Research of Filtering based on Wavelet De-noising Method for Revolution-Modulation North-Finder

摘要

In order to improve the precision of revolution-modulation north-finder, a filtering method based on wavelet de-noising method is studied. The continuous revolution in constant rate is used to modulate the output signals of gyroscopes and accelerometers, and the signals are demodulated by using the integrated method. The filtering method based on wavelet de-noising method is used to filter the signals of gyroscopes. The disturbance is filtered by maximum modulus, whose Lipschitz exponent is bigger than zero. The mean value of non-maximum modulus adjacent to the selected maximum modulus is used to alternate the maximum value. The noise is filtered by soft-thresholding, whose Lipschitz exponent is smaller than zero. The experimental results show the filtering method based on wavelet de-noising method effectively restrains the effect of random drift of gyroscopes and disturbance of system, and improves the precision of the revolution-modulation north-finder.