Study On The Algorithm Of GNSS Single-Frequency Precise Point Positioning

This paper studied the algorithm of GNSS Single-Frequency precise point positioning,used the measured data to verify the feasibility of the algorithm,and achieved certain results.The innovation and main work of this paper are included:a)The paper proposed a new method on detection and repair for cycle slips using epoch difference observations.Considering the misjudgment of cycle slip detection by Single-frequency preprocessing method based on double differences,a new method on detection and repair for cycle slips,which can effectively decrease misjudgment of cycle slip detection when there are cycle slips in reference satellite.The validity of the method is proved by the verification of the measured data.b)The method of using robust estimation and fuzzy weighted average detect cycle slip is proposed.The existing single-frequency cycle-slip detection methods usually verify the feasibility of the algorithm through the high sampling rate observation data,but there is no reasonable solution for the low sampling rate observation data.Especially for Beidou system,the post-normalized residuals of the epoch-difference observations can be up to meter level due to the instability of its satellite clock.If based on the conventional cycle-slip detection method,there will be a large amount of normal observations treated as cycle slips,decreasing the intensity of the coordinate solution.Therefore,based on the research of the characteristics of the posterior residual under different sampling rate for GPS and BDS epoch difference observations,a method for reasonable determination of the threshold of the cycle slip detection is proposed.The experimental results show that the method of using robust estimation and fuzzy weighted average detect cycle slip can effectively improve the positioning accuracy:for GPS,the accuracy can be increased by 24.5%,15.4%and 12.0%in the directions of NEU respectively;for BDS,the accuracy can be increased by 10.1%,37.6%,34.2%respectively.c)This paper realizes a set of single-frequency PPP algorithm and verifies the advantages of multi-system single-frequency PPP algorithm compared with single-system single-frequency PPP algorithm by setting different elevation angles to simulate urban canyons.Considering that square root inf'ormation filter has higher numerical precision and stronger stability,it is especially suitable for real-time data processing.Based on this method,a real-time data processing estimator for single-frequency PPP is constructed.This paper uses precise satellite orbit and satellite clock products,corrects GNSS errors of various sources,and treats slant ionosphere delay and zenith troposphere delay as unknown parameters which can be estimated by random walk method.Meanwhile,the ISB error is also considered in multi-system data processing.The experimental results show that,the positioning performance of BDS single frequency PPP is worse than that of GPS single frequency PPP,and the multi-system single-frequency PPP algorithm can effectively improve the positioning performance of single-system single-frequency PPP and the reliability of the results.For the application of relatively inexpensive and lightweight GNSS single-frequency receiver equipment in urban location services,precision agriculture,intelligent transportation system and other emerging areas,there will be a certain Promote the role.