Wavelength detection in spectrally overlapped FBG sensor network using extreme learning machine

• Main Authors: Jiang, Hao, Chen, Jing, Liu, Tundong
• Other Authors: School of Electrical and Electronic Engineering
• Format: Article
• Language: English
• Published: 2015
• Subjects: DRNTU::Engineering::Electrical and electronic engineering::Optics, optoelectronics, photonics
• Online Access: https://hdl.handle.net/10356/103790; http://hdl.handle.net/10220/24580
• Institution: Nanyang Technological University

This paper presents a novel learning-based method called extreme learning machine (ELM) to solve the Bragg wavelength detection problem in the fiber Bragg grating (FBG) sensor network. Based on building up a regression model, the proposed approach is divided into two phases: offline training phase and online detection phase. Due to the good generalization capability of ELM, the well-trained detection model can directly and accurately determine the Bragg wavelengths of the sensors even when the spectra of FBGs are completely overlapped. The results demonstrate that the proposed method is efficient and stable. It has shown competitive advantages in terms of the detection accuracy, the offline training speed as well as the real time detection efficiency.