Wavelet variance analysis of random error properties for fiber optic gyroscope

Different random errors of fiber optic gyroscope (FOG) have different frequency characteristics. With the function of multi-resolution analysis, time-frequency domain and dimensional analysis, the wavelet variance method can be used to characterize the changes of noise in different frequency components that it was suitable for providing reference for fault diagnosis and error analysis of FOG. Compared with Allan variance which was the traditional way to analyze the random errors of FOG, the wavelet variance method could deal with a problem of power leakage occurring in Allan variance as long as the supporting interval of wavelet basis function was small enough. The experiments on the entire fiber optic gyroscope were carried out by simulating the radiation with ⁶⁰Co radiation source at different dose rate. Experimental result proves that wavelet variance can give more accurate analysis of the changes in random errors of FOG in frequency domain than Allan variance does.