Research On Integer Phase Resolution Algorithm Base On Differenced Carrier Phase

With the vigorous development of the satellite navigation and positioning industry,various related industries have higher requirements for satellite positioning accuracy.High-precision positioning technology based on carrier phase difference has gradually become a hot issue.The most important step in carrier phase differential positioning is to obtaion high-precision carrier phase,of which the core issues are cycle slip detection and correction and integer ambiguity resolution.In the thesis,the basic theory and classical algorithms of cycle slip detection and integer ambiguity resolution are studied and analyzed.The shortcomings of traditional algorithms are modified and the performance of the modified algorithm is verified by simulation experiments.The main contents of the thesis are as follows:Firstly,the mathematical model of carrier phase observation is derived and analyzed,and the concepts of cycle slip and integer ambiguity are introduced.The linear combination of GNSS observations is studied,the combination formula of each method is given,and their error suppression capability is analyzed.Then the error sources in GNSS observation are studied and classified.The causes of various error sources and their correction methods are discussed and analyzed,which lays a theoretical foundation for the cycle slip detection and the integer ambiguity resolution research in the following.Then,the problem of cycle slip detection based on dual-frequency is studied in the thesis.Firstly,the generation principle of the cycle slip detection value is analyzed.Then,the theory of classical dual-frequency cycle slip detection and correction algorithms is studied and deduced,and the advantages and disadvantages of these algorithms are analyzed.Considering the shortcomings of M-W combination,a modified M-W combination is proposed by introducing a dual-frequency geometry-free combination as the supplementary detection value,and the pseudo-range observation value is modified by polynomial fitting.The results of simulation experiments show that the modified algorithm avoids the detection blind area of traditional algorithm and is more practical.Finally,the problem of integer ambiguity resolution in high dimension is studied in the thesis.The least square model of double-difference ambiguity is deduced,several classical ambiguity resolution methods based on ambiguity domain are analyzed,and the limitation of these algorithms in high dimensions is analyzed.The standard differential evolution algorithm is modified in two aspects of search space and strategy according to the characteristic of high dimension integer ambiguity searching,and an adaptive differential evolution algorithm?(ADE)is proposed.In the aspect of search space,the decorrelation algorithm is adaptively selected according to the different dimensions of ambiguity,and the search space is fixed by the covariance matrix of the float solution,so as to improve the calculation efficiency.In the aspect of search strategy,the variation factor of the algorithm decreases adaptively according to the Logistic model law,thus improving the convergence speed of the algorithm.Random disturbance of crossover factors and partial population elimination strategy effectively prevent the algorithm from falling into a local optimal solution.The results of simulation experiments show that the convergence speed of the algorithm is faster than some commonly used intelligent optimization search algorithms,does not fall into local optimal solutions,and has higher solution efficiency than LAMBDA method in high-dimensional conditions.