Machine learning methods for identification and classification of events in ϕ-OTDR systems: a review

Analysis of Field Trial Results for Excavation-Activities Monitoring with φ-OTDR

Underground telecommunication cables are highly susceptible to damage from excavation activities. Preventing accidental damage to underground telecommunication cables is critical and necessary. In this study, we present field trial results of monitoring excavation activities near underground fiber cables using an intensity-based phase-sensitive optical time-domain reflectometer (φ-OTDR). The reasons for choosing intensity-based φ-OTDR for excavation monitoring are presented and analyzed. The vibration signals generated by four typical individual excavation events, i.e., cutting, hammering, digging, and tamping at five different field trial sites, as well as five different mixed events in the fifth field trial site were investigated. The findings indicate that various types of events can generate vibration signals with different features. Typically, fundamental peak frequencies of cutting, hammering and tamping events ranged from 30 to 40 Hz, 11 to 15 Hz, and 30 to 40 Hz, respectively. Digging events, on the other hand, presented a broadband frequency spectrum without a distinct peak frequency. Moreover, due to differences in environmental conditions, even identical excavation events conducted with the same machine may also generate vibration signals with different characteristics. The diverse field trial results presented offer valuable insights for both research and the practical implementation of excavation monitoring techniques for underground cables.

Instability Compensation of Recording Interferometer in Phase-Sensitive OTDR

In the paper, a new method of phase measurement error suppression in a phase-sensitive optical time domain reflectometer is proposed and experimentally proved. The main causes of phase measurement errors are identified and considered, such as the influence of the recording interferometer instabilities and laser wavelength instability, which can cause inaccuracies in phase unwrapping. The use of a Mach-Zender interferometer made by 3 × 3 fiber couplers is proposed and tested to provide insensitivity to the recording interferometer and laser source instabilities. It is shown that using all three available photodetectors of the interferometer, instead of just one pair, achieves significantly better accuracy in the phase unwrapping. A novel compensation scheme for accurate phase measurements in a phase-sensitive optical time domain reflectometer is proposed, and a comparison of the measurement signals with or without such compensation is shown and discussed. The proposed method, using three photodetectors, allows for very good compensation of the phase measurement errors arising from common-mode noise from the interferometer and laser source, providing a significant improvement in signal detection. In addition, the method allows the tracking of slow temperature changes in the monitored fiber/object, which is not obtainable when using a simple low-pass filter for phase unwrapping error reduction, as is customary in several systems of this kind.

Field trial of concurrent co-cable and co-trench optical fiber online identification based on ensemble learning

The co-route optical fibers, comprising both co-cable and co-trench fibers, pose a significant potential risk to network service quality assurance by operators. They are incapable of achieving high-precision recognition and visual state management. In this study, we gathered both static and dynamic optical fiber data using a linewidth tunable light source (LTLS) and introduced a multimodal detection architecture that applies ensemble learning to the collected data. This constitutes what we believe to be the first field trial of concurrent recognition of optical fibers found both in co-cables and co-trenches. To identify co-cable fibers, we employed a double-layer cascaded Random Forest (DLC-RF) model based on the static features of fibers. For co-trench fiber, the dynamic characteristics of fiber vibrations are utilized in combination with multiple independent curve similarity contrast learners for classifying tasks. The proposed architecture is capable of automatically detecting the condition of the optical fiber and actively identifying the same routing segment within the network, eliminating the need for human intervention and enabling the visualization of passive optical fiber resources. Finally, after rigorous testing and validation across 11 sites in a typical urban area, including aggregation and backbone scenarios within the operator's live network environments, we have confirmed that the solution's ability to identify co-routes is accurate, exceeding 95%. This provides strong empirical evidence of its effectiveness.

