Research On Urban High Precision Multipath Error Model

GNSS system can provide users with high-precision 3D position information at any time.By adding correction information,GNSS can potentially provide centimeter level positioning accuracy.But at the same time,it is also vulnerable to various external factors.When it is blocked,the receiving opportunity is affected by the reflected signal from the occlusion,resulting in multi-path error,which is difficult to meet the needs of high-precision positioning.In order to meet the requirements of high-precision navigation and positioning,this paper first studies the generation of multipath error and the factors that affect the multipath error,and extracts and analyzes the multipath error through pseudo range carrier phase.Then,the multipath error correction models of empirical mode decomposition and wavelet transform are established.Aiming at the shortcomings of the two kinds of filtering,the empirical mode decomposition wavelet transform method is proposed to extract the multipath,and the polynomial fitting error correction model is established to process the multipath data in real time.This model uses wavelet transform to extract noise in high-frequency signal,which avoids the loss of some effective signals of high-frequency components decomposed in EMD method,and makes use of the adaptive characteristics of EMD to enhance the filtering effect.Because there is no correlation between the multipath errors of different observation points,there is no way to correct the coordinate sequence of other observation points in a region.In this paper,a method of multi-path error distribution based on multi-path distribution function is proposed.Firstly,the multi-path error information of multiple observation points in a region is extracted by filtering method,and then the function model is constructed by polynomial fitting,and the observation data of adjacent days are processed by the function model.The experimental results show that the multipath error is effectively suppressed,and the function model established by this method can weaken the influence of multipath error in a certain range.Finally,in order to solve the problem that the conventional mathematical model cannot be established in the dynamic navigation environment,the neural network processing method is introduced,and the dynamic multi-path error modeling method based on back propagation(BP)neural network is proposed.Firstly,the observation data are trained to obtain the trained neural network model,and then the neural network model can be predicted To the next period of multipath error data,and then the next period of the original coordinate sequence correction.Through the experiment,it is found that the neural network model can well predict the dynamic multi-path error,and the predicted data is almost consistent with the actual data,which achieves the purpose of high-precision navigation and positioning.The effectiveness of the method is verified.