GPS高程异常拟合粗差剔除以及精度评价方法研究  

Research on removed GPS elevation based on abnormal fitting gross errors and accuracy evaluation methods

作　　者：李浩[1] 吴祖海[1] 

机构地区：[1]中南大学土木工程学院,湖南长沙410075

出　　处：《铁道科学与工程学报》2015年第1期99-104,共6页Journal of Railway Science and Engineering

基　　金：国家自然科学基金资助项目(41171326)

摘　　要：提出一种GPS高程拟合公共点粗差定位的新方法,即利用最小二乘法平差所得方差,采用假设检验法,完成GPS高程异常拟合模型参数计算过程中的粗差定位与剔除。对多种高程拟合模型的拟合精度进行评价,引入均方误差、信噪比和互相关系数评价指标,提出一种多指标融合评价方法,实现对GPS高程异常拟合精度的评价。结合实例分析,证明所提方法的有效性。A novel method was presented in this paper where variance after least square adjustment was used to locate the gross errors in the calculation of GPS elevation fitting model parameters using hypothesis testing. The accuracy of different elevation fitting models was evaluated by introducing evaluation index of mean square error, signal- to- noise ratio, mutual correlation coefficient, and a hybrid evaluation method was proposed to assess the accuracy of GPS elevation anomaly fitting. Combined with some practical projects, the results calculated in this literature prove that these methods are valid.

关 键 词：GPS高程异常 拟合模型 粗差定位 精度评价 

分 类 号：P228[天文地球—大地测量学与测量工程]