The phase sensitive optical time-domain reflectometer (φ-OTDR), or in some applications called distributed acoustic sensing (DAS), has been a popularly used technology for long-distance monitoring of vibrational signals in recent years. Sinceφ-OTDR systems usually operate in complicated and dynamic environments, there have been multiple intrusion event signals and also numerous noise interferences, which have been a major stumbling block toward the system’s efficiency and effectiveness. Many studies have proposed different techniques to mitigate this problem mainly inφ-OTDR setup upgrades and improvements in data processing techniques. Most recently, machine learning methods for event classifications in order to help identify and categorize intrusion events have become the heated spot. In this paper, we provide a review of recent technologies from conventional machine learning algorithms to deep neural networks for event classifications aimed at increasing the recognition/classification accuracy and reducing nuisance alarm rates (NARs) inφ-OTDR systems. We present a comparative analysis of the current classification methods and then evaluate their performance in terms of classification accuracy, NAR, precision, recall, identification time, and other parameters.

This manuscript proposes a method based on back propagation (BP) neural network and the spectral subtraction method to quickly obtain sensing information in Brillouin fiber optics sensors. BP neural network’s characteristics which can realize any complex nonlinear mapping help to determine the frequency shift section(s) information. The training function, transfer function and number of hidden layer nodes of BP neural network are determined with experimental data. The experimental results show that comparing with traditional Lorentz fitting algorithm and edge detection with Sobel operator, the BP neural network is about 1/12 in terms of time complexity with the Lorentz algorithm, about 1/9 with the edge detection based on Sobel operator; while the respective accuracy on determine the frequency shifted section(s) has improved by 79.4% and 27.9%.

Anisotropic diffusion (AD) is employed to enhance the signal-to-noise ratio (SNR) of Brillouin distributed optical fiber sensor. A Brillouin optical time-domain analyzer (BOTDA) with 99-km-long fiber under test is set up, where a section of 2.3 m and another 182 m section at the end of the fiber are heated for experimental verification. The SNR of experimental data sets with different sampling point numbers are enhanced to several improvement levels by AD and three other methods for comparison. Results show that the 2.3 m section and the temperature transition region of the 182 m section are better preserved by AD than other methods for the same SNR improvement. Objective criteria analysis shows that AD achieves the best measured spatial resolution and temperature accuracy among the four methods, the location of temperature transition can be detected more accurately for data with low SPNs after AD denoising. In addition, the processing time of AD is 1/3 that of non-local means (NLM) and 6% that of block-matching and 3D filtering (BM3D). The edge-preserving quality and fast processing speed allow the proposed algorithm to be a competitive denoising alternative for sensors based on Brillouin scattering.

We propose a hybrid model named channel attention based temporal convolutional network combined with spatial attention and bidirectional long short-term memory network (ATCN-SA-BiLSTM) for phase sensitive optical time domain reflectometry signal recognition. This hybrid model consists of three parts: ATCN, which extracts temporal features and preserves causality of time domain signals, the SA mechanism, which re-weights spatial sequences for better feature extraction, and BiLSTM, which extracts spatial relationships considering the bidirectional propagation characteristics of disturbances in space domain signals. Experimental results show that our method achieves better classification performance with an accuracy of 93.4% and zero nuisance alarm rate.