| Abstract: | The presence of selective availability (SA) on GPS has led to a variety of proposals for mitigation of its effects, for example combining GPS with other sensors or by identifying and removing the SA using redundant information in the position solution. A further method for reducing the impact of SA is the combination of GPS and GLONASS measurements. GLONASS does not have any deliberate degradation. This paper describes several algorithms for combining GPS and GLONASS measurements in real time on a single receiver. There are two least squares solutions using the code phase measurements alone: one with the measurements from both systems weighted equally, and one with the measurements weighted by the covariance of the measurements. Kalman filters with no SA model, and first- and second-order models, using both the code and carrier phase to estimate user position and velocity, are also implemented. Two different adaptive schemes are compared which attempt to identify SA model parameters in real time. The algorithms are compared in terms of their positioning accuracy, computational overhead and robustness. A ten-channel GPS/GLONASS receiver developed at The University of Leeds is used to provide satellite data for the evaluation of the different methods. |
