BeiDou Reliable Integer Ambiguity Resolution In The Presence Of Time-Correlated Observation Errors

Ambiguity resolution is a prerequisite for obtaining high-precision positioning results of Beidou carrier phase difference technique.The main factors affecting ambiguity resolution include the precision of ambiguity float solution and the accuracy of random model.In order to improve the precision of ambiguity float solution,the method of multi-epoch accumulation is generally used.The method can increase the redundancy of observation equation and weaken the influence of observation noise,which enhances the model and improves the precision of ambiguity float solution.In the case of multi-epoch accumulation,the traditional random model does not take into account the time correlation between epochs,but the satellite observation contains errors with a feature of time correlation.In the case of non-zero baseline,if the error can not be completely eliminated,there will be time correlation error in the observation,which negatively affects the ambiguity resolution.Accurate random model is the premise of ambiguity resolution with high efficiency,so it is significant to analyze the time-correlated characteristics of Beidou observation and build the time-correlated random model.Based on the traditional multi-epoch accumulation random model,this paper constructs a time-correlated random model and tests it on the conditions of short baseline,medium baseline and dynamic separately.The main contents of this article include:(1)Through research on the carrier phase difference technique,a single-difference function model is constructed to collect observation residuals,which makes observations independent of each other,avoids the influence of reference satellite and facilitates the analysis of satellite-related random characteristics.Based on the model,the GPS and Beidou observation measurement accuracy are comparatively analyzed.It is found that the observation accuracy of the Beidou system and the GPS system is basically the same on the condition of zero baseline,and the observation accuracies of satellites with different types are different.(2)Based on the extraction of residual errors,the time correlation coefficient is used to evaluate the degree of temporal correlation of observation.The actual measurement results show that there is obviously time correlation under the condition of non-zero baseline,and the time correlation of Beidou system is stronger than that of GPS.The main error that affects time correlation is considered as a first-order Gaussian Markov process for simulation analysis,and the effect of time correlation on the ambiguity solution is evaluated by Monte Carlo simulation.The results show that the traditional random model ignores the correlation between epochs,resulting in a virtually high success rate of ambiguity resolution and missed detection in the ambiguity solution that passes the test,which affects the navigation accuracy and reliability of the system.(3)In order to solve the problem that the traditional multi-epoch accumulation random model does not consider the time correlation between epochs,a time-correlated random model is constructed based on that model.Compared with the traditional random model,the time-correlated random model which considers the time correlation between epochs,describes the random characteristics of observations more accurately,which can improve the success rate of ambiguity resolution and reduce the probability of false alarm and missed detection.The actual measurement results show that when the traditional random model misses some detections,the time-correlated random model can still guarantee that there is no missed situation.At the same time,the success rate of ambiguity is increased by 3.51% and the false alarm rate is reduced by 26.93%.(4)In view of the problem that ionospheric delay cannot be completely eliminated by difference,the ionospheric delay can be eliminated by using the ionospheric-free combination.At the same time,multi-epoch accumulation is used to enhance the model,and time-correlated random model is adopted.The actual measurement results show that the proposed algorithm can effectively improve the success rate of ambiguity,reduce the occurrence of missed detection and false alarms,and achieve a better balance in terms of system continuity and reliability performance.