GPS动态监测形变信息提取技术  

Deformation Information Extraction of GPS Dynamic Monitoring in Mining Area

作　　者：赵辉[1] 张书毕[1] 鲍国[2] 姜维为 

机构地区：[1]中国矿业大学环境与测绘学院,江苏徐州221116 [2]徐州空军学院,江苏徐州221116 [3]上海曙光爆破技术工程有限公司,上海200437

出　　处：《测绘科学技术学报》2012年第5期339-343,共5页Journal of Geomatics Science and Technology

基　　金：江苏省普通高校研究生科研创新计划项目(CXLX11-0323);江苏高校优势学科建设工程资助项目(PAPD SA1102)

摘　　要：采用GPS技术对矿区地表移动进行动态监测,针对长距离差分动态定位精度受多因素残余误差的影响,利用主成分分析法(PCA)进行空间滤波,分离区域站点相关的共模误差,以提高坐标序列精度。为进一步分析矿区地表移动受多源影响,采用解决盲信号分离问题的独立分量分析法(ICA),提取各测站形变信息。通过实例与仿真分析,结果表明,利用PCA空间滤波可提高3个分量坐标序列精度分别为46.1%,71.0%及36.8%;基于特征矩阵联合似对角化算法(JADE)进行ICA分解,能够实现各测站形变信息的提取。In order to improve accuracy in differential dynamic positioning for long distances, the Principal Com- ponent Analysis（PCA） was used to extract the Common Mode Error（CME）, the coordinates series precision was improved. For further analysis surface movement by multi-source effects, the Independent Component A nalysis（ICA）, a method to solve blind signal separation problem, was used to extract deformation information about each station. Both the simulation test and measured data analysis showed that accuracy of three coordi- nate components were improved 46.1%, 71.0% and 36.8% by PCA, and the extraction of deformation information was achieved based on ICA of joint approximate diagonalization of eigenmatrix（JADE） algorithm.

关 键 词：动态监测 形变提取 共模误差 主成分分析 独立分量分析 

分 类 号：P228[天文地球—大地测量学与测量工程]