[APG Public List] Regarding GPS

John Yates john at jytangledweb.org
Sun Oct 31 18:09:51 MDT 2010

On 10/31/2010 3:51 PM, Jay at 1familytree.com wrote:
> A few items were mentioned in the last several postings regarding GPS.
>>  > I don't mean to harp on this, but it is important to remember, current
>>  > GPS devices (that I own, anyway) only have an accuracy of +/-15 feet,
>>  > at best. That isn't quite as accurate as your 10'x10'. :-)
>>  >
> The device has the accuracy, it just doesn't report to you the number of
> trailing decimal points in order to show you.
> Actually, the various forms of accuracy and monitoring of the various
> Global Positioning Satellites create an accuracy of LESS than 12 inches
> and in some cases LESS than 4 inches.
> (See Wikipedia for more info) ...including
> "The phase difference error in the normal GPS amounts to 2–3 metres
> (6.6–9.8 ft) of ambiguity. CPGPS working to within 1% of perfect
> transition reduces this error to 3 centimeters (1.2 in) of ambiguity. By
> eliminating this error source, CPGPS coupled with DGPS normally realizes
> between 20–30 centimetres *(7.9–12 in*) of absolute accuracy./Relative
> Kinematic Positioning/ (RKP) is a third alternative for a precise
> GPS-based positioning system. In this approach, determination of range
> signal can be resolved to a precision of less than 10 centimeters*(3.9 in).*

The device I have was a $250 device (Garmin eTrex Legend Cx). It doesn't
have the full theoretical accuracy of the technology, or of the
most expensive devices. One device I know of today would only
cost about $80 and have the accuracy of +/-15 feet. (lowest end
Garmin eTrex, no downloadable maps or turn by turn navigation, but
will record locations just fine).

And the +/-15 feet I see, is consistent with my experience of trying
to find a location again from just the GPS coordinates I have
recorded. The +/- displayed by the device is probably what I'd
call a theoretical accuracy. It would be based on the technology,
the trigonometric equations, the speed of light (of radio waves),
and the circuitry in the device. An experimental accuracy could
also be determined, and just like your car seldom sees the
theoretical gas mileage maximum on the sale sticker, and a computer
seldom sees its theoretical maximum speed, (real world variables
always end up in the way) you will see an experimental accuracy that
is not as good as the theoretical.

To measure the experimental accuracy, I'd do this experiment:
Place an underground marker in an open grassy area. Use a
GPS device to record its coordinates. Then, give a dozen,
a hundred, or more people with their own GPS units the
coordinates you determined. Have each of them place a marker
where they think the buried marker is. The more people the more
accurate your estimate of the uncertainty will be. When
they are done, unveil the hidden marker, and measure the
radius from the hidden marker to all their placed markers.
Calculate the RMSD (Root Mean Square Deviation) in feet.
I suspect it will converge to only +/-15 feet. No, I'm not
going to do the experiment. Maybe geocachers would have done
this. (one person could theoretically do this, but then
there would be a "bias" in it because you know where the
buried item is, if only subconsciously. And if only one
other person marks flags, their previous flag placements
may bias future ones. Hmm, so maybe everyone should also
bury their flags until the end... aren't experiments fun! :-)


> To me the intellectual appeal of latitude/longitude or GPS is the
> tantalizing thought that perhaps there's an useful universal reference
> system for all locations, so one can easily model spatial relationships.
>   Numbers are so appealing!   I do love them.   LOL   So, even if GPS
> descriptors in this case are less precise, their appeal would lie in the
> ease of use and communication to others, compared to my more elaborate
> explanation of where my Grandma Nancy's grave is located.   Of course, I
> do not have a GPS so it's a moot point, but if I did I think I would
> have taken a reading and recorded the info.   just in case.   GPS
> references require the user to also have a GPS, which may be an issue at
> this point in time, but less of an issue for future researchers.  For
> those with a GPS device, it certainly seems to be a useful "finding
> aid."   LOL   except in this situation where there *is no physical
> marker* just getting to the approximate location is not particularly
> useful because there's nothing to see directly representing the grave,
> one still would need the info about the reference grave, and the
> relative location to that reference.
> Every Smart Phone has GPS .... pretty soon every phone will........
> times change faster that most of us prepare for.
> Who would have imagined that when flying in an airplane you can no
> longer buy a drink with CASH......only debit/credit cards allowed!
> Use your GPS wherever you can and trust the set(s) of numbers being
> given to you.  You will be glad you did not too far down the road.
> Best To all,
> Jay

More information about the APGPublicList mailing list