Computing long term orbit and clock models with variable time-horizons

1. A method for generating time-horizons for long term orbit models for at least one satellite within a global navigation satellite system (GNSS) comprising:

computing a predicted orbit parameter in an orbit parameter prediction model;

producing an indicia of accuracy by comparing the predicted orbit parameter to a current orbit parameter;

if the indicia of accuracy indicates at least a predefined level of accuracy, setting a time-horizon for the orbit parameter prediction model at a first value;

and if the indicia of accuracy indicates less than the predefined level of accuracy, setting the time-horizon for the orbit parameter prediction model at a second value.

2. The method of claim 1 wherein the computing step uses historical ephemeris for computing the predicted orbit parameter.

3. The method of claim 1 wherein the computing step uses measured signals from the at least one satellite for computing the predicted orbit parameters.

4. The method of claim 1 wherein the orbit parameter prediction model is a clock model.

5. The method of claim 1 wherein the orbit parameter prediction model is an orbit model.

6. The method of claim 1 wherein the time-horizon is associated with the orbit parameter prediction model for each satellite in a GNSS satellite constellation.

7. The method of claim 1 wherein the at least one satellite is a satellite within a satellite constellation for GPS, GLONASS or GALILEO.

8. The method of claim 1 wherein the indicia of accuracy comprises residuals from a curve fit of historical ephemeris used in the orbit parameter prediction model.

9. The method of claim 1 further comprising identifying whether the at least one satellite is healthy or unhealthy.

10. The method of claim 9 wherein, if the at least one satellite is unhealthy, determining whether broadcast ephemeris is available for the at least one satellite, and using an orbit parameter from the broadcast ephemeris as a predicted orbit parameter within the orbit parameter prediction model.

11. The method of claim 1 wherein the orbit parameter prediction model is a clock model comprising an offset term, a rate term and an acceleration term, where the offset term is set to an af 0 term of the historical ephemeris and the rate term and acceleration term are set using a curve fit to an all term of the historical ephemeris.

12. The method of claim 1 wherein setting the time-horizon comprises setting the time-horizon prior to be prior to a scheduled change in at least one of satellite orbit and clock.

13. A method for generating long term orbit models for a plurality of satellites within a global navigation satellite system (GNSS) comprising:

generating an orbit parameter prediction model for each satellite in the plurality of satellites;

and generating a time-horizon, associated with each orbit parameter prediction model, within which the orbit parameter prediction model is accurate.

14. The method of claim 13 wherein the time-horizon generating step further comprises:

computing a predicted orbit parameter in an orbit parameter prediction model for each satellite in the plurality of satellites;

producing an indicia of accuracy by comparing a predicted orbit parameter to a current orbit parameter;

if the indicia of accuracy indicates at least a predefined level of accuracy, setting a time-horizon for the orbit parameter prediction model at a first value;

and if the indicia of accuracy indicates less than the predefined level of accuracy, setting the time-horizon for the orbit parameter prediction model at a second value, where the second value is less than the first value.

15. The method of claim 14 wherein the computing step uses historical ephemeris for computing the predicted orbit parameter.

16. The method of claim 14 wherein the computing step uses measured signal from the at least one satellite for computing the predicted orbit parameter.

17. The method of claim 14 wherein the orbit parameter prediction model is a clock model.

18. The method of claim 14 wherein the orbit parameter prediction model is an orbit model.

19. The method of claim 14 wherein each satellite is a satellite within a satellite constellation for GPS, GLONASS or GALILEO.

20. The method of claim 14 wherein the indicia of accuracy comprises residuals from a curve fit of historical ephemeris used in the orbit parameter prediction model.

21. The method of claim 14 further comprising identifying whether each satellite is healthy or unhealthy.

22. The method of claim 21 wherein, for each unhealthy satellite, determining whether current broadcast ephemeris is available for the unhealthy satellite, and using an orbit parameter from the current broadcast ephemeris as a predicted orbit parameter within the orbit parameter prediction model.

23. The method of claim 13 wherein the orbit parameter prediction model is a clock model comprising an offset term, a rate term and an acceleration term, where the offset term is set to an af 0 term of the historical ephemeris and the rate term and acceleration term are set using a curve fit to an afl term of the historical ephemeris.

24. The method of claim 13 wherein generating the time-horizon comprises setting the time-horizon prior to be prior to a scheduled change in at least one of satellite orbit and clock.

25. Apparatus for generating long term orbit models for a plurality of satellites within a global navigation satellite system (GNSS) comprising:

an orbit prediction module for generating an orbit parameter prediction model for each satellite in the plurality of satellites; and

a time horizon prediction module for generating a time- horizon, associated with each orbit parameter prediction model, within which the orbit parameter prediction model is accurate.