Research On Theory And Method Of Precise Single Epoch Data Processing By GNSS

GNSS single epoch positioning has the advantage of no need to detect cycle slips.It is especially suitable for dynamic measurement,deformation monitoring and other fields where cycle slips are frequently occur.However,there are still some problems such as the low success rate of ambiguity fixing and the low accuracy of single epoch positioning.Therefore,It is great significance to research the theory and methods of precise single epoch data processing.Using dual frequency data with constraints or using triple frequency data,we improve the success rate of single epoch ambiguity and enhance the reliability of positioning.Window wavelet denoising is used to improve the accuracy of single epoch positioning by choosing the appropriate extension method.The main research contents are as following:1.Both using a search algorithm or a direct algorithm,single epoch positioning requires a high accuracy prior coordinate.Firstly,we compare the accuracy of single epoch positioning using GPS,BDS and both pseudoranges.The results show that,whether static or kinematic observation,the pseudorange single epoch positioning accuracy based on the dual system is optimal.The RMS values of position errors are 0.51 m(kinematic)and 0.47m(static),respectively.2.The conventional single epoch position algorithm based on search algorithm is analyzed.The method of damped LAMBDA to fix the ambiguity is summarized for single frequency and dual frequency observation data.Based on the TCAR algorithm,we provide the algorithm which directly calculates the ultra-wide lane ambiguity using pseudorange differential positioning results for the triple frequency data.This algorithm avoids the problem that the ultra-wide lane ambiguity can not be calculated when the pseudorange observation is lost.And the results of the pseudorange differential positioning make full use of the observation of all satellites,which is less affected by a low quality satellite observation.Experimental examples show that when the BDS triple frequency data involved in the calculation,ambiguity success rate is 100%.It is very advantageous in the ambiguity resolution of kinematic positioning when dual system positioning is used.The number of available epochs increases due to the increase in the number of available satellites.3.A kinematic single epoch positioning algorithm with additional constraints is studied aiming at the problem that the ambiguity success rate is not high when using the conventional single epoch algorithm.Constraints can be divided into two categories according to the actual situation.The first one is space linear constrain when the receiver moves along a fixed trajectory such as a train track.The second category is to establish a plane linear constraint or slope constraint when only motion in plane is considered.Experimental examples show that additional constraints can improve the ambiguity success rate to a certain extent.The ambiguity fixed success rate is 95.87%when the conventional method is used.For three-dimensional linear constraints,when the constraints intensity is 2.9 cm,the success rate reached 100%.For the two-dimensional plane constraint,when the constraint strength is 2.3 cm,the success rate reached 97.46%.For the slope constraint,when the constraint intensity is 2.3cm,the success rate reached 99.05%.4.Because the situation that the direct algorithm is difficult to be applied in kinematic positioning,a kinematic single epoch direct algorithm with additional constraints is studied.The direct algorithm needs a high accuracy priori coordinates.When the constraint reaches sufficient accuracy,the pseudorange differential positioning accuracy with constraint condition is expected to meet the requirement of directly calculating the wide lane ambiguity,and then calculate the original ambiguity.In case of using tri-frequency observations,we can directly calculate the ultra-wide lane ambiguity,wide lane ambiguity and original ambiguity in order.Because the noise of triple frequency static observation data is small,when we use the direct algorithm to solve the ambiguity,the success rate can reach 99.96%without constraints.Compared with the search algorithm,the success rate is low.But its computational efficiency has been increased by at least 30%.5.In order to improve the kinematic single epoch positioning accuracy,a precise kinematic single epoch positioning algorithm using moving window wavelet denoising is studied.The moving window wavelet is used to denoise the double difference observation data of the current epoch.The accuracy of the positioning is improved by using the clean data after denoising.The end effect of window wavelet is controlled by selecting the appropriate extension method for data extension.Experimental examples show that the use of symmetric or 0-order extension has a better improvement on the end effect,which is consistent with the results of wavelet postprocessing.Compared with the conventional single epoch position results,the accuracy is improved by about 15%in the horizontal direction and 22%in the elevation direction.