Research On Signal Processing And Coding Methods In Fiber Sensing By Brillouin Optical Time Domain Reflectometry

Brillouin Optical Time Domain Reflectometry (BOTDR) is a spontaneous Brillouin scattering based distributed optical fiber sensing technology, which can be used to monitor temperature and strain along an optical fiber and has the merit of one-end-access and the ability of breakpoint measurement. The works in this thesis focus on signal processing and coding methods in BOTDR, and concentrate on the optimization and enhancement of BOTDR system.First, the time-frequency analysis of the broadband Brillouin signal in BOTDR is highlighted. A novel series of signal processing methods called Cohen’s class is proposed, in order to reduce the trade-off between the spatial resolution and the frequency accuracy of BOTDR. This trade-off is a drawback of the existing signal processing methods. The influence of the Heisenberg-Gabor uncertainty principle on the existing methods and the cause of the trade-off are interpreted, and the advantage of Cohen’s class over the existing ones is illuminated. The stimulated and experimental broadband Brillouin signals are analyzed with short-time Fourier transform and Cohen’s class, respectively. The comparison between the processing results verifies the validity and the advantage of Cohen’s class.Then the effect of direct current component of the pulse light on BOTDR is investigated theoretically and demonstrated experimentally. It is proved that in a frequency-scanning BOTDR using RF demodulation, the leakage light gives rise to additional noise term, and affects the measurement accuracy of Brillouin frequency shift.Finally, the coding methods in BOTDR are explored. The existing methods are summarized. The influence of the return-to-zero and non-return-to-zero codes on stimulated Brillouin scattering threshold is compared experimentally. Additionally, the feasibility of applying the optical orthogonal codes in BOTDR is discussed.