Research On Signal Processing Method In Solar Spectral Velocimetry Navigation

Solar spectral velocity measurement navigation is a new method of autonomous astronomical navigation,which uses the spectrometer onboard spacecraft to obtain real-time spectral information and combines with the optical Doppler redshift effect to directly obtain real-time radial velocity.This method makes full use of a wide range of light sources in the deep space environment as the navigation source,which can solve the limitations of traditional navigation that can not get rid of the support of ground base stations.It can make up for the shortcomings of traditional astronomical angle measurement navigation methods that can not directly obtain speed information and poor accuracy of instantaneous speed determination,and there is no need to model based on orbital dynamics,which reduces the complexity of autonomous navigation.Solar spectral velocimetry navigation uses sunlight as the navigation source,and the relative radial velocity between the spacecraft and the sun is derived from the Doppler redshift between the received spectrum and the standard solar spectrum.The sun is the only star in the solar system,with strong illumination energy and extremely rich spectral lines,but the solar spectrum is extremely unstable.Solar activities such as sunspots,corona and flares will cause spectral distortion,which will affect the accuracy of solar spectral velocity measurement and reduce the accuracy of navigation.Therefore,improving the measurement accuracy of solar spectral velocity is the premise of high precision astronomical velocity measurement and navigation.Based on the principle of solar spectral velocity measurement navigation,this paper explores the signal processing method for improving the two important links of solar spectral velocity measurement navigation(redshift measurement and radial velocity calculation).The specific work is as follows:Based on the principle of optical Doppler redshift,a method to accurately measure the solar spectral redshift is studied.The solar spectrum has an extremely broad continuous spectrum,containing tens of thousands of emission and absorption lines,and the radiated energy is mainly concentrated in the visible and infrared regions with the wavelength of the spectral line above 400 nm.The analysis of spectral data shows that a large number of infrared absorption peaks can be used for redshift measurement.However,due to the interference of skylight background,the distortion of actual received spectrum will affect the accuracy of redshift measurement.In order to improve the performance of redshift measurement,a solar spectral redshift measurement method based on dual tree complex wavelet analysis and kernel density estimation is proposed.Firstly,the dual tree complex wavelet transform is used to denoise the solar spectrum,then the feature absorption spectrum is normalized and extracted,and Parzen window density estimation method is used to calculate the red shift value.By analyzing the influence of sunspots activity and simulating different noise on the redshift measurement,the results show that the accuracy of the redshift measurement method is more than 96% within 0.001 allowable error.Aiming at the solar spectral velocity measurement navigation,the method to improve the accuracy of Doppler redshift velocity measurement is studied.At present,most of the research on velocity measurement and navigation is to improve the navigation accuracy from the perspective of navigation modeling,such as double satellite differential velocity measurement model.Based on the principle of solar spectral Doppler redshift velocity measurement,an adaptive EMD-NDFT doppler redshift velocity measurement method for solar spectral velocity measurement navigation is proposed in this paper.Firstly,according to the non-stationary characteristics and instabilities of solar spectral signals,the EMD algorithm is used to perform adaptive stratification,and then adaptive reconstruction signals were obtained by setting adaptive thresholds according to the eigen modal signals of each layer.Then,according to the characteristics of non-uniform sampling of solar spectrum,the spectral line is moved to the frequency domain by non-uniform Fourier transform,and the radial velocity of spacecraft relative to the sun is matched by Taylor formula.This method combines time domain noise reduction with frequency domain calculation.The simulation results show that the speed measurement accuracy is effectively improved and the computational complexity is reduced.