Research On Φ-OTDR Optical Fiber Sensing Technology Based On Compressed Sensing

The distributed optical fiber sensing system based on phase-sensitive optical time-domain reflectometry(Φ-OTDR)technology has been a research hotspot in related fields for many years because of its advantages such as long detection distance,high spatial resolution,strong corrosion resistance,and anti-interference ability.However,the Φ-OTDR system generates a huge amount of data when sampling the Rayleigh backscattering signals,which is not conducive to data processing and transmission.Meanwhile,its maximum detectable frequency is limited by the Nyquist sampling theorem and the sensing distance.In this thesis,the compressed sensing algorithm is used to reduce the required data storage,and the additive random sampling is used to improve the frequency detection range of the system.A Φ-OTDR optical fiber sensing system was built,and vibration signal sensing was carried out using uniform sampling method and moving average-moving differential signal positioning algorithm,which verified the performance of the sensor system and the feasibility of the positioning algorithm.Using the compressed sensing theory,the vibration sensing data of the Φ-OTDR system was compressed and recovered,and the data storage and recovery were realized,which are much lower than the Nyquist sampling rate.The success probabilities of the signal recovery using different sparse matrices and observation matrices are compared,and the reasons for the different success probabilities are analyzed.Without changing the original hardware equipment,the sensing signal was sampled by the additive random undersampling method,and the feasibility of using additive random sampling to detect high frequency vibration signals in Φ-OTDR sensor was verified from both numerical calculation and experiment.In this thesis,good calculation and experiment results have been achieved on both using the compressed sensing algorithm for data compression and using the additive random sampling for frequency expansion without changing the hardware structure and the processing bandwidth.This reveals the feasibility of applying these two techniques in Φ-OTDR sensing and provides a basis for their practical applications.