[OSM-talk] All radio begins with the antenna

Nick Black nickblack1 at gmail.com
Fri Sep 29 12:55:31 BST 2006


On 9/29/06, SteveC <steve at asklater.com> wrote:
> * @ 28/09/06 09:51:13 PM lars at aronsson.se wrote:
> >
> > My new GPS receiver is a PDA, Mio P550, based on the new chip
> > SiRFstar III.  It's really a lot sharper than my old one, at least
> > as long as I measure latitude and longitude.  But the improvement
> > is not so evident when it comes to altitude.  In my car where the
> > GPS receiver is close to the windshield, I know reception is
> > always worse when I'm heading north, since the roof shades signals
> > from the south where many satellites are(*) if you're driving at
> > northern latitudes.  But this became very clear when I was driving
> > home from the Göteborg Book Fair this Saturday and turned north on
> > E4 after Huskvarna (57.93 N, 14.33 E).  The steep hills go up and
> > down, but my altitude reading varied almost independent of this.
> > Later I've been able to verify that lat-long are really sharp.
> >
> > Today I went to an open place and measured the altitude with both
> > the old and new GPS receivers, while sitting still for almost two
> > hours.  My results are shown in a diagram in my OSM diary,
> > http://wiki.openstreetmap.org/index.php/User:LA2
> >
> > My unexpected finding was how great a difference was made just by
> > positioning the device (and its built-in antenna).  Apparently,
> > the PDA must be held upright and should not lay flat on a table.
>
> I sent similar graphs to the list years ago
>
> http://www.fractalus.com/steve/stuff/gpstracks/index.html
>
> It comes, I think, from the geometry. You can imagine the spehere of the
> eath and the satellites flattened out to two planes. The range of values
> you get x & y (lat, lon) is much greater than the range of values you
> get in z (because they all fly at the same altitude).

Ionospheric interference also contributes to these kind of regular
errors that you see in GPS data.  It would be cool to do a similar
comparison with EGNOS enabled and EGNOS disabled on two receivers and
see what the resulting errors looked like.  In theory EGNOS should
compensate for a lot of the Ionospheric error, altough as this type of
error varies quite slowly you would have to test over a week or so.
Anyone have a roof?

Lars:  Have you used you SirfIII setup in urban areas?  I'd also like
to look at the different errors that occur in built up areas using
Sirf III and other chipsets.  I suspect that because Sirf III is more
sensitve, it would pick up more multipath and perhaps be less
accurate.

Nick


>
> Thats the analogy I use, anyway.
>
> >
> > (*) The web that GPS satellites spin around the globe has large
> > holes over both poles.  GPS satellites never go higher than 55 N.
> > I covered this in my diary of February 27, 2006,
> > http://wiki.openstreetmap.org/index.php/User:LA2/Diary_for_Q1_2006
> > If you are in Belfast, Copenhagen or Moscow (55 N), you
> > occasionally see satellites at zenith and close to the northern
> > horizon (over the Pacific), but there is a dark hole between
> > there.  I live at 58.4 N, so I never see GPS satellites at zenith.
> >
> >
> > --
> >   Lars Aronsson (lars at aronsson.se)
> >   Aronsson Datateknik - http://aronsson.se
> >
> > _______________________________________________
> > talk mailing list
> > talk at openstreetmap.org
> > http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>
> have fun,
>
> SteveC steve at asklater.com http://www.asklater.com/steve/
>
> _______________________________________________
> talk mailing list
> talk at openstreetmap.org
> http://lists.openstreetmap.org/cgi-bin/mailman/listinfo/talk
>




More information about the talk mailing list