doi: 10.1098/rspa.1998.0193 |
[29] | Liu K, Tian M, Liu T G, et al. A high-efficiency multiple events discrimination method in optical fiber perimeter security system[J]. J Light Technol, 2015, 33(23): 4885-4890. doi: 10.1109/JLT.2015.2494158 |
[30] | 蒋立辉, 盖井艳, 王维波, 等. 基于总体平均经验模态分解的光纤周界预警系统模式识别方法[J]. 光学学报, 2015, 35(10): 1006002. Jiang L H, Gai J Y, Wang W B, et al. Ensemble empirical mode decomposition based event classification method for the fiber-optic intrusion monitoring system[J]. Acta Opt Sin, 2015, 35(10): 1006002. |
[31] | 李静云, 安博文, 陈元林, 等. 基于时频特征的光纤振动模式识别研究[J]. 光通信技术, 2018, 42(7): 55-59. Li J Y, An B W, Chen Y L, et al. Research on optical fiber vibration pattern recognition based on time-frequency characteristics[J]. Opt Commun Technol, 2018, 42(7): 55-59. |
[32] | 朱程辉, 朱睿, 王建平, 等. 基于自适应EMD的光纤安防系统入侵信号识别[J]. 传感器与微系统, 2020, 39(4): 26-30. Zhu C H, Zhu R, Wang J P, et al. Intrusion signal recognition of optical fiber security & protection system based on adaptive EMD[J]. Transducer and Microsystem Technologies, 2020, 39(4): 26-30. |
[33] | 张景川, 曾周末, 赖平, 等. 基于小波能谱和小波信息熵的管道异常振动事件识别方法[J]. 振动与冲击, 2010, 29(5): 1-4. doi: 10.3969/j.issn.1000-3835.2010.05.001 Zhang J C, Zeng Z M, Lai P, et al. A recognition method with wavelet energy spectrum and wavelet information entropy for abnormal vibration events of a petroleum pipeline[J]. J Vib Shock, 2010, 29(5): 1-4. doi: 10.3969/j.issn.1000-3835.2010.05.001 |
[34] | 李彦, 梁正桃, 李立京, 等. 基于小波和支持向量机的光纤微振动传感器模式识别[J]. 传感器与微系统, 2013, 32(2): 43-45, 49. doi: 10.3969/j.issn.1000-9787.2013.02.013 Li Y, Liang Z T, Li L J, et al. Pattern recognition of fiber-optic micro vibration sensor based on wavelet and SVM[J]. Transducer Microsyst Technol, 2013, 32(2): 43-45, 49. doi: 10.3969/j.issn.1000-9787.2013.02.013 |
[35] | Wu H J, Xiao S K, Li X Y, et al. Separation and determination of the disturbing signals in phase-sensitive optical time domain reflectometry (Φ-OTDR)[J]. J Light Technol, 2015, 33(15): 3156-3162. doi: 10.1109/JLT.2015.2421953 |
[36] | 喻骁芒, 罗光明, 朱珍民, 等. 分布式光纤传感器周界安防入侵信号的多目标识别[J]. 光电工程, 2014, 41(1): 36-41. doi: 10.3969/j.issn.1003-501X.2014.01.007 Yu X M, Luo G M, Zhu Z M, et al. The multi target recognition of intrusion signal of perimeter security with distributed fiber-optic sensor[J]. Opto-Electron Eng, 2014, 41(1): 36-41. doi: 10.3969/j.issn.1003-501X.2014.01.007 |
[37] | 李凯彦, 赵兴群, 孙小菡, 等. 一种用于光纤链路振动信号模式识别的规整化复合特征提取方法[J]. 物理学报, 2015, 64(5): 054304. Li K Y, Zhao X Q, Sun X H, et al. A regular composite feature extraction method for vibration signal pattern recognition in optical fiber link system[J]. Acta Phys Sin, 2015, 64(5): 054304. |
[38] | Wu H J, Qian Y, Zhang W, et al. Feature extraction and identification in distributed optical-fiber vibration sensing system for oil pipeline safety monitoring[J]. Photonic Sens, 2017, 7(4): 305-310. doi: 10.1007/s13320-017-0360-1 |
