Enhanced frequency and time domain feature extraction for communication infrastructure type classification using optical fiber sensing

• Published in: Optical fiber technology 2024-10, Vol.87, p.103859, Article 103859
• Main Authors: Inoue, Masaaki, Koshikiya, Yusuke
• Format: Article
• Language: English
• Subjects: Fiber sensing; Neural networks; Optical fiber communication; Vibration measurement
• ISSN: 1068-5200
• DOI: 10.1016/j.yofte.2024.103859

•Multi-class classification targeting infrastructure types in the field was performed.•Feature extraction designed for a sensing technique with conventional optical fiber.•Spectral envelope shape offers the prospect of improved accuracy.•Classification using both frequency and time-domain features show good performance.•Reclassification by considering the similarity of neighboring classes. We present a feature extraction method for a feedforward-type neural network (FNN) designed to realize a distributed acoustic sensing (DAS) technique that suits conventional optical fiber. This is, to the best of our knowledge, the first trial in which an FNN is used to interpret the field communication infrastructure type surrounding optical cables. Three classes are taken to represent the field environment: the cable tunnel (class 1), circular duct (class 2), and overhead area (class 3). We investigate and compare frequency- and time-domain feature extraction. We also show that the frequency-domain features yielded by spectral envelope shape (SES) processing have better performance than simple fast Fourier transform features. Two types of time-domain features are verified: one is the short-time maximum magnitude (STMM), which shows the largest value in the time frame, and the other is the short-time average magnitude (STAM), which indicates the average value in a time frame. Note that all features are optimized for multi-class classification. In this paper, we present the suitable number of both features and the number of training iterations. An accuracy rate of 79.0% is achieved using FNN analysis with the features studied here. Furthermore, by considering the similarity of neighboring classes, classes are refined into higher probability classes. As a result, accuracy is improved to 87.2%.