Ken wrote:
As a design goal, it would be helpful to the OD process if the delay could be made consistent within about +/-5 micro-sec -- i.e., 1500 meters (smaller is better). It is also important to quantify the standard deviation to that bias is since that is also factored into the OD process.
Let me chime in since this is one of my specialty areas.
It has been my hope/plan that the delays will be very accurately calibrated, and that we will have "certified" ground stations that are also well calibrated. I've been pushing to have ALL local oscillators on EAGLE and all timing clocks (remember that the 100-200 kb/s sampling clock is yet one more LO in the system) derived from a single, high stability clock. In addition to ranging, it is my hope that we can have an amateur, global time synchronization capability at the usec level.
When (occasionally) it is time to do the ranging task, it might be necessary to swap in a different, optimized software into the SDX -- this would be my backup just in case the delay is not deterministic when normal SDX is running.
Even if the delays are not as accurately determined as I hope, I see no reason that they will not be stable over times of minutes. With this being true, then we can do very accurate DIFFERENTIAL delays using N "certified" ground stations. If you do this at ~20,000 km range with stations ~1 earth radii, the difference observable only suffers from a factor ~3 degradation. Every ground station pair you add (the number of pairs M increases with the number of stations N as M=(N)*(N-1)/2) gives a statistical improvement of of sqrt(M-1); hence the geometrical loss from using a differenced observable is easily made up on the ground.
73, Tom