High-Precision Single-Frequency GPS Point Positioning

The objective of the research described in this paper is to develop a Kalman filter-based functional model for single-frequency positioning which is suitable for low dynamic platforms such as cars or boats. No external sensors are used and the precise velocity information comes from the time-differenced carrier-phase measurements. The Kalman filter combines these measurements with undifferenced code data to solve for user position, velocity, clock errors and major biases such as ambiguities. Utilizing the velocity information with range measurements provides fast convergence of the filter. The advantage over sequential least squares processing techniques is that it relies on an accurate model of the system dynamics. Imposing a near-constant acceleration constraint, we achieved high performance in terms of position accuracy by tuning the filter process noise parameters. Initial results from processing of geodetic-quality data show decimeter level accuracy, which is close to that of real-time standalone dualfrequency point positioning. Results of processing lowquality data are of similar accuracy, but they include biases due to multipath, residual atmospheric effects and some point positioning modeling considerations, which remain to be addressed.