Distributed Brillouin frequency shift extraction via a convolutional neural network

作     者：YIQING CHANG HAO WU CAN ZHAO LI SHEN SONGNIAN FU MING TANG 

作者机构：Wuhan National Laboratory for Optoelectronics(WNLO)&National Engineering Laboratory for Next Generation Internet Access SystemSchool of Optical and Electronic InformationHuazhong University of Science and TechnologyWuhan 430074China 

出 版 物：《Photonics Research》 (光子学研究（英文版）)

年 卷 期：2020年第8卷第5期

页      面：690-697页

核心收录：

学科分类：08[工学] 080202[工学-机械电子工程] 0802[工学-机械工程] 0702[理学-物理学] 

基　　金：National Key Research and Development Program of China(2018YFB1801002) National Natural Science Foundation of China(61722108,61931010) Innovation Fund of WNLO。 

主　　题：neural network individual 

摘      要：Distributed optical fiber Brillouin sensors detect the temperature and strain along a fiber according to the local Brillouin frequency shift(BFS),which is usually calculated by the measured Brillouin spectrum using Lorentzian curve fitting.In addition,cross-correlation,principal component analysis,and machine learning methods have been proposed for the more efficient extraction of BFS.However,existing methods only process the Brillouin spectrum individually,ignoring the correlation in the time domain,indicating that there is still room for improvement.Here,we propose and experimentally demonstrate a BFS extraction convolutional neural network(BFSCNN)to retrieve the distributed BFS directly from the measured two-dimensional data.Simulated ideal Brillouin spectra with various parameters are used to train the BFSCNN.Both the simulation and experimental results show that the extraction accuracy of the BFSCNN is better than that of the traditional curve fitting algorithm with a much shorter processing time.The BFSCNN has good universality and robustness and can effectively improve the performances of existing Brillouin sensors.