An Algorithm for Estimation and Separation of Ephemeris & Clock Errors in SBAS
- Paper number
IAC-07-B2.1.10
- Author
Mr. Saumyaketu Mishra, Indian Space Research Organisation (ISRO), India
- Coauthor
Mr. Rajneesh Gupta, Indian Space Research Organisation (ISRO), India
- Coauthor
Mr. A.S. Ganeshan, Indian Space Research Organisation (ISRO), India
- Year
2007
- Abstract
The estimation and separation of ephemeris and clock errors is an integral part of a GPS based system like SBAS. Generally, the method is based on the full state approach to solve for satellite errors (ephemeris and clock) and station errors simultaneously with a least square estimator, or the other way is to sequentially estimate the satellite ephemeris through a Kalman filter, using a complex model of the satellite dynamics. In this paper, the estimation and separation of ephemeris and clock errors is addressed through a unique approach of combining both the methods, in a simple and elegant approach. This algorithm estimates these errors as corrections to the broadcast GPS navigation data, required by an SBAS user for accuracy improvement. It is important to note that the measurements are pre-processed for removal of errors in phase measurements and wild points in pseudo range measurements; dual frequency smoothing of carrier phase with code measurements, followed by the removal of other system errors (troposphere, broadcast GPS clock, geometric range and receiver clock). In the algorithm, a single difference technique is employed first on the pre-processed measurements, to separate the ephemeris and clock components and estimate these later using two different Kalman filters. In the algorithm, the method uses a priori information of ephemeris errors along with pre-processed measurements through a minimum variance estimator and provides initial estimates of the ephemeris corrections and their variance. A Kalman filter is designed to further smooth the ephemeris corrections, using minimum variance estimates as measurements and a simple kinematic orbit model for state propagation. These ephemeris corrections are used in conjunction with pre-processed measurements, to form the clock error measurements for all satellites. Similar to the ephemeris estimation approach, the clock error estimation process also incorporates two stages. First, a weighted least square method is used to provide initial estimate of satellite clock and its variance. Further, this clock estimate is treated as a measurement along with a second order Gauss Markov process as a state model for clock dynamics in the Kalman filter for clock. The clock Kalman filter provides the estimates of clock rate also simultaneously. For performance evaluation of the algorithm, the algorithm is qualified first through data simulation in terms of known errors in ephemeris/clock and their subsequent retrieval. The algorithm performance based on standard dual frequency code and carrier phase measurements at 1 Hz rate, has also been presented and discussed in the paper.
- Abstract document
- Manuscript document
IAC-07-B2.1.10.pdf (🔒 authorized access only).
To get the manuscript, please contact IAF Secretariat.