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

Article Engineering, Electrical & Electronic

Near-field target localisation based on the distributed acoustic sensing optical fibre in shallow water

Weihao Cao et al.

Summary: The distributed acoustic sensing optical fibre (DASF) is widely used for target detection and localisation. By using Phi-OTDR based on DASF, near-field targets can be effectively located.

OPTICAL FIBER TECHNOLOGY (2023)

Add to Collection

Article Chemistry, Analytical

Classification of Acoustic Influences Registered with Phase-Sensitive OTDR Using Pattern Recognition Methods

Ivan A. Barantsov et al.

Summary: This article focuses on the development of a classification method based on an artificial neural network architecture for recognizing acoustic influences recorded by a phase-sensitive OTDR. The use of a band-pass filter at the initial stage of signal processing improves the signal-to-noise ratio of the collected data sets. Experimental results show that the AlexNet and DenseNet169 architectures achieve accuracies above 90%. A novel CNN architecture based on AlexNet is proposed and achieves the best results, with an accuracy above 98%. The proposed model has the advantages of low power consumption (400 mW) and high speed (0.032 s per net evaluation).

SENSORS (2023)

Add to Collection

Article Engineering, Electrical & Electronic

Real-time classification for 0-OTDR vibration events in the case of small sample size datasets

Nachuan Yang et al.

Summary: This paper proposes a fiber optic system vibration event recognition method based on a combination of image segmentation pre-processing, texture, statistical, morphological feature extraction, and weighted support vector machine (WSVM), which can effectively classify-five types of vibration events in high-speed railway perimeter intrusion detection with small sample data and no parallel processing units. The method achieves 99 FPS and 98.8% accuracy on the test set with 330 vibration images as the training set to build the model without GPU and in the presence of interference signals. It provides a generalized 0-OTDR vibration event recognition method for small samples.

OPTICAL FIBER TECHNOLOGY (2023)

Add to Collection

Article Engineering, Electrical & Electronic

Markov Transition Fields and Deep Learning-Based Event-Classification and Vibration-Frequency Measurement for φ-OTDR

Xiaoting Zhao et al.

Summary: This paper proposes a novel method using Markov Transition Fields and deep learning to classify vibration events and measure vibration frequency. The method converts the normalized time series of a signal detected by phi-OTDR into MTF images, which are then classified using convolutional neural networks and fully connected neural networks. The experimental results demonstrate the effectiveness of this low-cost and fast method.

IEEE SENSORS JOURNAL (2022)

Add to Collection

Article Chemistry, Analytical

Real-Time φ-OTDR Vibration Event Recognition Based on Image Target Detection

Nachuan Yang et al.

Summary: This study proposes a real-time phi-OTDR multi-vibration events detection method based on computer vision, which enhances vibration signal features and trains an improved target detection model using a customized dataset, resulting in improved system performance and accuracy.

SENSORS (2022)

Add to Collection

Review Optics

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

Deus F. Kandamali et al.

Summary: The paper provides a review of the application of phase sensitive optical time-domain reflectometer (phi-OTDR) in monitoring vibrational signals, highlighting the challenges faced by the system and various techniques to overcome them. It focuses on the use of machine learning methods for event classification and evaluates the performance of different classification methods.

APPLIED OPTICS (2022)

Add to Collection

Article Chemistry, Analytical

Real-Time Multi-Class Disturbance Detection for Φ-OTDR Based on YOLO Algorithm

Weijie Xu et al.

Summary: This paper proposes a real-time multi-class disturbance detection algorithm based on YOLO for distributed fiber vibration sensing. The algorithm achieves real-time detection and classification of external intrusions sensed by distributed optical fiber sensing system (DOFS) based on phase-sensitive optical time-domain reflectometry (Phi -OTDR). The proposed model simplifies the traditional two-step object detection into a one-step process, using the Darknet53 network for both event localization and classification, resulting in improved detection speed and real-time operation. Experimental results show that this scheme provides a solution to real-time, multi-class external intrusion events detection and classification for practical applications of Phi -OTDR-based DOFS.

SENSORS (2022)

Add to Collection

Article Chemistry, Analytical

High-Accuracy Event Classification of Distributed Optical Fiber Vibration Sensing Based on Time-Space Analysis

Zhao Ge et al.

Summary: In this study, a convolutional neural network (CNN) approach is proposed for analyzing distributed optical fiber vibration sensing (DVS) data and achieving high-accuracy event recognition. Through extensive training with large datasets, the CNN is able to accurately classify multiple vibration events and achieve a recognition accuracy of 99.9% based on time-space data. Furthermore, the method demonstrates good generalization performance in practical applications.

SENSORS (2022)

Add to Collection

Article Optics

Temporal convolution network with a dual attention mechanism for phi-OTDR event classification

Manling Tian et al.

