Abstract：A new prediction method based on the combination of gray system model and wavelet neural network is proposed.Firstly，fitting given satellite clock bias values with gray model，and determining parameters of gray model with the values.Then，establishing model of wavelet neural network with fitting residuals.Finally，combining the prediction results obtained from the two models mentioned above to get prediction results.The results of prediction tests using the precise satellite clock bias data from IGS show that the method can get better satellite clock bias prediction results，compared to the quadratic polynomial model and the grey model.Especially for the rubidium clock of ⅡR type and ⅡF，the prediction precision can be provided with two times or more.
Wang Yupu,Lü Zhiping,Chen Zhengsheng et al. A METHOD OF SATELLITE CLOCK BIAS PREDICTION BASED ON
GRAY MODEL AND WAVELET NEURAL NETWORK[J]. jgg, 2014, 34(3): 155-159.
