Research On Precise Orbit Determination And Integrity Monitoring Of Beidou Satellite Navigation System

The precise orbit of Beidou Satellite Navigation System(BDS)is the basis for high-precision positioning of the single receiver users.The acquisition of precise orbits mainly depends on precise orbit determination technology.How to improve the precise orbit accuracy and ensure the orbit reliability are key issue for the application of BDS precise point positioning related to safety of life.This paper focuses on precise orbit determination and integrity monitoring technology based on the BDS hybrid constellation and the mature experience of Global Positioning System(GPS).The research is carried out from four aspects: initial value optimization of precise orbit determination,anomaly detection of signal in space,model optimization of precise orbit determination and integrity monitoring of real-time precise orbit.It aims to improve the accuracy and reliability of the precise orbit,and establish a technical framework suitable for BDS precise orbit determination.The main work and innovation of the paper are summarized as follows:(1)Aiming at the problem that the non-normal distribution of signal in space error reduces the parameter estimation accuracy,the trend-term model of orbit and clock error is established to improve the normality of initial value error of precise orbit determination.This method realizes the calculation of signal in space error based on the optimization of satellite antenna phase offset and clock reference bias,and the trend terms of orbit and clock errors are extracted based on the signal in space error mean and spectrum to realize the normality optimization of initial value error.The experimental results show that the compensation of error trend term can effectively improve the normality of initial value error compared with the original signal in space error.The optimized initial value error can be approximated as a zero-mean normal distribution,which provides a good stochastic model for precise orbit determination and significantly improve the accuracy of orbit estimation.(2)The non-normal distribution of BDS instantaneous user rang error is inconsistent with the assumption of traditional signal in space anomaly detection method.In order to solve this problem,a post signal in space anomaly detection method is proposed based on orbit and clock error decomposition.To realize the signal in space anomaly detection,the polynomial decomposition of orbit and clock errors is used to construct the test statistics which can be described by zero-mean normal distribution.Aiming at the low efficiency in traditional real-time signal in space anomaly detection caused by large pseudorange noise,a carrier phase smoothing pseudorange method is proposed based on Kalman filter.This method uses Kalman filtering to achieve adaptive selection of smoothing time,and uses the instantaneous user range error model to optimize the calculation of the receiver clock,thereby improving signal in space anomaly detection efficiency.Based on the data in past six years,the experimental results show that the BDS signal in space anomaly probability is much greater than that of the GPS,and signal in space anomalies must be eliminated in the process of precise orbit determination.Compared with the signal in space anomaly detection method with a fixed smoothing time,the proposed method can significantly improve the efficiency of signal in space anomaly detection and fully guarantee the strength of the precise orbit determination observation model.(3)In order to improve the orbit accuracy of Geostationary Earth orbit(GEO)and Inclined Geosynchronous Satellite Orbit(IGSO)satellites affected by low-precision dynamic model,a precise orbit prediction method constrained by the prior orbit error model of the broadcast ephemeris is proposed.A priori orbit error model is constructed based on the orbit error spectrum of broadcast ephemeris,and precise orbit prediction is realized by extrapolation of real-time broadcast ephemeris and prior orbit error models.Comparing with the prediction method using the dynamic model,the proposed method can make up for the shortcomings of the dynamic model,and significantly improves the orbit prediction accuracy of the GEO and IGSO satellites.(4)Aiming at the problem that the reliability of real-time precise orbit is difficult to guarantee due to the anomalous orbit parameters,the adjacent prior ephemeris is used to predict satellite position,and a test statistics is constructed based on the deviation between the real-time and predicted satellite position to realize the integrity monitoring of real-time precise orbit.Compared with the yesterday's ephemeris minus today's ephemeris method,the proposed method can effectively reduce the predicted value of position deviation and reduce the minimum detectable error corresponding to the designed test statistics.The detection performance can not only meet the requirements of false alarm and missed detection rate,but also improve the sensitivity of orbit anomaly detection.The reliability of real-time precision orbit is fully guaranteed.