EWT与MPE结合的GNSS坐标时间序列信号提取方法  

Signal Extraction Method of GNSS Coordinate Time Series Based on EWT-MPE

作　　者：陶蕊 鲁铁定[2] 周子琪 TAO Rui;LU Tieding;ZHOU Ziqi(Communications Design&Research Institute of Jiangxi Province,Nanchang 330052,China;School of Surveying and Geoinformation Engineering East China University of Technology,Nanchang 330013,China;Jiangxi Port&Navigation Design Institute Co.Ltd,Nanchang 330013,Chian)

机构地区：[1]江西省交通设计研究院有限责任公司,江西南昌330052 [2]东华理工大学测绘与空间信息工程学院,江西南昌330013 [3]江西省港航设计院有限公司,江西南昌330013

出　　处：《测绘科学技术学报》2024年第5期446-452,共7页Journal of Geomatics Science and Technology

基　　金：国家自然科学基金项目(42374040,42061077,42104023,42364002)。

摘　　要：针对GNSS坐标时间序列信号提取过程中部分有效信号在残差噪声中“湮没”的问题,提出一种经验小波变换(EWT)和多尺度排列熵(MPE)相结合的二次信号提取方法(EWT-MPE)。该方法对第一次滤波后残差噪声进行分解,产生经验尺度分量和一系列经验小波分量,并采用MPE作为阈值指标,将各分量划分为信息模态分量、混合模态分量和噪声模态分量,对混合模态分量再次提取,直至无混合分量后,再重构初始去噪信号和信息模态分量为最终提取信号,从而达到信号二次提取的目的。通过对中国境内8个陆态网台站原始数据进行实验研究,结果表明,该方法整体优于经验模态分解(EMD)和EWT,信噪比均值达到12.47 dB,能量百分比均值达到79.10%,与EMD和EWT相比,信噪比分别提升33.80%和13.78%,能量百分比分别提升16.15%和5.87%,可以有效提取出原始信号中局部趋势特征,进一步提升信号提取效果。Aiming at the problem of“annihilation”of some effective signals in residual noise in the process of GNSS coordinate time series signal extraction,a secondary signal extraction method(EWT-MPE)combining empirical wavelet transform(EWT)and multi-scale permutation entropy(MPE)is proposed.This method first decomposes the residual noise after the first filtering and generates empirical scale components and a series of empirical wavelet components,and then uses MPE as a threshold index to divide all component into information modal components,mixed modal components and noise modal components.The mixed modal components will then be extracted again until there is no mixed component,and then the initial denoising signal and the information modal component are reconstructed into the final extracted signal,so as to achieve the purpose of the second time signal extraction.Through experimental research on the original data of 8 terrestrial network stations in China,the results show that the method is overall better than empirical mode decomposition(EMD)and EWT,with an average signal-to-noise ratio of 12.47 dB and an average energy percentage of 79.10%.Compared with EMD and EWT,the signal-to-noise ratio is increased by 33.80%and 13.78%,and the energy percentage is increased by 16.15%and 5.87%respectively.It can effectively extract local trend features in the original signal and further improve the signal extraction effect.

关 键 词：全球导航卫星系统 坐标时间序列 经验小波变换 多尺度排列熵 信号提取 

分 类 号：P228[天文地球—大地测量学与测量工程]