In the processing of recognizing Φ-OTDR event signals by deep learning models, a large amount of annotated data is required for model training. However, the data annotation is a major challenge for a lot of unlabeled data and the labeling cost are high. This article proposes a valuable signal data selection method for Φ-OTDR sensing system. The data samples that are beneficial for training from the unlabeled data set is selected based on their uncertainty values. Through only labeling these valuable data samples and training the classification model with them, the labeling cost can be saved a lot and the classification model can keep its classification accuracy. The experiments show that this method can save 83.77% on manual labeling cost and the classification accuracy can still reach 96.37%, which is similar to the classification accuracy of the model trained by labeling all the data.

This paper proposes a multi-radial-distance event classification method based on deep learning. To the best of our knowledge, this is the first time that the $\Phi $ -OTDR can tell how far the target event from the sensing fiber is through deep learning approach. The temporal-spatial data matrix collected by the system is filtered by three different band-pass filters to form RGB images as the input of the Inception_V3 network trained by ImageNet dataset. The passband of three band-pass filters is selected by searching the maximum Euclidean distance in the frequency domain. Three kinds of filters with different frequency bands enhance the effective features of data samples in advance. The simulated annealing (SA) algorithm is applied to search the maximum Euclidean distance. Field experiment includes five kinds of events with four different radial distances, where there are 17 subclasses in total, has been carried out. The classification results show that the classification accuracy reaches 86% and the method can tell both the event type and radial distance.

The deep integration of communication and sensing technology in fiber-optic systems has been highly sought after in recent years, with the aim of rapid and cost-effective large-scale upgrading of existing communication cables in order to monitor ocean activities. As a proof-of-concept demonstration, a high-degree of compatibility was shown between forward-transmission distributed fiber-optic vibration sensing and an on–off keying (OOK)-based communication system. This type of deep integration allows distributed sensing to utilize the optical fiber communication cable, wavelength channel, optical signal and demodulation receiver. The addition of distributed sensing functionality does not have an impact on the communication performance, as sensing involves no hardware changes and does not occupy any bandwidth; instead, it non-intrusively analyzes inherent vibration-induced noise in the data transmitted. Likewise, the transmission of communication data does not affect the sensing performance. For data transmission, 150 Mb/s was demonstrated with a BER of 2.8 × 10⁻⁷ and a Q_dB of 14.1. For vibration sensing, the forward-transmission method offers distance, time, frequency, intensity and phase-resolved monitoring. The limit of detection (LoD) is 8.3 pε/Hz^1/2 at 1 kHz. The single-span sensing distance is 101.3 km (no optical amplification), with a spatial resolution of 0.08 m, and positioning accuracy can be as low as 10.1 m. No data averaging was performed during signal processing. The vibration frequency range tested is 10–1000 Hz.

The structural health of fiber-optic communication networks has become increasingly important due to their widespread deployment and reliance in interconnected cities. We demonstrate a smart upgrade of a communication system employing a dual-polarization-state polarization shift keying (2-PolSK) modulation format to enable distributed vibration monitoring. Sensing can be conducted without hardware changes or occupying additional communication bandwidth. Experimental results demonstrate that forward transmission-based distributed vibration sensing can coexist with PolSK data transmission without significant deterioration in performance. This proof-of-concept study achieved a sensitivity of 0.4141 μV/με with a limit of detection (LoD) of 563 pε/Hz^1/2@100 Hz. The single-span sensing distance can reach up to 121 km (no optical amplification) with a positioning accuracy as small as 874 m. The transmission rate is 300 Mb/s, the Q_dB is 16.78 dB, and the corresponding BER is 5.202 × 10⁻¹². For demonstration purposes, the tested vibration frequency range is between 100 and 200 Hz.

Fiber-optic sensing systems based on a forward transmission interferometric structure can achieve high sensitivity and a wide frequency response over long distances. However, there are still shortcomings in its ability to position multi-point vibrations and detect low-frequency vibrations, which limits its usefulness. To address these challenges, we study the viability of merging long-range forward-transmission distributed vibration sensing (FTDVS) with high spatial resolution optical frequency-domain reflectometry (OFDR), forming the first reported hybrid distributed sensing method between these two methods. The probe light source is shared between the two sub-systems, which utilizes stable linear optical frequency sweeping facilitated by high-order sideband injection locking. As a result, this is a new approach for the FTDVS method, which conventionally uses fixed-frequency continuous light. The method of nearest neighbor signal replacement (NSR) is proposed to address the issue of discontinuity in phase demodulation under periodic external modulation. The experimental results demonstrate that the hybrid system can determine the position of vibration signals between 0 and 900 Hz within a sensing distance of 21 km. When the sensing distance is extended to 71 km, the FTDVS module can still function adequately for high-frequency vibration signals. This hybrid architecture offers a fresh approach to simultaneously achieving long-distance sensing and wide frequency response, making it suitable for the combined measurement of dynamic (e.g., gas leakage, pipeline excavation warning) and quasi-static (e.g., pipeline displacement) events in long-distance applications.