Question

I'm using ordGen to evaluate observed range differences for my receiver, and I'm running into some problems for any other recevier than the Trimble NETRS. To summarize, OrdGen seems to work with Trimble receivers only, and nothing else.

First: I'm running the latest version from SVN, on OSX. Everything built fine with out errors, using make. I have a topcon receiver, and I used the tps2rin tool to produce rinex (provided by topco

I ran ordGen on a file to evaluate the observed range differences for a receiver, using a precise ephemeris. .

~/gpstk/dev/apps/reszilla/ordGen -o rin3.08o -e igs15022.sp3 -d -v | more

# program:ordGen

# debugLevel: 1

# verboseLevel: 1

# Sending output to stdout

# Reading ords from stdin

Reading rin3.08o as RINEX obs data. # Antenna Position: (-2691463.8, -4263160.5, 3893614.7) Reading rin3.08o as RINEX obs data. # OrdEngine using smart mode

# OrdEngine using L1 P range and L2 P range

# Time Type PRN Elev Azimuth ORD(m) wonky

2008 295 23:40:53.1 0 2 59.75 46.43 -35735.17686 0

2008 295 23:40:53.1 0 4 15.09 48.45 -29101595.95112 0

2008 295 23:40:53.1 0 5 77.84 247.99 -35825.72259 0

2008 295 23:40:53.1 0 9 11.46 199.84 -35886.67430 0

Something seems wrong here, as these ORD values seem huge. So, I checked recever position, and it is correct. My receiver is at lat 37.86416447510, lon-122.26672944150, which corresponds to the ecef position. Now, my receiver is moving, but is entirely static for the processed observations, in addition, the recevier position corresponds to the static position. .

So, to check the ordGen program, I ran on some other files. Running ordGen on the example files included with GPSTK, I get the following output (truncated).

../apps/reszilla/ordGen -o bahr1620.04o -e bahr1620.04n -d -v | more

Reading bahr1620.04o as RINEX obs data.

# Antenna Position: (3633909.1, 4425275.5, 2799861.3)

Reading bahr1620.04o as RINEX obs data.

# OrdEngine using smart mode

# OrdEngine using L1 P range and L2 P range

# Time Type PRN Elev Azimuth ORD(m) wonky

2004 162 00:00:30.0 0 4 16.06 39.57 111891.07397 0

2004 162 00:00:30.0 0 5 69.33 352.17 111889.60966 0

2004 162 00:00:30.0 0 6 24.46 282.25 111894.00296 0

2004 162 00:00:30.0 0 9 43.54 203.61 111891.38343 0

Observed range differences of 10^6 meters seems very, very large. Is this a junk dataset? To further test, I ran on some data I took from CORS. Running the following command:

./ordGen -o ~/nt_data/131/chab2740.08o -e ~/nt_data/131/chab2740.08n -d -v | more

# Antenna Position: (-2685744.3, -4278242.1, 3881399.5)

Reading /home/matlab/nt_data/131/chab2740.08o as RINEX obs data.

# OrdEngine using smart mode

# OrdEngine using L1 P range and L2 P range

# Time Type PRN Elev Azimuth ORD(m) wonky

2008 274 00:00:00.0 0 4 44.16 51.80 -55090831.73518 1

2008 274 00:00:00.0 0 9 39.66 217.84 -56119569.08878 1

2008 274 00:00:00.0 0 10 12.09 151.85 -62041386.75626 1

2008 274 00:00:00.0 0 12 70.87 327.52 -52546372.58886 1

2008 274 00:00:00.0 0 24 10.71 212.22 -62092214.88948 1

2008 274 00:00:00.0 0 29 16.60 268.80 -61234669.95226 1

I again received what looks like junk data. This time, however it reports wonky. Changing modes seems to help, with the CORS data.

./ordGen -o ~/nt_data/131/chab2740.08o -e ~/nt_data/131/chab2740.08n -d -v --omode c1| more

# OrdEngine using c1 mode

# OrdEngine using L1 C/A range

# Time Type PRN Elev Azimuth ORD(m) wonky

2008 274 00:00:00.0 0 4 44.17 51.80 4.58663 0

2008 274 00:00:00.0 0 9 39.67 217.84 4.27813 0

2008 274 00:00:30.0 0 4 43.97 51.67 4.81643 0

2008 274 00:00:30.0 0 9 39.46 217.66 3.90172 0

2008 274 00:00:30.0 0 10 12.27 151.73 13.01645 0

2008 274 00:00:30.0 0 12 71.10 327.62 1.29552 0

So, for the CORS data, changing modes helped. But for my data, and the example data, it didn't. None of the modes I tried (c1, p1, smart, etc) seemed to help.

Any ideas on why most of my ORDs are coming out wrong, and what I can do to correct it?

Follow-Up:

I re-ran a lot of CORS data, and I found that OrdGen seems to produce junk output for any receiver other than the Trimble NETRS receiver. I have tried data from Topcon Receivers, trimble 5700 and Ashtech receivers, and ordgen produces junk.

Once the data is exported to Rinex, shouldn't it be receiver independent? Any ideas on how I can make ordGen work with a TopCon receiver.  

Thanks,

-- JackTisdale - 27 Nov 2008

Answer

ALERT! If you answer a question - or have a question you asked answered by someone - please remember to edit the page and set the status to answered. The status is in a drop-down list below the edit box.

You are correct in thinking that ordgen is not receiver specific. The developer has looked at this and is not sure what is doing on. I'm going to pass along your email to him so he can start a dialog and figure out what is going on.

-- RickMach - 01 Dec 2008

I think what you are seeing here is a feature and not a bug smile . ordGen was never intended to be used alone. At the very least you have to pipe its output through ordClock, which solves for the clock bias. There is a detailed writeup in the new GPSTk user manual.

-- BenHarris - 20 Jan 2009 No such template def TMPL:DEF{PROMPT:supportquery}
Topic revision: r4 - 20 Jan 2009, BenHarris
 

This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding Foswiki? Send feedback