Research On Ambiguity Resolution Of BDS/GNSS Multifrequency Carrier Phase Based On Quadrangle Operation

In GNSS carrier phase measurement,the solution of integer ambiguity is the key step of positioning.Fast and accurate solution of integer ambiguity is of great significance to improve positioning accuracy and reduce positioning time.Therefore,starting from the positioning double difference observation equation,this paper studies the principle and process of double difference solution of integer ambiguity,deeply analyzes the least squares ambiguity adjustment(lambda)algorithm,which is most widely used in engineering,and puts forward two improved integer ambiguity solution methods.Finally,it is analyzed and verified by simulation experiments and measured experiments.The main research contents of this paper are as follows:(1)This paper introduces the GNSS carrier phase observation equation,constructs the theoretical basis of difference and linearization,focuses on the principle of integer ambiguity resolution,deeply studies the algorithm theory of integer ambiguity decorrelation process and search process,and analyzes the advantages and disadvantages of its algorithm solution process.(2)Aiming at the problem of decorrelation delay caused by a large number of matrix operations in the current lambda algorithm,an improved decorrelation algorithm based on matrix sorting method is proposed.The square difference covariance matrix is presorted to reduce the probability that the diagonal elements in the diagonal matrix after Cholesky decomposition are less than zero,reduce the gap between the diagonal elements after conversion,and improve the efficiency of decorrelation.The experimental results show that MOM method can reduce the average solution time from 43690μs to 40970μs while maintaining the same accuracy as LAMBDA algorithm。(3)In the LDL decomposition process of lambda algorithm,the formation process of L,D and Lt submatrices depends too much on the calculation formula,and in the actual programming and calculation,more storage matrices are required for the storage of L,D and Lt matrix elements,resulting in the complexity and time-consuming of the calculation process.The concept of composite matrix is introduced,and an improved Cholesky decomposition is proposed to simplify the relationship between L,D and Lt submatrices;The four corner rule is used to simplify the calculation process of each sub matrix element and improve the efficiency of decomposition calculation.The experimental results show that this method reduces the calculation time by about 15% on the premise of ensuring the whole cycle ambiguity resolution accuracy.