Getting at the underlying data...

I would like to access the actual positional data that a Garmin etrex
reciever uses to compute a position.

For example: My Garmin Mariner only reports latitude-longitude values
to the nearest 1/10 of a second, which is about 10 FT.

I'd like to access the "raw" data used to compute this position so I
can increase the precision of the lat-longs, and then average
collections of these positions to come up with a more accurate point.

In essence, I'm wondering if there is a way to "hack" into the lower
level data string the etrex is using. Is there a way to get a NMEA
string out of the receiver?

Scott Huey

No need to hack. The precision is actually available in the track and
waypoint data downloaded to Mapsource (but Mapsource won't show it to you
or export it for you). To get at it I use G7toWin; other non-garmin
programs probably can be used, but this is the one I've used successfully -
Google G7toWin to find where you can download it if you don't have it
already. You can download your data to it directly from the GPSr or import
it from a gdb file that you've previously created in Mapsource. Once you
have the data in G7toWin you will see that it is displaying the data to
about 2 more significant figures than does Mapsource; if you then export the
data to a text file and load it into for instance Excel, you will see that
the data is actually carrying much more precision -- down to the resolution
in the NMEA string.

John

Thanks for the response John. I had a couple of more questions...

I'm just curious, how many decimal places does the NMEA string output
by the Garmin etrex receivers carry by default?

Can I increase the precision of the position by averaging the number of
points collected by the receiver? Or am I running into some limit of
the hardware and onboard algorithms used to compute the position?

My initial thought was to use the increased precision of the lat/long
values to set up some custom differential correction using another
receiver and a point with a known lat/long value. Would this work?

Scott Huey

Scott -

I should confess that I am not an authority on NMEA. I too would be
interested in the exact format and resolution of the NMEA data for the
Garmins if someone with more expertize could wade in with an explanation.
My own observation is that the (decimal degree) positions exported by
G7toWin have 7 decimal point precision, and that variations of 10^-7 occur
in the data. (This is for a 60CS and a 60CSx, but I think the eTrex would
be the same).

Averaging will improve your results; first of all, I don't think the eTrex
does waypoint averaging (unless this has added to the more recent units - 60
and 76 series units do do waypoint averaging). Even with a unit that does
waypoint averaging you can improve the results by doing your own averageing,
since you have the flexiblility of looking and data over different time
intervals and can exclude outliers.

Differential correction with consumer grade GPSrs such as the eTrex and
60/76 units isn't practical; the positions of two identical units sitting
side-by side can easily differ by several meters (there have been previous
discussions of what is needed to do differential correction in this
newsgroup and/or in sci.geo.satellite-nav). Also, the increased precision
you see in data exported through G7toWin doesn't really help much over using
the same data exported directly from Mapsource (but it's nice to see
"beyond" the precision of the Mapsource data).

John