Sub-Meter Accuracy

  • Thread starter Thread starter deecee
  • Start date Start date
Carsten said:
Ron Wilson schrieb:




Don't get me wrong here. It would be too good if reliable submeter is
possible using only WAAS. But with an internal, rather simple antenna,
and L1 only?
Click to expand...

The "rather simple" antenna's are all tuned individually, if you look
closely at the patch you can see the tuning stubs in the centre of each
side. I'd recommend just taking my word for it rather than doing that,
you have to take a lot apart to get to the antenna. Plus it does have a
very large ground plane (and large antenna for that matter) for a
handheld internal antenna.

Also the internal antenna is a dual feed design which gives it better
multipath performance to start with. That is partly why in some
situations the internal antenna can be better than the standard geo
external antenna.


____________________________________________
reply address : newspost at . org
talon . uk
____________________________________________
I'm not paranoid about spam.
Well ok, maybe a little bit.
 
Ron said:
Why don't you try to borrow or rent a GeoXT and
do some testing for yourself? I think you will be amazed with its accuracy,
as I was. I was a doubter too, before trying it.
Click to expand...

Thanks for the vote of confidence, we try our best :)
 
Andy said:
The "rather simple" antenna's are all tuned individually, if you look
closely at the patch you can see the tuning stubs in the centre of each
side. I'd recommend just taking my word for it rather than doing that,
you have to take a lot apart to get to the antenna. Plus it does have a
very large ground plane (and large antenna for that matter) for a
handheld internal antenna.

Also the internal antenna is a dual feed design which gives it better
multipath performance to start with. That is partly why in some
situations the internal antenna can be better than the standard geo
external antenna.
Click to expand...

Hi Andy--could you elaborate on the dual feed design antenna a bit?
Thanks,
-Sam
 
Andy said:
You get 1cm rms on a Geo III? I'm impressed. And I'm fairly sure the guy
who designed it will be when I tell him ;)
Click to expand...

There's nothing like running a real test to find out. If Ron Wilson,
or someone else, wants to collect a little data I'll process it. With
a single test it's always possible to run into some problem that
screws things up so I'd suggest recording ~30 minutes of static
data at some control point within 10 km of a CORS station.
That should be enough data to obtain a cm-level position with
a high degree of confidence assuming the receiver is capable.
Just to make it interesting the receiver could be setup, say, one
decimeter due south of the control point or whatever and
if cm-level accuracy is possible I should be able to determine
the correct offset - without knowing it beforehand of course.
 
John Bonde said:
There's nothing like running a real test to find out. If Ron Wilson,
or someone else, wants to collect a little data I'll process it. With
a single test it's always possible to run into some problem that
screws things up so I'd suggest recording ~30 minutes of static
data at some control point within 10 km of a CORS station.
That should be enough data to obtain a cm-level position with
a high degree of confidence assuming the receiver is capable.
Just to make it interesting the receiver could be setup, say, one
decimeter due south of the control point or whatever and
if cm-level accuracy is possible I should be able to determine
the correct offset - without knowing it beforehand of course.
Click to expand...
 
John Bonde said:
There's nothing like running a real test to find out. If Ron Wilson,
or someone else, wants to collect a little data I'll process it. With
a single test it's always possible to run into some problem that
screws things up so I'd suggest recording ~30 minutes of static
data at some control point within 10 km of a CORS station.
That should be enough data to obtain a cm-level position with
a high degree of confidence assuming the receiver is capable.
Just to make it interesting the receiver could be setup, say, one
decimeter due south of the control point or whatever and
if cm-level accuracy is possible I should be able to determine
the correct offset - without knowing it beforehand of course.
Click to expand...


John,

I would love to. I tried to email you, but it bounced. Send me an email
and we can set up the details of the data collection.
 
Sam said:
Hi Andy--could you elaborate on the dual feed design antenna a bit?
Thanks,
-Sam
Click to expand...

Most standard patch antennas are single feed, this means that there is a
single connection from the patch to the amplifier. The GeoXT/XM use a
dual feed patch, two connections from the antenna to a differential
amplifier. Since ground bounce and most other multipath signals have the
opposite polarisation they can be filtered out by the diferential amplifier.
I know that's not really a huge amount more, RF isn't exactly my strong
point. Filtering and shielding I can cope with, antenna and amplifier
design I leave to the experts, trimble has plenty of them.


