Iterative filtering for Time-Frequency Localised Pulse Optimisation in STFT-BOTDR

Dynamic strain measurement in distributed fibre optic sensing (DFOS) is essential for structural health monitoring (SHM) of the strain changes induced by construction failure and other activities in infrastructure’s life cycle. Among different DFOS systems, the Short Time Fourier Transform-Brillouin Optical Time Domain Reflectometry (STFT-BOTDR) takes the advantages of STFT obtaining full frequency spectrum to improve the performance of conventional BOTDR, providing an opportunity for dynamic sensing. The key parameters of distributed fibre optic sensors, spatial and frequency resolution, are strongly linked with the pulse’s time-frequency localisation. In this paper, a set of Kaiser-Bessel functions is used to simulate different pulse shapes and compare their parameters in terms of Time-Frequency Localisation (TFL) and their Brillouin scattering spectrum. A method using iterative filtering algorithm to achieve the optimised pulse in terms of TFL is introduced to converge the Effective-pulse Width (TEW) in time-domain and Effective-pulse Linewidth (FEL) in the frequency domain, respectively, to the fundamental limitation. The optimised pulse can be fitted with 7^th order Gaussian (super-Gaussian) shape and offer the best experimental performance compared to Rectangular pulseperformance compared to Rectangular pulse.