Item

Abstract

Since the orbit repeat cycle of Galileo satellites is about 10 days, existing spatial–temporal repeatability-based station multipath mitigation methods such as the modified sidereal filtering (MSF) and the multipath hemispherical map need more than one week of data in order to model the Galileo phase multipath correction. From this perspective, these methods are improper for most applications. We proposed a random walk multipath method (RWM) to resolve this problem. Since the elevation and azimuth angles between adjacent epochs do not change too much, the low-frequency phase multipath effects at adjacent epochs are similar. Multipath correction value can be estimated by the random walk model. Because the GPS satellite repeat cycle is short, GPS observations easily mitigate the multipath effect and have common coordinate parameters with Galileo. Multipath-reduced GPS signals can be a constant for separating coordinate parameters and multipath parameters. Galileo phase observation residuals of the latest day are used to calculate the variance for the random walk model. Experiment results show that compared to the traditional MSF model, the proposed RWM method can improve the Galileo residual reduction and positioning precision nearly without the need of any historical data. As for the practical real-time GNSS monitoring application in a severe multipath environment, the result shows that the new method can significantly reduce the Galileo multipath effect and subsequently yield a precise positioning solution. Moreover, the RWM method is invulnerable to the sampling rate and observation environment.