[teqc] latest version of teqc on-line (11 Oct 2011)

Lou Estey lou at unavco.org
Tue Oct 11 11:14:22 MDT 2011


The latest version (11 Oct 2011) of teqc is (finally!) ready and on-line:
As usual, save a copy of your current executable in case there
is some unexpected problem with the new executable.

Stuart Wier at UNAVCO is now a co-developer of teqc, and his main additions
this iteration are the new GLONASS orbit fitting now done by default during
mixed GPS+GLONASS qc runs.  Please contact Stu, or the whole teqc at unavco.org
list (Stu's on the teqc list), if you have specific questions regarding this
new teqc and doing qc with GLONASS navigation data.

The main fixes and changes since 21 Oct 2010 are given in detail here:
but here's a very brief summary:

* teqc with the '+qc' or '+qcq' options can now do a "qc full" (see
http://facility.unavco.org/software/teqc/tutorial.html#sec_11 for more
information) using GLONASS navigation data and GLONASS pseudoranges, by
interpolation of GLONASS SV broadcast orbital locations, thus allowing
qc processing for GLONASS just as previous versions did for only GPS.
By default, '+qc' or '+qcq' will use GPS+GLONASS data for a joint point position
solution if both types are available (i.e. mixed GPS+GLONASS observations
in RINEX obs, and both GPS RINEX nav and GLONASS RINEX nav).

* a related new command line option, '-no_orbit' or '-no_orb', is used to
suppress either GPS(G) or GLONASS(R) navigation data during the antenna
point positioning with '+qc' or '+qcq'.  Thus, '+qc -no_orb R' should revert
to a qc analogous to the GPS-only point-positioning qc of previous versions,
or, similarly, '+qc -no_orb G' does a GLONASS-only point-positioning qc, or,
'+qc -no_orb GR' should do the equivalent of a 'qc lite' run even if, say,
GPS RINEX nav and GLONASS RINEX nav files are available.

Note: In the implementation currently used in teqc, GLONASS orbit interpolation
requires seven or more contiguous ephemerides for a good solution; nine or
more is slightly better. If GLONASS ephemerides are spaced half an hour part
(as is currently being broadcast) then you will need more than 3 hours of
GLONASS navigation data to do a qc full with GLONASS.  If doing a qc of
less than 3 hours of observation data, use one or more daily combined GLONASS
RINEX nav files (available at CDDIS, SOPAC, NGS, ...) for optimal results.

* There is now a second build for Windows (2000/XP/...) using gcc MinGW
(see http://en.wikipedia.org/wiki/MinGW for more information).  This should
be more compatible for Cygwin or related UNIX-emulation environments on

* ASCII output of raw navigation messages (esp. GPS CA 300-bit subframes)
with '+raw_nav' option if raw data is collected with the following:

   - Topcon & Javad [GD] - GPS CA
   - Javad [gd] - GPS CA, L2C, and L5
   - Septentrio 4017, 4018, 4019 (& deprecated 5894, 5927) - GPS CA, L2C, and L5
   - Trimble 0x57-5 and .dat record 26 - GPS CA

You can also experiment looking at GLONASS CA (SA) nav messages from
Septentrio 4026 and Trimble .dat record 26 -- although these are just
experimental at this point.  Similarly you can experimenting looking
at GPS L2C and L5, SBAS L1, and Galileo/GIOVE nav messages with Septentrio
4018 (& deprecated 5947), 4019, 4020, 4022, 4023, 4024, 4025, and 4029.
(Output of GPS L2C and L5 nav messages via Trimble .dat 26 is currently
blocked due to problems with Trimble's storage of those messages.)

* fix of the +/-G, +/-R, +/-E, +/-S, +/-ch, and +/-NaN_obs options
when values are >= 32 (esp. useful for GIOVE PRN values > 32; see
or GPS PRN 32 filtering with '-G32' or '+G32'

* yet more generalization for reading of Trimble RT27 in .dat record 26
and 0x57-6 (this takes care of all edge cases that we've encountered)

* reading of new capabilities in Ashtech B-file, e.g. GPS L2C data

* many, many other little odds and ends

Note: There is no build in this release for Solaris 5.9 x86 (our machine
here finally died -- so upgrade to 5.10!), nor a build that will run on
OSX 10.7 "Lion", see


One set of executables that are built but not generally made available
are those compiled with gcc on Solaris (Sparc and x86).  (The released
builds for Solaris use Sun's native compilers.)  I do these to find yet
more fuzz in the code, but generally we don't use gcc-compiled executables
on Solaris here at UNAVCO -- we find they are slower than the Sun-compiled
builds.  But if any one is interested, let me know.


Louis H. Estey, Ph.D.              office:  [+001] 303-381-7456
UNAVCO, 6350 Nautilus Drive           FAX:  [+001] 303-381-7451
Boulder, CO  80301-5554            e-mail:  lou  unavco.org
    WWW:  http://www.unavco.org   http://jules.unavco.org

"If the universe is the answer, what is the question?"
                                                -- Leon Lederman

More information about the teqc mailing list