| Integer Ambiguity constraint is essential in precise GPS positioning. The performance and reliability of the ambiguity resolution process are being hampered by the current culmination (Y2000) of the eleven-year solar cycle. The traditional approach to mitigate the high ionospheric effect has been either to reduce the inter-station separation or to form ionosphere-free observables. Neither is satisfactory: the first restricts the operating range, and the second no longer possesses the “integerness” of the ambiguities. A third generalized approach is introduced herein, whereby the zero ionosphere weight constraint, or pseudo-observables, with an appropriate weight is added to the Kalman Filter algorithm.; The weight can be tightly fixed yielding the model equivalence of an independent L1/L2 dual-band model. At the other extreme, an infinite floated weight gives the equivalence of an ionosphere-free model, yet preserves the ambiguity integerness. A stochastically tuned, or weighted model provides a compromise between the two extremes. The reliability of ambiguity estimates relies on many factors, including an accurate functional model, a realistic stochastic model, and a subsequent, efficient integer search algorithm. These are examined closely in this research. (Abstract shortened by UMI.)... |
