Distributed Sensor Calibration by Gaussian Approximation

Abstract

The calibration process is the key factor that contributes to the accuracy and repeatability of the sensor operation in general. To improve the performance of distributed sensors in optical fibers, we will present in this paper an alternative approach to performing their calibration. The proposed calibration method consists of an algorithm that represents the sensor signal by a sum of Gaussian functions and then determines the new Gaussian parameters for the signal to be corrected. Such methodology allows obtaining different spatial resolutions according to the precision range required. For example, spatial resolutions of 29.82 cm and 1.8 cm were obtained for hot spots of 50 cm and 3 cm, respectively.