Title: Evaluation of Orbit and Clock Models for Real Time WAAS
Author(s): Yeou-jyh Tsai, Yi-chung Chao, Todd Walter, Changdon Kee, David Powell, Per Enge, and Brad Parkinson
Published in: Proceedings of the 1995 National Technical Meeting of The Institute of Navigation
January 18 - 20, 1995
Disneyland Hotel
Anaheim, CA
Pages: 539 - 547
Cite this article: Tsai, Yeou-jyh, Chao, Yi-chung, Walter, Todd, Kee, Changdon, Powell, David, Enge, Per, Parkinson, Brad, "Evaluation of Orbit and Clock Models for Real Time WAAS," Proceedings of the 1995 National Technical Meeting of The Institute of Navigation, Anaheim, CA, January 1995, pp. 539-547.
Full Paper: ION Members/Non-Members: 1 Download Credit
Sign In
Abstract: The Wide-Area Augmentation System (WAAS) will be a primary navigation aid for all phases of flight down to Category I precision approach. Previous results of flight trials done by the Stanford WADGPS labora- tory in the summer of 1994 show that vertical accura- cies better than 3.1 meters can be achieved 95 percent of the time. In our early trials, vector corrections for each GPS satellite were based on the snapshot process- ing of pseudorange residuals sent from our Wide-Area Reference Stations (WRSs). In other words, the cor- rections did not take advantage of any process models for either the ephemeris or the.satellite clock errors. In this paper, we use a simple kinematic model for the GPS satellite ephemeris error and a second order Gauss-Markov process for the satellite clock error. A Kalman filter approach is employed to smooth the es- timate of error corrections. In addition, we preprocess our pseudorange residuals to separate the ephemeris estimate from the clock estimate. Separating the er- rors caused by SA (fast) from those caused by satel- lite ephemeris error (slow) enables prompt and efficient messages to be broadcast to the WAAS user. Conse- quently, WAAS accuracy will be improved. Compar- ison between WAAS performance with and without these process models will be given, where this com- parison is based on the actual data collected by our real time experimental WAAS network.