Accuracy improvement in peak positioning of spectrally distorted fiber Bragg grating sensors by Gaussian curve fitting.

To improve measurement accuracy of spectrally distorted fiber Bragg grating temperature sensors, reflection profiles were curve fitted to Gaussian shapes, of which center positions were transformed into temperature information. By applying the Gaussian curve-fitting algorithm in a tunable bandpass filter demodulation scheme, approximately 0.3 degrees C temperature resolution was obtained with a severely distorted grating sensor, which was much better than that obtained using the highest peak search algorithm. A binary search was also used to retrieve the optimal fitting curves with the least amount of processing time.