Abstract

Distributed acoustic sensors based on chirped-pulse phase sensitive-optical time-domain reflectometry (chirped-pulse FOTDR) have proven capable of performing fully distributed, single shot measurements of true strain or temperature perturbations, with no need for frequency scanning or phase detection methods. The corresponding refractive index variations in the fiber are revealed in the chirped-pulse FOTDR trace through a local temporal shift, which is evaluated using trace-to-trace correlations. The accuracy in the detection of this perturbation depends upon the correlation noise and the coherence of the laser source. In this paper, we theoretically and experimentally analyze the impact of the laser phase noise in chirped-pulse FOTDR. In particular, it is shown that the noise in the readings of strain/temperature variations scales directly with the frequency noise power spectral density of the laser. To validate the developed model, an experimental study has been performed using three lasers with different static linewidths (5 MHz, 50 kHz, and 25 kHz), i.e., with different phase noise. Besides, we present a simple technique to mitigate the effect of the laser phase noise in chirped-pulse FOTDR measurements. The proposed procedure enables to detect perturbations with high signal-to-noise ratio even when using relatively broad linewidth (i.e., comparatively high phase noise) lasers. Up to 17 dB increase in signal-to-noise ratio has been experimentally achieved by applying the proposed noise cancellation technique.