Research on Prediction and Characteristics of Periodic in BDS/GPS Satellite Clocks

摘要

The prediction accuracy of satellite clock bias (SCB) directly affects the performance of navigation system. According to the spectrum analysis method, the periodic characteristics of satellite clocks are analyzed, and using precise SCB data of BDS/GPS satellite the prediction tests are conducted, and then we build the forecast model considering periodic items. Based on this, the influence of the number of periodic items on the result of SCB prediction is analyzed. The effect of SCB prediction between periodic term model and quadratic polynomial model is compared, and the SCB prediction performance of different types of satellites is summarized. Experimental results show that the periodic variation of satellites are significant. The periodic items of BDS are 24, 12 and 8 h, and the periodic items of GPS are 24, 12 and 6 h. The number of periodic items affects the prediction results. The prediction accuracy of BDS using 2-3 main periodic items is the best, and the prediction accuracy of GPS using 3-4 main periodic items is the best. Compared with the quadratic polynomial prediction model, the prediction accuracy of the proposed model with additional periodic items can be improved by 1-6 ns.