基于CEEMD的GPS多路径改正模型研究

摘要

利用总体完备经验模态分解从X、Y、Z三坐标分量残差中分离并提取多路径信息,形成多路径改正模型,对第二天坐标序列进行改正,并与经验模态分解及移动平均去噪法进行对比.结果表明,利用总体完备经验模态分解可以有效地从坐标残差中提取多路径误差,且经改正后的坐标序列RMS明显小于改正前,说明在一定程度上削弱了多路径误差影响,且处理效果优于移动平均去噪法及经验模态分解法.%This paper separates and distills the multipath model from three directions residual coordinate of X,Y,Z based on complete empirical mode decomposition (CEEMD).We use the repetition characteristics of multipath to correct the coordinate sequence for the next day.The results show that,CEEMD can effectively extract the multipath error correction model from residual coordinate,and the RMS of coordinates sequence after correction is much smaller than before.This signifies that the method weakens the influence of multipath error to a certain extent.Furthermore,the processing result is superior to the traditional moving average de-noising method and EMD.