This paper presents a vehicle detection and classification system using distributed fiber-optic acoustic sensing (DAS) technology and describes a comprehensive classification method including signal processing and feature extraction. This sensing device is based on Rayleigh scattering light and is used for real-time vehicle detection, classification, and speed estimation. Distributed acoustic signals from an arbitrary point can be detected and located through DAS technology which can provide fully distributed acoustic information along the entire fiber link. This technology utilizes sensing fiber in the form of distributed sensors to collect traffic vibration signals and then extracts several features from the signals to estimate the vehicle count and identify vehicle categories. According to the vehicle vibration signal characteristics, the wavelet-denoising algorithm and dual-threshold algorithm are improved. The improved algorithm is used to reconstruct the signal for feature extraction, and the vehicle count and speed are obtained. When all features have been extracted, the classification of vehicle types is implemented by a support vector machine classifier. The validation data (using a distributed fiber-optic acoustic sensor) demonstrate that the vehicle detection accuracy is higher than 80%, the speed estimation error is less than 5%, and the vehicle classification accuracy is higher than 70%.

An improved phase-sensitive optical time-domain reflectometry (φ-OTDR) system with sub-meter spatial resolution is demonstrated. Two Michelson interferometers (MIs) with different path length differences are used in the proposed system. One is 10 m, the other is 9.2 m. Two Rayleigh backscattering phase traces with different spatial resolution are obtained by a phase generated carrier (PGC) algorithm at adjacent times. After using differencing and adaptive 2-D bilateral filtering algorithms, a 0.8-m spatial resolution over 2 km is achieved. Experimental results indicate that the system shows an extraordinary linearity as high as 99.94% with amplitude-modulation and acquires a detection frequency from 5 to 500 Hz.

The key point on analyzing the data stream measured by fiber optic distributed acoustic sensing (DAS) is signal activity detection separating measured signals from environmental noise. The inability to calculate the threshold for signal activity detection accurately and efficiently without affecting the measured signals is a bottleneck problem for current methods. In this article, a novel signal activity detection method with the adaptive-calculated threshold is proposed to solve the problem. With the analysis of the time-varying random noise’s statistical commonality and the short-term energy (STE) of real-time data stream, the top range of the total STE distribution of the noise is found accurately for real-time data stream’s ascending STE, thus the adaptive dividing level of signals and noise is obtained as the threshold. Experiments are implemented with simulated database and urban field database with complex noise. The average detection accuracies of the two databases are 97.34% and 90.94% only consuming 0.0057 s for a data stream of 10 s, which demonstrates the proposed method is accurate and high efficiency for signal activity detection.