Research On Ambiguity Resolution Methods Based On BDS Triple-frequency

China's Beidou satellite navigation system is progressing steadily from the domestic region,to the Asia-Pacific region,and then to the global region covering the “three-step” plan.As the world's first full-constellation broadcasting triple-frequency signal system,Beidou's triple-frequency observations can form more optimal combination observations with longer wavelengths,weak ionospheric delay,and small combined observation noise,which contributes to ambiguity resolution and improving the accuracy of Beidou navigation positioning.Therefore,this thesis mainly studies the theory of three-frequency combined observations,discusses and analyzes several methods of solving the Beidou triple-frequency ambiguity resolution.The main contents and research results are as follows:(1)The working principle,basic observation,pseudorange and carrier phase basic observation equations of BDS are introduced.And the Beidou single,double and triple difference observation equations with their applications are given.Then,the various error of Beidou observations are discussed and general methods for weakening and correcting related errors are summarized.(2)The triple-frequency combined observation theory is studied,and the correlation definition of the triple-frequency combined observations is introduced.The main error items of the combined observations are analyzed by integer linear transformation: ionospheric delay,combined observation noise,tropospheric delay;The selection of the optimal linear combination of the Beidou triple-frequency is discussed.The optimal three sets of linear independent combinations in the short baseline and long baseline scenarios are given.And then the commonly used linear combination methods and their applications are introduced.(3)Two methods of triple-frequency ambiguity resolution without geometric model are studied: CIR method and matrix transformation method.It is verified by the measured data that both methods can obtain good fixation effect under short baseline conditions: the single epoch is fixed;in the case of long baseline,the fixed effects of the two methods are affected by ionospheric delay and combined noise,and the success rate is not very high.The overall effect of the CIR method is slightly better than the matrix transformation method.(4)The method of step-by-step fixing of three-frequency ambiguity based on geometric model is studied.It can be seen from the verification of measured data that due to the influence of geometric noise,even for short baseline,narrow lane NL ambiguity is not easy to be directly fixed,and then adopted The correct narrow lane ambiguity can be obtained by smoothing multiple epochs before and after.(5)The non-geometric and ionospheric delay CIR correction method is studied.It can be seen from the verification of measured data that the ionospheric error is eliminated and the reliability of the ambiguity resolution is guaranteed,but the fluctuation of the narrow-channel ambiguity floating-point solution is become bigger.Two main reasons for noise amplification are analyzed: one is due to the amplification of noise during the ionospheric delay correction;the second is that the ionospheric correction method is not accurate.(6)For the influence of the delay of the ionosphere,the classical CIR is difficult to achieve long baseline ambiguity resolution,and an improved CIR algorithm for long baselines with constraints is proposed.The performance of the algorithm is performed by using BDS triple-frequency data.Experiments show that the proposed new algorithm can improve the fixed success rate of long baseline wide lane ambiguity,and can greatly reduce the search space of narrow lane ambiguity,so that the valuation fluctuation can be within 4 weeks.