lou at unavco.org
Fri Sep 3 09:50:25 MDT 2004
>>> Are the Terrasat .OBS files at http://www.gps-survey.com
>>> in GeoTracer format?
>> It certainly looks like GeoTracer (receiver) output. But not fully
>> identical to
>> the GeoTracer 2000 (L1 only) which I implemented. I think the Z-XII is a
>> dual frequency receiver, this might account for the differences.
>> You can contact Herbert at Herbert_Landau at trimble.com
> Thanks. It's probably time to contact a source. :)
> The other item that I've noticed about the Z-12 RINEX at this
> site is it doesn't have bit 2 of the LLI flag set for A/S on
> (which was on all SVs for the day in question).
> If the Terrasat software is re-translating Z-12 data into
> a GeoTracer format (the .OBS files), and then GeoGenius is
> translating that format into RINEX, the situation might be
> that there are not the necessary fields in the GeoTracer format
> to fully reconstruct RINEX as it would be if one were to
> directly translate the Z-12 data into RINEX with ASRINEXO,
> ash2rin, teqc, etc. (Thus, potentially, the ms reset problem,
> missing A/S flag, and whatever else.)
Sam sent me a Word doc dated 5 July 1994 from Herbert Landau
which describes (as best I can tell from a quick scan) the .OBS
format found at the above site (output by the GeoTracer receivers,
though my impression is that Dr. Landau would call this the
A couple of quick observations:
Even though Sam was interested in this for the L1-only GeoTracer 2000,
it appears that the format could hold dual-frequency observations
as well, so no problem for dual-frequency Z-12 data.
There doesn't appear to be any "A/S in effect" flag for GPS SVs.
There is no explicit storing of receiver millisecond resets or the
receiver clock offset value, though this could be stored in the
epoch time. Phase and pseudorange are stored as 8-byte floats.
My guess, given the later translation to RINEX (by GeoGenius, we
think), is that the Z-12 data translation to this format is just
a direct translation of the raw phase, pseudorange, and time tag
values. Thus reconstruction of proper RINEX from this format wrt
to receiver millisecond clock resets is not done. (It could be done
-- detection of ms resets by looking for ms jumps in the pseudoranges
-- but we know from the resulting RINEX that this step isn't done.)
More information about the teqc