Fiber Bragg grating interrogation using FBG filters and artificial neural network

Optical fiber sensors have become a popular alternative to traditional electronic sensors due to their numerous advantages. An important challenge in deploying optical sensors is the interrogation of the sensor, that is, recovering the measured value from the sensor output. This paper aims to present a simple yet effective way of interrogating a fiber Bragg grating (FBG) temperature sensor using optical filters and an artificial neural network (ANN). This interrogation system is capable of giving the precise temperature value without directly measuring the resonance wavelength shift or performing any Fourier calculations. The network was implemented and the training was accomplished using simulated data. Simulated results are presented and compared to traditional methods of interrogation. The system proposed in this paper showed excellent performance in identifying the temperature from the sensor output and showed more precision than the traditional method.