[39] | 彭宽, 冯诚, 王森懋, 等. 基于时/频域综合特征提取的分布式光纤入侵监测系统事件识别方法[J]. 光学学报, 2019, 39(6): 0628002. Peng K, Feng C, Wang S M, et al. Event discrimination method for distributed optical fiber intrusion sensing system based on integrated time/frequency domain feature extraction[J]. Acta Opt Sin, 2019, 39(6): 0628002. |
[40] | Sun Q, Feng H, Yan X Y, et al. Recognition of a phase-sensitivity OTDR sensing system based on morphologic feature extraction[J]. Sensors, 2015, 15(7): 15179-15197. doi: 10.3390/s150715179 |
[41] | Aslangul S A. Detecting tunnels for border security based on fiber optical distributed acoustic sensor data using DBSCAN[C]//Proceedings of the 9th International Conference on Sensor Networks, 2020: 78-84. |
[42] | Cortes C, Vapnik V. Support-vector networks[J]. Mach Learn, 1995, 20(3): 273-297. |
[43] | Qi X X, Ji J W, Han X W, et al. An Approach of passive vehicle type recognition by acoustic signal based on SVM[C]//Proceedings of the 2009 Third International Conference on Genetic and Evolutionary Computing, 2009: 545-548. |
[44] | King D, Lyons W B, Flanagan C, et al. A multipoint optical fibre sensor system for use in process water systems based on artificial neural network pattern recognition techniques[J]. Sens Actuator A Phys, 2004, 115(2-3): 293-302. doi: 10.1016/j.sna.2004.03.068 |
[45] | Lewis E, Sheridan C, O'Farrell M, et al. Principal component analysis and artificial neural network based approach to analysing optical fibre sensors signals[J]. Sens Actuator A Phys, 2007, 136(1): 28-38. doi: 10.1016/j.sna.2007.02.012 |
[46] | 张俊楠, 娄淑琴, 梁生. 基于SVM算法的φ-OTDR分布式光纤扰动传感系统模式识别研究[J]. 红外与激光工程, 2017, 46(4): 0422003. Zhang J N, Lou S Q, Liang S. Study of pattern recognition based on SVM algorithm for φ-OTDR distributed optical fiber disturbance sensing system[J]. Infrared Laser Eng, 2017, 46(4): 0422003. |
[47] | Tipping M E. The relevance vector machine[C]//Advances in Neural Information Processing Systems, 2000: 652-658. |
[48] | 朱永利, 尹金良. 组合核相关向量机在电力变压器故障诊断中的应用研究[J]. 中国电机工程学报, 2013, 33(22): 68-74. Zhu Y L, Yin J L. Study on application of multi-kernel learning relevance vector machines in fault diagnosis of power transformers[J]. Proc IEEE Inst Electr Electron Eng, 2013, 33(22): 68-74. |
[49] | 孙茜, 曾周末, 李健. 相关向量机在光纤预警系统模式识别中的应用[J]. 天津大学学报(自然科学与工程技术版), 2014, 47(12): 1115-1120. Sun Q, Zeng Z M, Li J. Application of relevance vector machine in pattern recognition of optical fiber pre-warning system[J]. J Tianjin Univ (Sci Technol), 2014, 47(12): 1115-1120. |
[50] | Rumelhart D E, Hinton G E, Williams R J. Learning internal representations by error propagation[M]//Parallel Distributed Processing: Explorations in the Microstructure Of Cognition, Vol. 1: Foundations. Cambridge: MIT Press, 1986: 318-362. |
[51] | 李小玉, 吴慧娟, 彭正谱, 等. 基于时间序列奇异谱特征的Φ-OTDR扰动检测方法[J]. 光子学报, 2014, 43(4): 0428001. Li X Y, Wu H J, Peng Z P, et al. A novel time sequence singular spectrum analysis method for Φ-OTDR disturbance detection system[J]. Acta Photo Sin, 2014, 43(4): 0428001. |
