Positioning performance improvements with European multiple-frequency satellite navigation system-Galileo

Abstract

The rapid development of the Global Positioning System has demonstrated the advantages of satellite based navigation systems. In near future, there will be a number of Global Navigation Satellite System (GNSS) available, i.e. modernized GPS, Galileo, restored GLONASS, BeiDou and many other regional GNSS augmentation systems. Undoubtedly, the new GNSS systems will significantly improve navigation performance over current GPS, with a better satellite coverage and multiple satellite signal bands. In this dissertation, the positioning performance improvement of new GNSS has been investigated based on both theoretical analysis and numerical study. First of all, the navigation performance of new GNSS systems has been analyzed, particularly for urban applications. The study has demonstrated that Receiver Autonomous Integrity Monitoring (RAIM) performance can be significantly improved with multiple satellite constellations, although the position accuracy improvement is limited. Based on a three-dimensional urban building model in Hong Kong streets, it is found that positioning availability is still very low in high-rising urban areas, even with three GNSS systems. On the other hand, the discontinuity of navigation solutions is significantly reduced with the combined constellations. Therefore, it is possible to use cheap DR systems to bridge the gaps of GNSS positioning, with high accuracy. Secondly, the ambiguity resolution performance has been investigated with Galileo multiple frequency band signals. The ambiguity resolution performance of three different algorithms is compared, including CAR, ILS and improved CAR methods (a new method proposed in this study). For short baselines, with four frequency Galileo data, it is highly possible to achieve reliable single epoch ambiguity resolution, when the carrier phase noise level is reasonably low (i.e. less than 6mm). For long baselines (up to 800 km), the integer ambiguity can be determined within 1 min on average. Ambiguity validation is crucial for any ambiguity resolution algorithm using searching method. This study has proposed to use both Ellipsoidal Integer Aperture (EIA) estimator and R-ratio test for ambiguity validation. Using real GPS data and simulated Galileo data, it has been demonstrated that the new method performs better than the use of EIA or the R-ratio test alone, with much less ambiguity mis-fixed rate.