ABSTRACT
Since the 1970s, based on the development of the optical fiber communication technology, many of the world’s universities and research institutions attach great importance to the research of fiber-optic gyroscope (FOG). FOG has a series of advantages, such as no moving parts, low cost, long service life, light quality, small size, large dynamic range, resistance to electromagnetic interference, and no acceleration caused by drift. So it has become one of the mainstream instruments in the field of inertial measurement and guidance in the 21st century. But the FOG also has its technical difficulties, its accuracy is also short of the existing level of laser gyro, so we have to eliminate or reduce the output error of FOG as much as possible (Zhang Guicai, 2008). Studies have shown that the coil of FOG has the nonreciprocal phase shift (i.e., Shupe error) due to the temperature disturbance and the Sagnac phase shift caused by rotation, and they cannot be distinguished. And this causes the output of the FOG to have a large bias error. This is the toughest temperature drift of FOG. Wavelet transform is a time frequency localization analysis, through the telescopic translation operations to multi-scale refinement of signal step by step, finally to achieve high frequency time segment, the low frequency in the frequency segment, can automatically adapt to the requirement of time-frequency signal analysis. Wavelet analysis has been widely used in signal processing (compression, denoising, frequency detection), digital watermarking, fault diagnosis, and so on (Han Junliang, 2010).
