Cost-Effective Wavelength Demodulation for Dual-Parameter FBG Sensors Using LSTM Networks

Fiber Bragg Grating (FBG) sensors are widely used for dual-parameter monitoring of strain and temperature. Traditional demodulation methods, however, are often complex and costly. In this work, we present a demodulation framework that converts wavelength shifts into transmitted intensity using an arrayed waveguide grating, followed by an LSTM network to establish the intensity-wavelength relationship. The experimental results show that the model achieves a root-mean-square error of 1.28 pm for the wavelength demodulation task for temperature and strain, a result that validates the high accuracy of the system.