Features of the Implementation of the Extremal Filtration Method in the Distributed Optic-Fiber Temperature Sensor

The signal processing method based on extremal filtration in low signal-to-noise ratio conditions was proposed. The extremal filtration is similar to the empirical mode decomposition (EMD) method, but complex mathematical calculations are absent that is the main advantage. Using it in real measuring devices significantly increases the speed rate. The extremal filtration process is an algorithm that involves sequentially repeating until an optimal result. The disadvantage of this method is decreasing the spatial resolution since every iteration due to a reduction in the number of samples. It is an especially negative fact in areas of sharp signal change. This paper demonstrates how to solve this problem using linear interpolation and to improve the result obtained by simple extremal filtration.