Highly discriminative and adaptive feature extraction method based on NMF-MFCC for event recognition of Φ-OTDR

To enhance the capability of phase-sensitive optical time domain reflectometers (Φ-OTDR) to recognize disturbance events, an improved adaptive feature extraction method based on NMF-MFCC is proposed, which replaces the fixed filter bank used in the traditional method to extract the mel-frequency cepstral coefficient (MFCC) features by a spectral structure obtained from the Φ-OTDR signal spectrum using nonnegative matrix factorization (NMF). Three typical events on fences are set as recognition targets in our experiments, and the results show that the NMF-MFCC features have higher distinguishability, with the corresponding recognition accuracy reaching 98.47%, which is 7% higher than that using the traditional MFCC features.

Intelligent water perimeter security event recognition based on NAM-MAE and distributed optic fiber acoustic sensing system

Distributed optical acoustic sensing (DAS) based on phase-sensitive optical time-domain reflectometry can realize the distributed monitoring of multi-point disturbances along an optical fiber, thus making it suitable for water perimeter security applications. However, owing to the complex environment and the production of various noises by the system, continuous and effective recognition of disturbance signals becomes difficult. In this study, we propose a Noise Adaptive Mask-Masked Autoencoders (NAM-MAE) algorithm based on the novel mask mode of a Masked Autoencoders (MAE) and applies it to the intelligent event recognition in DAS. In this method, fewer but more accurate features are fed into the deep learning model for recognition by directly shielding the noise. Taking the fading noise generated by the system as an example, data on water perimeter security events collected in DAS underwater acoustic experiments are used. The NAM-MAE is compared with other models. The results indicate higher training accuracy and higher convergence speed of NAM-MAE than other models. Further, the final test accuracy reaches 96.6134%. It can be demonstrated that the proposed method has feasibility and superiority.

Hybrid Distributed Optical Fiber Sensor for the Multi-Parameter Measurements

Distributed optical fiber sensors (DOFSs) are a promising technology for their unique advantage of long-distance distributed measurements in industrial applications. In recent years, modern industrial monitoring has called for comprehensive multi-parameter measurements to accurately identify fault events. The hybrid DOFS technology, which combines the Rayleigh, Brillouin, and Raman scattering mechanisms and integrates multiple DOFS systems in a single configuration, has attracted growing attention and has been developed rapidly. Compared to a single DOFS system, the multi-parameter measurements based on hybrid DOFS offer multidimensional valuable information to prevent misjudgments and false alarms. The highly integrated sensing structure enables more efficient and cost-effective monitoring in engineering. This review highlights the latest progress of the hybrid DOFS technology for multi-parameter measurements. The basic principles of the light-scattering-based DOFSs are initially introduced, and then the methods and sensing performances of various techniques are successively described. The challenges and prospects of the hybrid DOFS technology are discussed in the end, aiming to pave the way for a vaster range of applications.

Finding Well-Coupled Optical Fiber Locations for Railway Monitoring Using Distributed Acoustic Sensing

Distributed acoustic sensors (DAS) utilize optical fibers to monitor vibrations across thousands of independent locations. However, the measured acoustic waveforms experience significant variations along the sensing fiber. These differences primarily arise from changes in coupling between the fiber and its surrounding medium as well as acoustic interferences. Here, a correlation-based method is proposed to automatically find the spatial locations of DAS where temporal waveforms are repeatable. Signal repeatability is directly associated with spatial monitoring locations with both good coupling and low acoustic interference. The DAS interrogator employed is connected to an over 30-year-old optical fiber installed alongside a railway track. Thus, the optical fiber exhibits large coupling changes and different installation types along its path. The results indicate that spatial monitoring locations with good temporal waveform repeatability can be automatically discriminated using the proposed method. The correlation between the temporal waveforms acquired at locations selected by the algorithm proved to be very high considering measurements taken for three days, the first two on consecutive days and the third one a month after the first measurement.

