GPS Satellite Clock Bias Prediction Method Considering Random Items of Clocks Bias
YU Ye, ZHANG Huijun, LI Xiaohui
2018, 0(6):
1-6.
doi:10.13474/j.cnki.11-2246.2018.0166
Asbtract
(
)
HTML
PDF (1417KB)
(
)
References |
Related Articles |
Metrics
Based on the characteristics of the trend and random items of the satellite clock bias(SCB),a combination prediction model based on grey model(GM(1,1)) and Autoregressive Integrated Moving Average(ARIMA) is proposed.First,the model uses GM (1,1) to predict the trend of SCB.Then,the residual sequence of GM(1,1) model is modeled and predicted by using the ARIMA.Finally,the prediction results of GM (1,1) and ARIMA are added to obtain the final prediction value of clock bias.In addition,the predictive tests are carried out by using the precision SCB published by IGS(International GNSS Service),and the results are compared with those of the quadratic polynomial model(QPM)commonly used in SCB forecast and using the modified exponential curve method (MECM).The results show that this method can make high-precision short-term and mid-term forecast of GPS SCB.When modeling with 12 h clock bias data to predict the next 6,12,18 and 24 h,the average prediction accuracy of the model we proposed is 0.71,1.17,1.93 and 4.38 ns,respectively.Compared with the mean prediction accuracy of QPM,the accuracy of the prediction was increased by 29.70%,43.75%,67.62% and 76.21%,respectively.Compared with the mean prediction accuracy of MECM,the accuracy of the prediction was increased by 18.39%,33.90%,61.40% and 70.49%,respectively.