Summary: We propose a hybrid model named ATCN-SA-BiLSTM for phase sensitive optical time domain reflectometry signal recognition. The model combines channel attention based temporal convolutional network, spatial attention, and bidirectional long short-term memory network. Experimental results show that our method achieves better classification performance.

APPLIED OPTICS (2022)

Add to Collection

Article Optics

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

Yi Huang et al.

Summary: We propose a disturbance event recognition method based on region segmentation using differential phase signals of a phi-OTDR. The method achieves efficient recognition of disturbance events, with an average recognition accuracy of up to 92.9%. Compared to the method without region segmentation, this approach improves the average recognition accuracy by 8% and has a recognition time of only 0.39 s for three disturbance events on a 14.8-km sensing system.

APPLIED OPTICS (2022)

Add to Collection

Article Engineering, Electrical & Electronic

Unsupervised learning method for events identification in φ-OTDR

Jie Zhang et al.

Summary: An unsupervised-learning method for events-identification in phi-OTDR fiber-optic distributed vibration sensor is proposed in this paper. Different vibration-events, including blowing, raining, direct and indirect hitting, and noise-induced false vibration, are clustered using the k-means algorithm. An equivalent classification accuracy of 99.4% is achieved compared to the actual classes of vibration-events in the experiment. With a cluster-number of 3, the maximal Calinski-Harabaz index and Silhouette coefficient are obtained as 2653 and 0.7206, respectively. The research findings show that the clustering method is effective for events-identification in phi-OTDR without any prior labels, providing an interesting application of unsupervised-learning in self-classification of vibration-events for phi-OTDR.

OPTICAL AND QUANTUM ELECTRONICS (2022)

Add to Collection

Article Optics

Spatio-temporal joint oversampling-downsampling technique for ultra-high resolution fiber optic distributed acoustic sensing

Hao Li et al.

Summary: This study proposes a spatio-temporal joint oversampling-downsampling technique to suppress noise in coherent fiber distributed acoustic sensing systems. By utilizing spatial oversampling and averaging downsampling, noise decreases with increasing downsampling coefficient, ensuring correct quantization of noise frequency. Experimental results show significant improvement in DAS system performance with this technique.

OPTICS EXPRESS (2022)

Add to Collection

Article Chemistry, Analytical

Unsupervised Anomaly Detection Applied to Φ-OTDR

Antonio Almudevar et al.

Summary: This paper introduces a method based on Unsupervised Anomaly Detection techniques for processing signals from Distributed Acoustic Sensors (DASs) based on direct-detection Phi-OTDR. The method does not require human-labeled data for training, only event-free data is used, and has been tested with real data showing promising results.

SENSORS (2022)

Add to Collection

Article Optics

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

Yingzhe Huang et al.

Summary: This paper proposes a pattern recognition method for 9-OTDR based on self-reference features and applies machine learning to classify vibration signals. By analyzing the influence of different sample definition methods and feature parameters, the accuracy of pattern recognition is optimized. The paper also introduces a filtering algorithm for abnormal signal processing and conducts pattern recognition research on various events and mining equipment with high recognition rates.

APPLIED OPTICS (2022)

Add to Collection

Article Engineering, Electrical & Electronic

A Hybrid Model Integrating MPSE and IGNN for Events Recognition Along Submarine Cables

Tabi Fouda Bernard Marie et al.

Summary: This article proposes a developed hybrid model combining modern power spectrum estimation and improved gradient neural networks, to accurately identify and organize events along submarine cables. Signal characteristics are extracted in the frequency and time domains, and a multiclass IGNN classifier is used to determine the extent and position of collision events on the cable, resulting in real-time recognition and classification.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2022)

Add to Collection

Article Optics

Fading noise reduction in distributed acoustic sensing using an optimal weighted average algorithm

Yang Lu et al.

Summary: We present a novel approach for reducing fading noise in distributed acoustic sensing (DAS) by collecting Rayleigh signals in multiple optical modes and using an optimal weighted average algorithm to synthesize phase demodulated signals. The effectiveness of the proposed method is experimentally confirmed by comparing noise reduction in a conventional DAS system and one employing the new approach.

APPLIED OPTICS (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Pattern Recognition of Optical Fiber Vibration Signal of the Submarine Cable for Its Safety

Bernard Marie Tabi Fouda et al.

Summary: This article presents a method for detecting and measuring shock events on submarine cables using vibration signals from optical fiber sensors, which shows high accuracy and reliability in recognizing and determining events on the submarine cable.

IEEE SENSORS JOURNAL (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Research on Recognition Method of Railway Perimeter Intrusions Based on -OTDR Optical Fiber Sensing Technology

He Meng et al.

