[teqc] Questions regarding TEQC qc-reports

Lou Estey lou at unavco.org
Fri May 8 07:42:14 MDT 2009

The following questions came in to me personally, but they are probably
of enough interest to the entire list that I'm sending the questions
and my answers on to everyone.


 > we sometimes discover an 'a' in the qc-plot, most likely when a satellite
 > just is rising or disappearing. I can't find an explanation for this in the
 > TEQC manual. This behaviour seems to connect with strange results in
 > Bernese calculations.

You can always get a listing of the markers in the ASCII time plot by
executing `teqc ++sym`.  The markers have evolved over time, but the
current listing is:

[2754] teqc ++sym

QC Symbol codes for ASCII plot:

Symbol codes for "SV":
(hierarchy is left-to-right, top-to_bottom)
C  receiver clock slip                m  n-msec multipath jump
I  ionospheric phase slip             M  MP1 and MP2 slips
1  multipath MP1 slip only            2  multipath MP2 slip only
-  SV above elev mask, but no data    L  Bit 0 of LLI set (rx lost lock)
+  SV data, but below elev mask       ^  partial SV data below elev mask
.  no A/S; C/A                        c  no A/S; L1 C/A
:  no A/S; L1 P1                      =  L1 C/A L2C
~  no A/S; L1 C/A L2 P2               *  no A/S; L1 P1 L2 P2
,  A/S on; C/A                        a  A/S on; L1 C/A
;  A/S on; L1 P1                      e  L1 C/A L2 L2C
o  A/S on; L1 C/A|P1 L2 P2            y  A/S on; L1 P1 L2 P2
N  GLONASS: unknown freq chan #       _  no SV data and below elev mask
note: "no A/S" == GPS antispoofing off or unknown, or not GPS SV

So 'a' means there is at least one epoch in that column where antispoofing
is on, no (full) tracking on L2 (i.e. there might be pseudorange, but no
L2 phase), but there is L1 phase and C/A pseudorange.  At low elevations
this typically means the receiver is not tracking L2 but is tracking L1.
It's very common to see this.

 > There is also a line in the qc-report which says:
 > other msec mp events
 > What does this line indicate?

One thing that teqc's qc is looking for is millisecond jumps in the observables.
If present, this might indicate a clock problem in the receiver or a problem
with the translator.  One metric used to look for these is jumps in the multipath
bias (i.e. a multipath "slip") were the magnitude of the jump is 1 msec.
Obviously one has to have a little + or - delta in detecting such a jump.
The built-in delta is +/- 1/100th of a millisecond.  So in a random population
of multipath slips, there is a 1:50 chance that a normal multipath jump
is N*msec +/- 1/100th of a millisecond.  So, all multipath jumps that
fall within this bound are added up and displayed on that line, along
with the total number of all multipath jumps.  (Furthermore, the multipath
jumps that fall within this bound are shown with a 'm' marker; see above.)
If the ratio is about 1:50, then everything is as expected.  If the ratio
is abnormally skewed, say, 30:100, then something abnormal is going on.
Is this observed? Absolutely.  Here's a qc from a randomly selected site
from EarthScope's PBO array using Trimble's NetRS:



other msec mp events    : 40 (: 103)   {expect ~= 1:50}

and if you look at the ASCII time plot you'll see the 'm' markers
typically at low elevations.  With 103 detected multipath slips,
one would expect maybe 1-3 jumps that would be N*msec +/- 1/100th msec,
not 40.  So something strange is going on with the NetRS, and this isn't
anything new because I reported this behaviour to PBO about their
NetRS data back in late 2003 or early 2004.


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