基于多通道奇异谱的GNSS坐标序列粗差探测与数据插值  

Gross error detection and data interpolation for GNSS coordinates time series based on multichannel singular spectrum

作　　者：蔡晓军 杨建华[1] CAI Xiaojun;YANG Jianhua(College of Geology Engineering and Geomatics,Chang an University,Xi'an 710054,China)

机构地区：[1]长安大学地质工程与测绘学院

出　　处：《测绘工程》2019年第5期20-28,34,共10页Engineering of Surveying and Mapping

摘　　要：由于接收机故障、天线更换以及一些未知外界环境因素的影响,导致GPS时间序列中不可避免地存在数据缺失和粗差,数据缺失和粗差会产生各种问题,因此需要鲁棒探测粗差和缺失数据插值来获得连续完整统一的时间序列。传统方法可能需要针对不同类型的数据和不同长度的数据间隙研究不同的插值方法,这使得缺失数据的插值较为困难。针对上述问题,文中采用多通道奇异谱分析(Multichannel Singular Spectrum Analysis,MSSA)对时间序列进行粗差探测和缺失数据插值,重建非均匀采样时间序列的连续可靠模型,且不需要关于时间序列的任何先验信息。在该方法中,粗差探测与数据插值同时进行。模拟数据和实测GNSS时间序列数据测试结果都表明该方法的有效性。Due to receiver failure, antenna replacement, and the influence of some unknown external environmental factors, the data loss and gross errors in the GPS time series are unavoidable.The missing data can cause various problems, thus obtaining a continuous, complete and unified time series, robust detection of gross errors and missing data interpolation are needed.Traditional methods may require different interpolation methods for different types of data and different lengths of data gaps, which makes interpolation of missing data more difficult.In view of the above problems, multichannel singular spectrum analysis (MSSA) is used in this paper to perform gross errors detection and missing data interpolation about time series, and to reconstruct a continuous reliable model of non-uniform sampling time series without any need for prior information of time series.In this method, the gross error detection is performed simultaneously with data interpolation.Both the simulated data and the measured GNSS time series data test results show the effectiveness of the method.

关 键 词：GNSS时间序列 数据插值 粗差探测 多通道奇异谱分析 

分 类 号：P228[天文地球—大地测量学与测量工程]