EMD与VMD组合站坐标时间序列降噪方法  

Noise reduction method of EMD and VMD combined for station coordinate time series

作　　者：沈友东 贺小星 张云涛 SHEN Youdong;HE Xiaoxing;ZHANG Yuntao(School of Architectural Engineering,Jiujiang Vocational and Technical College,Jiujiang 332007,China;Jiangxi Research Center on Hydraulic Structures,Jiangxi Provincial Institute of Water Sciences,Nanchang 330029,China;School of Civil and Surveying&Mapping Engineering,Jiangxi University of Science and Technology,Ganzhou 341000,China;Hebei Water Conservancy Engineering Bureau Group Limited,Shijiazhuang 050011,China)

机构地区：[1]九江职业技术学院建筑工程学院,江西九江332007 [2]江西省水利科学研究院江西省水工安全工程技术研究中心,江西南昌330029 [3]江西理工大学土木与测绘工程学院,江西赣州341000 [4]河北省水利工程局集团有限公司,河北石家庄050011

出　　处：《海洋测绘》2023年第1期44-48,共5页Hydrographic Surveying and Charting

基　　金：国家自然科学基金(42104023);江西理工大学高层次人才科研启动项目(2021205200100564);江西省水工安全工程技术研究中心开放研究基金(2020GGCZX04);江西省教育厅科学技术研究项目(GJJ204015,GJJ200639)。

摘　　要：针对全球卫星导航系统(GNSS)站坐标时间序列成分复杂难分离,而传统经验模态分解(EMD)方法存在模态混叠问题,提出一种EMD与变分模态分解(VMD)组合降噪方法,可更好地提取有用信号。该方法以EMD方法确定VMD方法的分解层数,首先采用EMD降噪方法分离高频噪声与低频信号,接着采用VMD降噪方法从高频噪声中再次提取信号分量,两次降噪的信号分量之和为降噪后站坐标时间序列。以模拟数据和20个实测站坐标时间序列进行实验分析,多种传统评价指标进行定量评估,结果表明:EMD与VMD的组合降噪方法其降噪效果在一定程度上优于传统EMD降噪方法。To address the problem that the time series components of the Global Navigation Satellite System(CHMSS)station coordinates are complicated and difficult to be separated,and the traditional empirical modal decomposition(EMD)method suffers from modal confusion,a combined noise reduction method of EMD and variable modal decomposition(VMD)is proposed to better extract the useful signals.The method uses EMD method to determine the number of decomposition layers of VMD method,firstly separates high frequency noise and low frequency signal by EMD noise reduction method,then extracts signal components from high frequency noise again by VMD noise reduction method,and the sum of signal components of two noise reduction is the time series of station coordinates after noise reduction.The results show that the noise reduction effect of the combined EMD and VMD noise reduction method is better than the traditional EMD noise reduction method to a certain extent.

关 键 词：全球卫星导航系统 坐标时间序列 经验模态分解 变分模态分解 降噪分析 

分 类 号：P228[天文地球—大地测量学与测量工程]