[52] | 谢鑫, 吴慧娟, 饶云江. 一种基于光纤布喇格光栅振动传感器的光纤围栏入侵监测系统及其模式识别[J]. 光子学报, 2014, 43(5): 0506005. Xie X, Wu H J, Rao Y J. A fiber-optical perimeter intrusion detection system based on the fiber Bragg grating vibration sensors and its identification method[J]. Acta Photo Sin, 2014, 43(5): 0506005. |
[53] | 沈隆翔, 封皓, 沙洲, 等. 基于下变频和IQ解调的外差型相位敏感光时域反射技术的模式识别[J]. 光学学报, 2017, 37(8): 0806005. Shen L X, Feng H, Sha Z, et al. Pattern recognition of heterodyne phase-sensitive optical time-domain reflection technique based on down conversion and IQ demodulation[J]. Acta Opt Sin, 2017, 37(8): 0806005. |
[54] | Aktas M, Akgun T, Demircin M U, et al. Deep learning based threat classification in distributed acoustic sensing systems[C]//Proceedings of the 2017 25th Signal Processing and Communications Applications Conference (SIU), 2017. |
[55] | Xu C J, Guan J J, Bao M, et al. Pattern recognition based on time-frequency analysis and convolutional neural networks for vibrational events in φ-OTDR[J]. Opt Eng, 2018, 57(1): 016103. |
[56] | Shi Y, Wang Y Y, Zhao L, et al. An event recognition method for Φ-OTDR sensing system based on deep learning[J]. Sensors (Basel), 2019, 19(15): 3421. doi: 10.3390/s19153421 |
[57] | 吴俊, 管鲁阳, 鲍明, 等. 基于多尺度一维卷积神经网络的光纤振动事件识别[J]. 光电工程, 2019, 46(5): 180493. doi: 10.12086/oee.2019.180493 Wu J, Guan L Y, Bao M, et al. Vibration events recognition of optical fiber based on multi-scale 1-D CNN[J]. Opto-Electron Eng, 2019, 46(5): 180493. doi: 10.12086/oee.2019.180493 |
[58] | Goodfellow I J, Pouget-Abadie J, Mirza M, et al. Generative adversarial nets[C]//Proceedings of the 27th International Conference on Neural Information Processing Systems, 2014: 2672-2680. |
[59] | Shiloh L, Eyal A, Giryes R. Deep learning approach for processing fiber-optic DAS seismic data[C]//Proceedings of the 26th International Conference on Optical Fiber Sensors, 2018: ThE22. |
[60] | Li W, Zeng Z Q, Qu H Q, et al. A novel fiber intrusion signal recognition method for ofps based on SCN with dropout[J]. J Light Technol, 2019, 37(20): 5221-5230. doi: 10.1109/JLT.2019.2930624 |
[61] | Wang Z Y, Zheng H R, Li L C, et al. Practical multi-class event classification approach for distributed vibration sensing using deep dual path network[J]. Opt Express, 2019, 27(17): 23682-23692. doi: 10.1364/OE.27.023682 |
[62] | Chen X, Xu C J. Disturbance pattern recognition based on an ALSTM in a long‐distance φ‐OTDR sensing system[J]. Microw Opt Technol Lett, 2020, 62(1): 168-175. doi: 10.1002/mop.32025 |
[63] | Li Z Q, Zhang J W, Wang M N, et al. Fiber distributed acoustic sensing using convolutional long short-term memory network: a field test on high-speed railway intrusion detection[J]. Opt Express, 2020, 28(3): 2925-2938. doi: 10.1364/OE.28.002925 |
Overview: Developed on the basis of phase-sensitive optical time domain reflectrometer(Φ-OTDR), distributed acoustic sensing(DAS) is a new type of distributed optical fiber sensing technology. It is a hot topic that how to accurately distinguish the type of intrusion events from complex signals, which attachs greater importance to the research on the pattern recognition technology based on DAS.
