Research On Performance Improvement Of Distributed Optical Fiber Sensing System By Using Digital Signal Processing

The paper mainly discusses the research on the performance improvement of the distributed optical fiber sensing system by using the digital signal processing(DSP)technique. In the paper, the distributed optical fiber sensing system int study consists of the phase demodulation of phase sensitive optical time domain reflectometry(?-OTDR), coherent optical time domain reflectometry(COTDR) and Brillouin optical time domain analysis(BOTDA). And, these methods and means used in the above three kinds of distributed sensing system are not all the same.In the phase demodulation of ?-OTDR system, system model is composed of infinitesimal scattering model, I/Q demodulation and phase demodulation algorithm. It is simulated by using numerical simulation technology. In the simulation, it needs to make the parameters in the mathematical expression which is converted from system model consistent with the actual physical values before successful demodulation the strain information imposed on the ordinary single mode fiber. In order to improve the signal-to-noise ratio(SNR) of the strain demodulation signal, the author of the paper puts forward the phase estimation phase enhancement algorithm, and it is applied to the performance study of the phase demodulation of ?-OTDR system about light source line width. Both the direct phase difference of two points and the average phase estimation phase enhancement algorithms have a consistent conclusion, namely, with the increase of the light source linewidth, the SNRs of the demodulation signals demodulated by both algorithms are declining. Another conclusion is that the SNRs of signals demodulated by the average phase estimation phase enhancement algorithm are larger than that of the direct phase difference of two points. The author of the paper programmes the real-time phase demodulation ?-OTDR software system based on Labview. With the help of the software system, the PZT experiments have been done.And, the obtained experimental conclusion is as follows, within a certain frequency response and amplitude response range, the frequency response values is approximately equal and amplitude response has a good linear characteristic.In the study of COTDR system, at first, the author of the paper establishes corresponding COTDR system model, and then applies the cross-correlation data processing algorithm. At last, the author realizes the COTDR system simulation. In thispaper, the author puts forward an improved cross-correlation data processing algorithm.When compared with the cross-correlation data processing algorithm, the improved cross-correlation data processing algorithm expands the measuring ranges of the coherent optical time domain reflectometry system nearly 2 times. In this paper, the author also puts forward the optimized optical path, it can get higher quality of demodulation signals theoretically. In the experimental part, the author programmes the improved cross-correlation data processing algorithm in COTDR system software based on Labview and obtains theory verification through experimental results.At this part, the author also has carried on the experimental BOTDA study. The classic nonlocal mean filtering algorithm in digital image processing technique is introduced to Simplex coding BOTDA data processing algorithm. The experimental data collected from BOTDA experimental system is processed by conventional algorithm and conventional algorithm added the nonlocal mean filtering algorithm,respectively. The conclusion is that, not only the advantage of the classic NLM filtering algorithm is observed from the image results processed by conventional algorithm and conventional algorithm added the nonlocal mean filtering algorithm, but also it is quantitatively highlighted by the computational measurement uncertainty of Brillouin frequency shift.