Wavelet-based data processing for distributed fiber optic sensors

Distributed fiber optic sensing technique is considered as a promising tool to monitor continuously structural state in civil engineering. Distributed sensors produce large quantities of tempo-spatial data. The ability of wavelet transforms to detect abrupt changes, make them well suited for the analysis of monitoring data. This paper presents tempo-spatial wavelet-based data processing method for strain data measured by BOTDR (Brilliouin optical time domain reflectometer), a distributed fiber optic sensing technique based on Brilliouin scattering. First, Wavelet analysis is used to reduce the noise of the measured data, Then, Wavelet transform is used to detect abrupt strain changes in spatial and time domain respectively. A numerical example and monitoring strain data of some concrete tunnel are used to demonstrate feasibility of the proposed method.