[teqc] Does SNV message must be the 1st record at the beginning of raw data file

[Lou Estey]

hi Jerry,

> My PC is connected to GPS recevier directly. On a FC 3 box,  I made a
> PERL script to log the real time output from the ASHTECH Z-Surveyor
> receiver in 1-hour segments. I have enabled the output of raw data
> messages CBEN, DBEN, MBN, PBN and SNV.  Except the first 1-hour
> segments,  C1, D1 and D2  code are missing after the 1-hour segments is
> translated to RINEX file using TEQC,  only there are L1, L2, P1 and P2
> four observation types.
> 
> For example, Logging script launches at 03:10,  the first 1-hour segments
> will be created at 04:00,  it has L1, L2, C1, P1, P2, D1 and D2 seven
> observation types after translating to RINEX format using TEQC. The
> second 1-hour segments will be created at 05:00,  but it only has L1, L2,
> P1 and P2 four observation types after translating to RINEX format using
> TEQC, C1, D1 and D2 code are missing.  In the header section, there is a
> comment "note: cannot distinguish between L1(CA)=C1 and L1(P1)=P1".
> And same to  the next 1-hour segments.
> 
> I checked all of the raw data files and found that the 1st records at the
> beginning of the raw data file wasn't SNV message except the 1st 1-hour
> segments. So I made a test, edited one of the raw data files and removed
> records before SNV message at the begin of the file. Now there is no a
> comment "note: cannot distinguish between L1(CA)=C1 and L1(P1)=P1"
> in the header. But I found another thing.  In the O-file, only the 1st section
> has seven observation types, for the other secionts, there are only four
> observation types.
> 
> I'd like to make real-time estimation of atmospheric water vapor. So I have
> several questons,
> 
> 1.  What's the side effect for me if the 1st record at the beginning of raw
> data file isn't SNV messages?
> 2.  What's the meaning for "cannot distinguish between L1(CA)=C1 and
> L1(P1)=P1"?
> 3.  Should I have to make sure that the 1st record at the beginning of raw
> data file must be SNV message?
> 4.  Why only the 1st section has seven observation types and the other
> sections only have four observation types in the O-file?

I suspect that the leading SNV record is somewhat of a red herring.  My
guess (reading between the lines) is that the problem is due to the mixture
of data record types. There are basically two configurations of Ashtech
stream records that teqc will read in a unambiguous fashion:

multiple MBEN (or MBN) and one PBN per epoch
OR
DBEN and PBN per epoch (correct in versions 2005 May 6 or later)

(SNV can be present or not -- they would get translated to RINEX NAV
if they are there, but they are not necessary.)  I have no idea what
CBEN records are.

The data in MBEN (or MBN) and DBEN records are different, MBEN/MBN
being the more complete and, for example, contains doppler.  DBEN
does not distinguish between L1(CA)=C1 and L1(P1)=P1 and does not
supply doppler.

So here's my guess: You are allowing teqc to auto-identify the data,
i.e. you are not using a `-ash XX` option, and you are allowing teqc
to auto-obs the data, i.e. you are not using the `-O.obs XX` option.

Thus when the SNV record is first, teqc says "this is a MBEN/PBN
file -- it will have fully loaded observables" and it then reads
each MBEN/PBN record set and gives you C1 and doppler -- 7 observables
total.

I'm then guessing that when you strip off the SNV record, then the
next record is DBEN, in which case teqc says "this is a DBEN/PBN
file -- it has reduced observables" and it then read the DBEN/PBN
record set and supplies the 4 observables L1, L2, P1 and P2 -- which
are all you can get out of DBEN because it is a very compressed data
scheme.  You'd also get the header comment: "cannot distinguish between
L1(CA)=C1 and L1(P1)=P1" which is true for DBEN.

If you want the full 7 observables, just log the data without
DBEN (or CBEN), and translate the way you've been doing.  Leave SNV
records in; no problem there.

Or to try to get the full 7 observables from the MBEN/PBN record set
from you collected data, retranslate but additionally supply the
options: `-ash s` and/or `-ash s -O.obs L1+L2+C1+P1+P2+D1+D2`.
(Offhand, though, I don't remember if there really is a D2 in
MBEN.  Early versions of teqc "constructed" a D2 by rescaling D1;
this was eliminated by May 2001.)

hope this helps,
--lou

~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
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
~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~