Noise Reduction for Time-Domain Sensing Signal of Brillouin Scattering Based on Time Series Analysis and Kalman Filter Algorithm

To improve the accuracy of Brillouin time-domain reflection sensing signal to address precise strain and temperature sensing issues of overhead transmission lines or fibers, we propose a hybrid data noise reduction algorithm combining time series analysis with Kalman filter. The basic idea of the algorithm is to utilize the time series analysis theory to build a non-stationary model for compound strain and temperature variance rules, which can be considered as the state transition equation of Kalman filter. Secondly, we can obtain the observation equation of the Kalman filter via BOTDR technology measuring the strain and temperature of the sensing fiber. Then, the strain and temperature sensing and sampling data are processed by the recursive filtering of Kalman prediction. In a word, one can promptly and accurately locate or predict the icing and lightning strike on the overhead transmission lines or fibers with our proposed algorithm. According to the experimental results, the accuracy and rapid convergence of our algorithm is verified.