Summary: A novel data-driven intrusion identification method based on the XGBoost model is proposed to reduce the nuisance alarm rate in railway safety monitoring. The study successfully implemented distributed optical fiber sensing in the railway sector, achieving an average recognition accuracy of 98.5% in frequent external intrusion events and outperforming other popular methods.

IEEE SENSORS JOURNAL (2021)

Add to Collection

Article Instruments & Instrumentation

Event identification of a phase-sensitive OTDR sensing system based on principal component analysis and probabilistic neural network*

Xin Wang et al.

Summary: In order to reduce the nuisance alarm rate in phase-sensitive OTDR sensing system, a novel event identification model based on PCA and PNN was proposed. By training this model, five kinds of disturbance events can be effectively identified with an average identification rate of 97.74% and an average response time of 0.93 s. The high identification rate and fast response time make this method more adaptable in practical application.

INFRARED PHYSICS & TECHNOLOGY (2021)

Add to Collection

Article Optics

Interference fading suppression in φ-OTDR using space-division multiplexed probes

Zhiyong Zhao et al.

Summary: A novel interference fading suppression method for phi-OTDR using space-division multiplexed pulse probes in a few-mode fiber is proposed and experimentally demonstrated. The method significantly improves performance by reducing fading probabilities over fiber length and time scale. By introducing the concept of SDM to phi-OTDR, the proposed method avoids complex modulation techniques, offering simplicity, effectiveness, and high reliability.

OPTICS EXPRESS (2021)

Add to Collection

Review Engineering, Electrical & Electronic

Principle and Application State of Fully Distributed Fiber Optic Vibration Detection Technology Based on φ-OTDR: A Review

Tabi Fouda Bernard Marie et al.

Summary: Phi-OTDR is a powerful sensing device for fully distributed optical fiber vibration detection, known for its fast response and high sensitivity. The article outlines some current gaps in Phi-OTDR technology and proposes solutions. Its applications include perimeter security monitoring systems, submarine cable surveillance, and more.

IEEE SENSORS JOURNAL (2021)

Add to Collection

Article Engineering, Electrical & Electronic

A Novel Method of Recognizing Disturbance Events in Φ-OTDR Based on Affinity Propagation Clustering and Perturbation Signal Selection

Shaohua Xu et al.

Summary: This paper proposed a novel recognition method for identifying perturbation signals in the Phi-OTDR system, using affinity propagation clustering and variational mode decomposition techniques; the effectiveness of a novel perturbation signal selection technique was experimentally verified; the method reliably and accurately identifies different perturbation signal categories.

IEEE SENSORS JOURNAL (2021)

Add to Collection

Article Engineering, Electrical & Electronic

An Easy Access Method for Event Recognition of φ-OTDR Sensing System Based on Transfer Learning

Yi Shi et al.

Summary: Traditional event recognition methods for signal collected by phi-OTDR sensing system struggle to accurately identify event categories, while deep learning methods require extensive computation and training. The proposed transfer training method in this paper efficiently builds a high-precision event recognition network, achieving high classification accuracy through simple preprocessing and training.

JOURNAL OF LIGHTWAVE TECHNOLOGY (2021)

Add to Collection

Article Optics

High-fidelity acoustic signal enhancement for phase-OTDR using supervised learning

Fei Jiang et al.

Summary: This paper proposes using supervised learning to simultaneously address noise issues caused by laser frequency drift, laser phase noise, and interference fading, by training an end-to-end convolutional neural network to enhance acoustic signals. After enhancement, the average signal-to-noise ratio (SNR) of the recovered PZT vibration signals increased from 13.4 dB to 42.8 dB, and the average scale-invariant signal-to-distortion ratio (SI-SDR) of the recovered speech signals improved by 7.7 dB.

OPTICS EXPRESS (2021)

Add to Collection

Review Engineering, Electrical & Electronic

Pattern Recognition for Distributed Optical Fiber Vibration Sensing: A Review

Junchan Li et al.

Summary: The paper discusses the pattern recognition technologies for distributed optical fiber vibration sensing, including sensor introduction, pattern recognition models, performance comparison, and application areas.

IEEE SENSORS JOURNAL (2021)

Add to Collection

Article Engineering, Electrical & Electronic

Man-Made Threat Event Recognition Based on Distributed Optical Fiber Vibration Sensing and SE-WaveNet

Mingyang Sun et al.

Summary: Real-time threat identification has been a focal point in the security field, with researchers developing the SE-WaveNet deep learning model to efficiently address risks in threat event identification. By introducing the SE structure of the attention mechanism, the SE-WaveNet adapts to matrix data with complete information. Experimental results show that SE-WaveNet outperforms other models in accuracy, model size reduction, and processing speed.

IEEE TRANSACTIONS ON INSTRUMENTATION AND MEASUREMENT (2021)

Add to Collection

Article Optics