[Qgis-user] wishing for accurate lattitude/longitude from a cell, phone

[Garth Fletcher]

Thanks for that very intriguing info, which I had missed.

For others interested the link mentioned is found in
Qgis-user Digest, Vol 171, Issue 50
  Message: 5
  From: Randal Hale <rjhale at northrivergeographic.com>
  where Randy provides this link:
<https://www.fs.fed.us/database/gps/mtdcrept/accuracy/index.htm>

Which provides important comparisons by model and by canopy ("open",
"light-medium" and "heavy-closed").

The website is searchable by manufacturer/model, or you can download
the entire 750 test results in PDF or XLS format.
Very valuable comparisons.

At first I was confused by the "Position" column in both PDF and XLS
formats, until the website clarified this to being the number of
positions that were averaged together.

There is also a useful "NSSDA Standards" link.

Looking specifically at the Bad Elf GNSS Surveyor:
it is interesting to note the relatively small improvement when going
from single measurement to 60 averaged measurements.  This confirms my
belief that the errors are not primarily random noise, which should
decrease almost 8 fold when 60 measurements are averaged. This is
consistent with my observation that the reported locations "wander"
slowly as the satellite geometry changes, i.e., are not random noise.

Looking at the Garmin eTrex 30 information:
Again there is negligible improvement from averaging.
The use of GLONASS does very significantly decrease the accuracy
(increases the errors), independent of whether WAAS is used.
The latter is quite surprising, very counter-intuitive, and very
intriguing!

As recommended I will compare some measurements with and without
GLONASS enabled.

On 5/31/20 3:44 PM, qgis-user at stripfamily.net wrote:
> On 5/31/2020 1:20 PM, Garth Fletcher wrote:
>> The eTrex was receiving both US GPS and Russian
>> GLONASS satellites, plus WAAS (as indicated by a "D" in the Garmin's
>> satellite signal bars).
> 
> earlier in this (very interesting) thread a pointer was provided to a 
> speadsheet that contained the results of the USFS tests of a wide range 
> of GPS units. One thing that was surprising was that for a number of 
> devices, including the eTrex30, the accuracy /decreased/ when GLONASS 
> was enabled in addition to GPS. You might want to repeat this experiment 
> without GLONASS to see if you experience the same effect.

-- 
Garth Fletcher