Multievent localization for loop-based Sagnac sensing system using machine learning

In optical sensing applications such as pipeline monitoring and intrusion detection systems, accurate localization of the event is crucial for timely and effective response. This paper experimentally demonstrates multievent localization for long perimeter monitoring using a Sagnac interferometer loop sensor and machine learning techniques. The proposed method considers the multievent localization problem as a multilabel multiclassification problem by dividing the optical fiber into 250 segments. A deep neural network (DNN) model is used to predict the likelihood of event occurrence in each segment and accurately locate the events. The sensing loop comprises 106.245 km of single-mode fiber, equivalent to ∼50 km of effective sensing distance. The training dataset is constructed in simulation using VPItransmissionMaker, and the proposed machine learning model's complexity is reduced by using discrete cosine transform (DCT). The designed DNN is tested for event localization in both simulation and experiment. The simulation results show that the proposed model achieves an accuracy of 99% in predicting the location of one event within one segment error, an accuracy of 95% in predicting the location of one event out of the two within one segment error, and an accuracy of 78% in predicting the location of the two events within one segment error. The experimental results validate the simulation ones, demonstrating the proposed model's effectiveness in accurately localizing events with high precision. In addition, the paper includes a discussion on extending the proposed model to sense more than two events simultaneously.

Multichannel Classifier for Recognizing Acoustic Impacts Recorded with a phi-OTDR

The purpose of this work is to increase the security of the perimeter of an area from unauthorized intrusions by creating an improved algorithm for classifying acoustic impacts recorded with a sensor system based on a phase-sensitive optical time reflectometer (phi-OTDR). The algorithm includes machine learning, so a dataset consisting of two classes was assembled. The dataset consists of two classes. The first class is the data of the steps, and the second class is other non-stepping influences (engine noise, a passing car, a passing cyclist, etc.). As an intrusion signal, a human walking signal is analyzed and recorded in frames of 5 s, which passed the threshold condition. Since, in most cases, the intruder moves on foot to overcome the perimeter, the analysis of the acoustic effects generated during the step will increase the efficiency of the perimeter detection tools. When walking quietly, step signals can be quite weak, and background signals can contain high energy and visually resemble the signals you are looking for. Therefore, an algorithm was created that processes space-time diagrams developed in real time, which are grayscale images. At the same time, during the processing of one image, two more images are calculated, which are the result of processing the denoised autoencoder and the created mathematical model of the adaptive correlation. Then, the three obtained images are fed to the input of the created three-channel neural network classifier, which includes convolutional layers for the automatic extraction of spatial features. The probability of correctly detecting steps is 98.3% and that of background actions is 97.93%.

Machine Learning Approaches in Brillouin Distributed Fiber Optic Sensors

This paper presents reported machine learning approaches in the field of Brillouin distributed fiber optic sensors (DFOSs). The increasing popularity of Brillouin DFOSs stems from their capability to continuously monitor temperature and strain along kilometer-long optical fibers, rendering them attractive for industrial applications, such as the structural health monitoring of large civil infrastructures and pipelines. In recent years, machine learning has been integrated into the Brillouin DFOS signal processing, resulting in fast and enhanced temperature, strain, and humidity measurements without increasing the system's cost. Machine learning has also contributed to enhanced spatial resolution in Brillouin optical time domain analysis (BOTDA) systems and shorter measurement times in Brillouin optical frequency domain analysis (BOFDA) systems. This paper provides an overview of the applied machine learning methodologies in Brillouin DFOSs, as well as future perspectives in this area.

Strain FBG-Based Sensor for Detecting Fence Intruders Using Machine Learning and Adaptive Thresholding

This paper demonstrates an intruder detection system using a strain-based optical fiber Bragg grating (FBG), machine learning (ML), and adaptive thresholding to classify the intruder as no intruder, intruder, or wind at low levels of signal-to-noise ratio. We demonstrate the intruder detection system using a portion of a real fence manufactured and installed around one of the engineering college's gardens at King Saud University. The experimental results show that adaptive thresholding can help improve the performance of machine learning classifiers, such as linear discriminant analysis (LDA) or logistic regression algorithms in identifying an intruder's existence at low optical signal-to-noise ratio (OSNR) scenarios. The proposed method can achieve an average accuracy of 99.17% when the OSNR level is <0.5 dB.

