| A fiber-optic accelerometer based on Mach-Zehnder(M-Z) interferometer is described in this paper, also a denoising method applied in the accelerometer signal. Compared with traditional accelerometers, this accelerometer has many advantages including high precision, simple structure, high sensitivity and stability, wide bandwidth and resisting electromagnetism interference etc, so the application of the fiber-optic accelerometer is significant.Firstly, the principle and structure of the fiber-optic accelerometer based on M-Z interferometer are deeply analyzed in this dissertation. Starting with the transducing mechanism, the relation between the phase change of the optical wave and stress-strain of optic-fiber is derived. The main working procedure of the fiber-optic sensor is: the sensing element detects the acceleration signal, this signal modulates the phase of the wave, and the phase change is corresponding to the acceleration signal.Secondly, since the output signals are usually corrupted by stochastic noises, so Digital Signal Processing is applied in denoising the acceleration signal. 1/f noise has poor effects on the precision of fiber-optic accelerometers, the method of wavelet multi-resolution analysis is studied in this dissertation together with some characteristic properties of 1/f noise in wavelet domain, and then several kinds of 1/f noise models are structured. The Mallat method for processing 1/f noise in acceleration signal is put forward, that is to choose suitable thresholds to filter 1/f noise. The simulation results reveal that the method is effective.Finally, in order to improve the precision, amplifying circuit and low pass filter are designed based on theoretical analysis. PSpice is used to simulate the designed circuits, the results show that the output signal meets the requirements, which proves the feasibility of the designed circuits.
|
PDF Full Text Request