Binary encoding to identify overlapped FBGs in a sensor network using neural network using SPDM

Abstract Fiber Bragg Gratings (FBGs) are used in series sensor networks for structural health monitoring. To increase the number of FBGs in a sensor network, overlapped FBGs are employed. However, the spectra of overlapped FBGs shadow one another and, therefore, cannot be interrogated using conventional multiplexing techniques. In the literature, multiplexing schemes employing machine learning techniques have been used to interrogate sensor networks with overlapped FBGs subjected to different strains or temperatures. However, to the best of the authors’ knowledge, there is very limited literature work on overlapped FBG sensor networks where more than one pair of FBG is subjected to equal strain or temperature. There is also very limited literature on identifying the overlapped FBGs undergoing equal strain. In this paper, we propose a binary encoding method to determine the position and the central wavelength of overlapped FBGs undergoing strain using a neural network and SPDM. The proposed method achieved 100% accuracy in simulations and 92.30% accuracy in experimental studies for measuring the central wavelength of FBG for strain measurement in a series sensor network with overlapped FBGs.