Distributed Temperature Sensor With Fine Gauge Length Based on Adaptive Morphological Processing

In this study, a fine gauge length distributed temperature sensor based on adaptive morphological processing (AMP) is proposed and experimentally demonstrated. During the data processing in optical frequency-domain reflectometry (OFDR), a reduction in gauge length leads to a decrease in the number of data points participating in the cross correlation calculation. This reduction in sample size negatively impacts the correlation quality, subsequently giving rise to increased demodulation errors, which are detrimental to the accuracy of the measurements. In the data processing flow of the proposed method, the correlation results of the reference and measured data within each gauge length are arranged as a results array along the sensing distance. Based on the spatial continuity and local similarity of cross correlation results for temperature sensing, cruciform and linear structural elements are chosen to perform AMP of the array, suppressing bad points and false peaks under fine gauge length. In experiments, the temperature gradient information from 43 °C to 55 °C is reconstructed free of outliers under the gauge length of 2 mm located at the end of 40-m-long sensing fiber. The mean absolute error (MAE) and standard deviation (STD) of the temperature measurement system using AMP with a gauge length of 2 mm has an approximate error level to conventional method with 6-mm gauge length, proving that the gauge length of the system is improved by a factor of three. The experimental results demonstrate that the proposed method provides a new solution idea for realizing distributed temperature measurement with fine gauge length based on OFDR.