Intrusion signal recognition mainly includes two parts, feature extraction and classification algorithm. This paper summarizes the currently widely used feature extraction methods. Generally speaking, the time-domain features are simple, intuitive, and fast in response, but they are susceptible to noise. The frequency-domain features can obtain the inherent spectrum characteristics of the signal, but cannot reflect the frequency changes of the signal at every moment. The time-frequency domain features can express the time-domain and frequency-domain information of the signal, and the extracted feature information is also more accurate.
Classification algorithms include two categories, unsupervised learning and supervised learning. Supervised learning need to collect a large amount of data for training and verification. Therefore, supervised learning algorithms are mostly used in the fiber intrusion detection applications. Support vector machines (SVM) and BP neural networks are relatively common models for classification. In recent years, with the development of deep learning technology, building deep neural network models is very helpful for classification recognition. Therefore, models such as convolutional neural network (CNN), recurrent neural network (RNN), and generative adversarial network (GAN) are used in the field of distributed optical fiber intrusion signal recognition, which have achieved great performance.
In summary, choosing the proper feature extraction method and classification algorithm will greatly enhance the accuracy of intrusion signal recognition. Facing the increasing actual demands, the pattern recognition technology based on the DAS system will definitely play a more important role and fulfill its potential in the future.
DAS system structure
Intrusion signal recognition process
Signal representing a climb during torrential rain as detected. (a) Time domain representations; (b) LC vs. block number[12]
Original signals of five events. (a) Cutting; (b) Waggling; (c) Climbing; (d) Knocking; (e) No intrusion[18]
Segment zero-crossing rates of five events. (a) Cutting; (b) Waggling; (c) Climbing; (d) Knocking; (e) No intrusion[18]
FFT feature extraction flow chart
MFCC feature extraction flow chart
STFT time-frequency diagrams of two kinds of window functions for processing four intrusion events. (a), (c), (e), (g) Time-frequency diagrams of knocking, shaking, winding, and raining signals after passing through the Hanning window; (b), (d), (f), (h) time-frequency diagrams of knocking, shaking, winding, and raining signals after passing through the Kaiser window[27]
Fence invasive signals and their IMF components through EMD. (a) EMD of climbing; (b) EMD of knocking; (c) EMD of waggling; (d) EMD of cutting[29]
Signals and their kurtosis eigenvectors of four cases. (a) Climbing signal; (b) Knocking signal; (c) Eigenvectors of climbing; (d) Eigenvectors of knocking; (e) Waggling signal; (f) Cutting signal; (g) Eigenvectors of waggling; (h) Eigenvectors of cutting
Multi-scale decomposition tree. (a) Wavelet decomposition; (b) Wavelet packet decomposition
WE distribution for three typical events[38]
WPE distribution for three typical events[38]
(a) Calculated signal of vehicle passing; (b) Experimentally measured signal of vehicle passing[40]
DBSCAN core and outlier points[41]
Directed acyclic graph of RVM[49]
Feature distribution of three events[40]
Three-layer BP neural network structure
Typical structure of CNN
The effect of spectral subtraction on the vibration signal. (a) The time-domain waveform of the knocking signal after noise reduction; (b) The spectrogram of the knocking signal after noise reduction[55]
The optimized network structure (the red cube denotes convolution operation and the blue cube denotes pooling operation)[56]
Confusion matrix of five events' classification[56]
GAN flow chart
Accuracy and loss of testing datasets at different training algorithms[61]
Cyclic unit structure of LSTM network