Machine Learning Estimation of the Phase at the Fading Points of an OFDR-Based Distributed Sensor

The paper reports a machine learning approach for estimating the phase in a distributed acoustic sensor implemented using optical frequency domain reflectometry, with enhanced robustness at the fading points. A neural network configuration was trained using a simulated set of optical signals that were modeled after the Rayleigh scattering pattern of a perturbed fiber. Firstly, the performance of the network was verified using another set of numerically generated scattering profiles to compare the achieved accuracy levels with the standard homodyne detection method. Then, the proposed method was tested on real experimental measurements, which indicated a detection improvement of at least 5.1 dB with respect to the standard approach.

Pattern recognition using self-reference feature extraction for φ-OTDR

This paper proposes a pattern recognition method for φ-OTDR based on self-reference features, where machine learning is applied to classify the vibration monitored. The φ-OTDR collects the light amplitude-time-space sequence, establishes a reference position in the spatial dimension, and combines the two dimensions of the vibration and reference positions to form self-reference features, which are then used as machine learning features. These self-reference features can effectively improve the pattern recognition accuracy. This paper selects a low sampling frequency for data collection, analyzes the influence of sample definition methods of different time lengths on the pattern recognition accuracy, and determines that the optimal sample length is 10 data points. The contribution of different feature parameters to pattern recognition is analyzed, and eight eigenvalues such as average, maximum, and minimum are finally determined to form self-reference features that are used as the input of the machine learning algorithm. The recognition accuracies of five machine learning algorithms including kNN, Decision Tree, Random Forest, LightGBM, and CatBoost are analyzed and compared, and the CatBoost algorithm in the integrated learning algorithm is finally determined as the optimal algorithm. On this basis, this paper proposes a filtering algorithm to deal with abnormal signals, which can effectively compensate for abnormal data and further improve the accuracy of pattern recognition. Finally, this paper conducts the pattern recognition study on four common events of tapping, bending, trampling, and blowing, and obtains the average recognition rate of 98%. In addition, this paper innovatively carried out pattern recognition research on five types of mining equipment, including ball mills, vibrating screens, conveyor belts, filters, and industrial pumps, and obtained the average recognition rate of 93.5%.

Event recognition method based on dual-augmentation for a Φ-OTDR system with a few training samples

Thanks to the development of machine learning and deep learning, data-driven pattern recognition based on neural network is a trend for Φ-OTDR system intrusion event recognition. The data-driven pattern recognition needs a large number of samples for training. However, in some scenarios, intrusion signals are difficult to collect, resulting in the lack of training samples. At the same time, labeling a large number of samples is also a very time-consuming work. This paper presents a few-shot learning classification method based on time series transfer and cycle generative adversarial network (CycleGAN) data augmentation for Φ-OTDR system. By expanding the rare samples based on time series transfer and CycleGAN, the number of samples in the dataset can finally meet the requirement of network training. The experimental result shows that even when the training set has two minor classes with only two samples, the average accuracy of the validation set with 5 classification tasks can still reach 90.84%, and the classification accuracy of minor classes can reach 79.28% with the proposed method.

High-efficient disturbance event recognition method of ϕ-OTDR utilizing region-segmentation differential phase signals

We demonstrate a disturbance event recognition method based on region segmentation, which utilizes differential phase signals of a phase-sensitive optical time-domain reflectometer (ϕ-OTDR) to recognize disturbance events efficiently. The long-haul sensing fiber is divided into subsensing regions; whereas the phase signals at the two end points of the subsensing regions are subtracted, unwrapped, and differenced to represent the disturbance information. Feature extraction and classification are performed separately on the subsensing regions datasets. The experimental results indicate that the average recognition accuracy of the region-segmentation-based event recognition method is up to 92.9%. Compared to the method without region segmentation, this proposed method improves the average recognition accuracy by 8%; whereas the recognition time of three disturbance events on a 14.8-km sensing system is only 0.39 s. The proposed method provides significant support for the development of disturbance event recognition of the ϕ-OTDR sensor system.