2.13 Satellite Tracking

There was a project to use the VLBA to provide positional data to help navigate interplanetary spacecraft. For this project, the VLBA must be able to point at the spacecraft so the ability to do so was added to SCHED as of March 2004. The spacecraft positions are obtained with the help of spice files that are typically from JPL. In Dec. 2014, this capability was augumented to utilize TLE (Two Line Ephemeris) files.

The NAIF software package from JPL is called to read the spice files and calculate positions. It is also used in processing the TLE files. The NAIF software significantly increases the size of a SCHED distribution, and the satellite tracking capability is unlikely to be needed outside the AOC. Therefore the tracking capability is not included in the default SCHED distribution.

To use the tracking capability, a Satellite Initialization section needs to be included in the main input file. That section contains a group of inputs for each satellite. There are four input parameters in each group:

Note that the satellite routines also set the velocity for the satellite for use with DOPPLER. The satellite frequencies can be specified with their rest frequencies in a LINEINIT section.

There seems to be an incompatibility between the NAIF software used for satellite tracking and the code used for tracking planets based on a JPL ephemeris that is used elsewhere in SCHED. It is best not to mix the two. The satellite ephemeris files typically also contain the planets so, if you wish to point at both satellites and planets, you can do it with the satellite files alone. Just don’t set ephfile.

Note that, according to notes in the code, this satellite tracking section of SCHED does not take into account diurnal aberration which it should, because it is also not taken into account in the on-line system. The planet section of sched does take it into account. This leads to different calculated positions when using the ephemeris and when using the satellite tracking. Some day, this should all be handled better, but the effect is under an arcsecond so it does not matter for pointing antennas.

The items in the SATINIT section are:

  1. SATNAME: The name of the satellite. This is only used internally in SCHED. It is the name that should be used as the SOURCE in the scan inputs. This name is not sent to the NAIF software.
  2. SATNUM: The number, used in the spice files, for the satellite (or other celestial body, for that matter). This number is assigned by JPL. You need to know this number but I’m not really sure how you get it. This number is sent to the NAIF software to tell it which satellite to process. For satellites, these numbers are negative. They are positive for planets etc.
  3. KERFILE: A spice kernel file that gives information such as leap seconds. It is likely to have the extension .tls. Standard versions at NRAO are in the $PLANET_DATA area.
  4. SATFILE: The spice file for the satellite. It is likely to have the extension .bsp. Note that it must contain the satellite (or asteroid or whatever) you want to observe AND any other bodies needed to calculate the vector from the antenna to the body. That will usually mean that the Earth should be included and might require others, especially if the satellite is orbiting around some other body.
  5. TLEFILE: The Two Line Ephemeris file for the satellite. It is likely to have the extension .bsp.

When groups have been given for all satellites, give a line that contains the word ENDSAT and a slash.

If the above section is provided and one of the satellites is a source in the schedule, SCHED will call the NAIF software once for a nominal position for the summary file etc, then again for every scan for every stations to get updated positions and rates. It will also calculate an approximate parallax correction for each station. This can amount to several arcseconds, and the calculation is believed to be good to an arcsecond or better. The scan/station dependent positions are written to the .crd files for the VLBA. See the note below about VEX files.

For a satellite (or any moving source, for that matter), SCHED plotting can help you see where the object is going. In the RD (RA/Dec) plots, a line will be plotted for each scan. A likely use for this capability would be to obtain the transmission schedule for a satellite over some days or weeks, make a schedule with a scan for each period that it is transmitting, then make the RD plot and show the calibrators. This will help identify times when the satellite is both on and near a likely phase reference source.

There is a SCHED example, egsat.com that demonstrates the use of the satellite capability. Interested users are recommended to start with that example.

The scheme for handling moving sources in VEX files is not yet established. However, to correlate such observations on the VLBA DiFX correlator, a Vex file is needed for all the information other than the positions. Thus VEX files can be written when there are moving positions, but several warnings will be written about the use of such files. The positions of the moving sources should be obtained from ephemeris information at correlation time, separately from the VEX file. For pointing, positions may or may not be good enough depending on the rates. Also note that solar system objects may require offset pointing positions between different stations and that is not described in the VEX file.