Abstract
Generally, neglecting higher-order ionospheric delay can cause a range bias of several centimetres depending on satellite elevation, azimuth, and the ionospheric and geomagnetic conditions. Recent research has shown that such residual errors are correlated, which contributes to the slow convergence of precise point positioning (PPP) solution. In this paper, we attempt to rigorously model all GPS errors and biases, including the second-order ionospheric delay. Raw GPS measurements from a global cluster of international GNSS services (IGS) stations are first corrected for the effect of second-order ionospheric delay. The corrected data sets are then used as input to the Bernese GPS software to estimate the precise orbit and satellite clock corrections. Such precise products have been used in all of our PPP trials. NRCan's GPSPace software is modified to accept the second-order ionospheric corrections as well as the new NOAA tropospheric correction product. To examine the effect of rigorous error modeling on the PPP solution, new data sets from several IGS stations were processed using the modified GPSPace software. It is shown that the effect of second-order ionospheric delay can significantly affect the PPP solution convergence and its accuracy estimation.