Abstract: | Stanford has developed a complete real-time WAAS test facility. Using the Stanford WAAS, we have achieved a better than 3-meter 3-D accuracy 95% of time. However, with the WAAS messages, the generation of ionospheric grid corrections and confidences remains one of the most challenging tasks. The modeling goal is to 1) generate the ionospheric corrections (minimize the modeling errors) for the WAAS users, 2) estimate error confidences to cover 99.9% of the user’s ionospheric errors. At the same time, this ionospheric confidence will be used in the WAAS-weighted navigation algorithm to improve both accuracy and integrity. To investigate the generation algorithms, we have compared weighted Least-Squares(WLS) and Minimum Mean Square Error (MMSE) estimators. These algorithms employ both experimentally-determined (from NSTB data) and empirical ionospheric vertical correlation functions to account for the spatial decorrelation. Although the accuracy of ionospheric corrections reveals little difference, as the results show, the major difference between these algorithms is the generation of the grid ionospheric vertical error (GIVE) and the user ionospheric vertical error (UIVE). To account for the solar maximum cycle, a conservative function of the form exp(-x2) is recommended for the WLS estimator. The MMSE estimator uses an apriori variance of 2 meter’ for prior knowledge of the nominal model. Both approaches generate satisfactory ionospheric vertical corrections and GIVE’s. |
Published in: |
Proceedings of the 52nd Annual Meeting of The Institute of Navigation (1996) June 19 - 21, 1996 Royal Sonesta Hotel Cambridge, MA |
Pages: | 139 - 146 |
Cite this article: | Chao, Y.C., Tsai, Y.J., Evans, J., Kee, C., Walter, T., P.K.Enge,, Powell, J.D., Parkinson, B. W., "Generation of Ionospheric Correction and Confidence Estimates for WAAS," Proceedings of the 52nd Annual Meeting of The Institute of Navigation (1996), Cambridge, MA, June 1996, pp. 139-146. |
Full Paper: |
ION Members/Non-Members: 1 Download Credit
Sign In |