Abstract In this paper, we propose a novel neural network model named by Multi-branch Long Short-Time Memory Convolution Neural Network (MLSTM-CNN) for identifying disturbance signals in distributed optical fiber sensing system based on phase-sensitive optical time domain reflectometry (φ-OTDR). By unifying feature extraction and classification in a framework, MLSTM-CNN automatically extracts features at different time scales leveraging multi-branch layer and learnable LSTM layers, and then the disturbance signals are identified in the learnable CNN layers. Through constructing 25.05 km φ-OTDR experimental system, four kinds of real disturbance events, including watering, climbing, knocking, and pressing, and a false disturbance event can be effectively identified. Experimental results show that the average identification rate can reach 95.7%, and nuisance alarm rate (NAR) is 4.3%. Compared with the LSTM and CNN model, the recognition accuracy of the proposed model can be improved and the signal processing time can be efficiently reduced as well.

中文翻译：

---

基于φ-OTDR的光纤分布式扰动传感器事件识别的多分支长短时记忆卷积神经网络

摘要 在本文中，我们提出了一种名为多分支长短时记忆卷积神经网络（MLSTM-CNN）的新型神经网络模型，用于基于相敏光时域反射法识别分布式光纤传感系统中的干扰信号。 φ-OTDR）。通过在一个框架中统一特征提取和分类，MLSTM-CNN 利用多分支层和可学习 LSTM 层自动提取不同时间尺度的特征，然后在可学习 CNN 层中识别干扰信号。通过搭建25.05kmφ-OTDR实验系统，可以有效识别4种真实扰动事件，包括浇水、爬升、敲击和挤压，以及虚假扰动事件。实验结果表明，平均识别率可达95.7%，滋扰警报率 (NAR) 为 4.3%。与 LSTM 和 CNN 模型相比，该模型的识别准确率可以得到提高，并且信号处理时间也可以有效减少。