GPS Multipath Correction Model Based on Adaptive Local Mean Decomposition

Aiming at the current situation of sifting pure frequency-modulated (FM) signal of ensemble local mean decomposition (ELMD) using hard threshold criterion, this paper proposes an improved ELMD method based on the adaptive sifting stopping criterion. The improved ELMD combined with the consecutive mean square error (CMSE) criterion, which is named as the adaptive local mean decomposition (ALMD), is used to separate the noise signal and extract the useful signal and applied to the research of GPS multipath effects. The results of numerical simulation show that compared with the traditional LMD using the hard threshold criterion and wavelet de-noising method, the de-noising method proposed in this paper is better and more stable. The usage of the proposed method, which aims to extract the multipath correction model from the coordinate residual time series, can weaken the influence of multipath effects to a certain extent, thus improve GPS positioning accuracy.