Research And Analysis Of CEEMD In Eliminating GPS Multipath

With its high precision,all-weather observation,easy operation and many other advantages,GPS are widely used in the field of deformation monitoring.In the actual observation process,due to the environment,the instrument and the factors considered,it is inevitable that the observed data will be doped with some noise and multipath errors.These errors can not be eliminated by the difference model,which can seriously affect the accuracy of the observation.This paper research and analysis based on this.(1)Based on time-frequency analysis method,we focus on empirical mode decomposition(EMD)and discuss it's problems.Then Propose the EEMD and CEEMD.The simulation results show that CEEMD has a great improvement over EMD and EEMD in the algebraic algebra and noise residue.(2)It was proved that CEEMD filter denoising is better than that of EMD,EEMD and CEEMD by the experiment.The wavelet threshold denoising based on CEEMD is introduced.Compared with the traditional wavelet threshold denoising and CEEMD filter denoising,the experimental results show that the wavelet threshold denoising based on CEEMD has a good denoising effect.A multi-method fusion denoising based on CEEMD is proposed,and the comparison test shows that the method has certain advantages.These methods are introduced into the application of weakening multi-path error.Experiments show that the above methods have certain effects,and CEEMD-based multi-method fusion is the best.(3)Multi-method fusion denoising based on CEEMD and Principal Component Analysis(PCA)are combined to correct multipath errors.PCA can extract the common error of multi-path multi-path error after denoising,then we can use the common error to correct the multipath errors.It shows that the combination of the method is more applicable and can improve the positioning accuracy through the experiment.