Researches On The Key Algorithms Of BDS Triple-frequency Precise Relative Positioning

Precise relative positioning take advantage of carrier differential technology to determine the relative position between the two or more stations.With the completion of China's Beidou satellite navigation system and the start of Beidou global network,the status of Beidou satellite navigation system grow better,so the theory of satellite navigation system needs to be researched further.Beidou satellite is satellite navigation system that provide triple-frequency signal service initially,and triple-frequency signal service has many advantages relative to dual-frequency and single-frequency.In order to meet the characteristics of the Beidou satellite navigation system triple-frequency signal,this paper mainly study the Beidou three frequency data processing.The main contents include triple-frequency observation combination theory,triple-frequency cycle detection and repair,triple-frequency ambiguity resolution and precise relative positioning.The main achievements and innovations are as follows:1.The original,single and double differential mathematical models and stochastic models of the observations are briefly introduced for the common pseudorange and carrier observations of Beidou satellite navigation systems.The error weakening methods that affect the relative positioning results are analyzed,and the parameter estimation method is introduced.The experimental analysis and comparison of different ionospheric treatment strategies and tropospheric treatment strategies are carried out.2.The three-frequency linear combination are analyzed and studied.The relationship among the combined ionospheric delay and the combined noise and the combined coefficients is analyzed for the geometry-free model.The relationship between the combined noise and the combined coefficient is analyzed for the geo-ion-free model.It is concluded that it is not feasible to find three linearly independent combinations of observations in the geometry-free pseudorange and carrier combination model to quickly solve the double difference ambiguity for the slightly longer baseline.3.The geometry-free pseudorange and carrier combination and geometry-free carrier combination cycle slip detection and repair method that weaken ionospheric delay are proposed.An optimal geo-ion-free combination observation and a geometry-free and weakening ionospheric pseudorange and carrier combination observations are selected as cycle detection and repair combinations for the Beidou three-frequency data and they can scarcely be affected by the ionospheric delay.We can see that combining noise and combining ionospheric delay of the geometry-free carrier combination that influence on the original frequency cycle are not adjustable for the geometry-free carrier combination through the analysis.Finally,twogeometry-free pseudorange and carrier combination which are not affected by the ionospheric delay and a geometry-free carrier combination based on the second difference are used to detect and repair cycle slip of Beidou triple-frequency data.The Beidou triple-frequency cycle slip detection and repair method that this paper proposes is basically not affected by the ionospheric delay and can correctly detect and repair the small cycle slip combination and insensitive cycle slip combination for large epoch interval data by the analysis of observation data,.4.The principle of TCAR ambiguity resolution and the principle of LAMBDA ambiguity resolution are deduced,and the traditional TCAR algorithm is analyzed by the observation data.The results show that the ultra wide lane and wide lane in TCAR algorithm fixed success rate is high and can be quickly solved for short and medium length baseline.But the third step of the narrow lane ambiguity solution,vulnerable to double differential ionospheric delay,becomes extremely unreliable.5.The results of LAMBDA ambiguity resolution and positioning of Beidou dual-frequency data and Beidou triple-frequency data under baseline of different lengths are compared and analyzed.It can be seen that the success rate of the triple-frequency data ambiguity resolution is not much improved in the case of the short baseline relative to dual-frequency data by calculating the data under baseline of different lengths,and the position convergence rate is similar to the dual-frequency data in the case of the medium length baseline.In addition,we can see that the accuracy of the triple-frequency data and the dual-frequency data is roughly equal by analysing accuracy of the above-mentioned baseline of the different lengths.6.Beidou triple-frequency mathematical model with constraint conditions is proposed.The geometric accuracy factor and the ambiguity accuracy factor of the model are superior to the traditional triple-frequency positioning model's by analysing measurement data.The amount of new model's independent parameter is greatly reduced due to the introduction of the constraint condition compared with the traditional triple-frequency positioning model,which is helpful to improve the fixed success rate of the ambiguity and the convergence speed of the position parameter.The new model is validated by two baseline of short lengths and two baseline of medium lengths.Through the results of position and ambiguity parameters,we can see that ambiguity fixed success rate of the Beidou triple-frequency mathematical positioning model with the constraint condition is higher than traditional model in the case of single-epoch resolution for baseline of short lengths and its ambiguity fixed success rate are higher than 99%.At the same time,the convergence velocity of the ambiguity and the convergence velocity of the position parameters is faster than traditional triple-frequency positioning model's for baseline of medium lengths and the Beidou triple-frequency positioning model with the constraints uses only 72 epochs and 5 epochs for baselines of 45 km and 72 km lengths,while the traditional model uses342 epochs and 90 epochs,respectively.