Investigation On The Theory And Application Of Gps Fitting Height Model

By using GPS technology, people can quickly and accurately obtain the planar positional precision of ground point which it's order of magnitude is 10-6 to 10-9, but the elevation information has not been given sufficient attention and application. and the potential of elevation precision has not yet been fully excavated because of many factors.With the further development and improvement of the number of satellite navigation and positioning system, enhancement of receiver's performance, improvement of data-processing software its applications will be more universal and in-depth, elevation precision of GPS will also be more important.GPS measurements aquire the geodetic height of the ground point which is based on the WGS-84 ellipsoid, but we always adopt normal height in our country.This relates to the conversion from geodetic height to normal high, and to gain the GPS point's normal elevation, the current general approach is using fitting model,regard as the points of the surface coordinate function in a certain scope, that the changed peacefully in this extent, we calculate the height anomaly in accordance with GPS geodetic height and their own level height which had measured by level surveying,then serve this function as a fitting coefficient to calculate the other GPS points' normal height and height anomaly.This paper focuses on solution of the value of height anomaly and the GPS height measuring principle.Useing Visual Basic (VB) as a language platform,called the MatrixvVB components,part of calculate function of Matlab computing power to achieve a fast, simple fitting system visualization elevation fitting system. Examples of relevant data using a variety of fitting methods (polynomial curve fitting, cubic spline curve fitting, flat-fitting,polynomial surface fitting) to fit through the analysis of the results of the study,given different terrain conditions,the precision of the different model fitting evaluation,and specific conclusions.The results show that GPS point zonal distribution in the longer term,the cubic spline curve fitting model is better than the effect of polynomial surface fitting; into facets region laid in the GPS-like region, the precision of polynomial surface fitting is better than planar fitting precision; in flat areas or the region of little changes in height anomaly,the appropriate choice of elevation of the GPS elevation model fitting to deal with the precision of its elevation to achieve the standard of four precision,etc.