____________________________________________
reply address : newspost at . org
talon . uk
____________________________________________
I'm not paranoid about spam.
Well ok, maybe a little bit.
 
Andy said:
Most standard patch antennas are single feed, this means that there is a
single connection from the patch to the amplifier. The GeoXT/XM use a
dual feed patch, two connections from the antenna to a differential
amplifier. Since ground bounce and most other multipath signals have the
opposite polarisation they can be filtered out by the diferential amplifier.
I know that's not really a huge amount more, RF isn't exactly my strong
point. Filtering and shielding I can cope with, antenna and amplifier
design I leave to the experts, trimble has plenty of them.
Click to expand...

Thanks Andy
 
In working with Ron Wilson to try and process some carrier phase
info from his GeoXT I came across the following:

http://trl.trimble.com/dscgi/ds.py/Get/File-29670/SprtNote_PfOffice_CarrierPhaseProcessing.pdf

So it does seem that the carrier phase output on the GeoExplorers
is crippled and they can't do cm-level work. About a year ago I
had received some RINEX files, which did process to cm-level,
that I thought were from a GeoExplorer IIIc. It seems that they
were most likely from a Pro XRS and that I was wrong about
the potential capability of the GeoXT.

It would be interesting to hear why Trimble did this. The
only reason I can think of is to sell more expensive receivers.
 
Sam said:
The Lassen OEM receivers output full carrier phase, but not the carrier
fraction. My guess is that this low cost receiver would otherwise
compete with Trimble's multi 10k$ products. Maybe Trimble added the same
feature to the GeoXT?
Click to expand...

The performance of the GeoXT is not intentionaly capped or limited in
any way. It is the best we could make it given the size, battery life
and unit price restrictions on the design.
Do you really think we would not take advantage of the marketing benefit
of having a cm level handheld if we could do it with just a software
change? Trust me, if we could do it we would. We'd probably end up
selling it for much the same price as the top end ProXR, that way there
would be no risk of cutting into other products from other business
units. If a higher accuracy GeoCE is ever released it's only going to be
after a large amount of development work.

Andy
____________________________________________
reply address : newspost at . org
talon . uk
____________________________________________
I'm not paranoid about spam.
Well ok, maybe a little bit.
 
Andy said:
The performance of the GeoXT is not intentionaly capped or limited in
any way. It is the best we could make it given the size, battery life
and unit price restrictions on the design.
Do you really think we would not take advantage of the marketing benefit
of having a cm level handheld if we could do it with just a software
change? Trust me, if we could do it we would. We'd probably end up
selling it for much the same price as the top end ProXR, that way there
would be no risk of cutting into other products from other business
units. If a higher accuracy GeoCE is ever released it's only going to be
after a large amount of development work.
Click to expand...

Oh, come on Andy. It is intentionally limited to keep it from hurting
the Pro XR market. As long as your company believes it can make
more money by running two entry line products it will. Nothing
really wrong with that as you are in the business to make money,
but don't pretend it's a technical problem.

Motorola had a receiver, the LGT1000, which was remarkably
similar to the GeoXT in size, weight, and functionality. Even the
costs were fairly close. It was marketed for GIS data collection,
like the GeoXT. Its interface was a bit more primitive, being
based on DOS instead of WinCE, but that was to be expected
since it was released in 1992. Oh yeah, and it provided carrier
info with enough resolution for cm-level work. So you are
going to need a large amount of development work just to
match 1992 Motorola technology?
 
John said:
Andy wrote:



Oh, come on Andy. It is intentionally limited to keep it from hurting
the Pro XR market. As long as your company believes it can make
more money by running two entry line products it will. Nothing
really wrong with that as you are in the business to make money,
but don't pretend it's a technical problem.

...........
Click to expand...

I'm not going to pretend to be an expert on the inner workings of our
GPS recievers. But I do know what is discussed in the GIS engineering
group and I do know what the GeoXT reciever can do, one of the plus
sides to designing a product is knowing what it can and can't do.
Trust me, cutting into the ProXR market isn't an issue, the Geo has
already done that, ProXR sales have dropped a huge amount. And if we
could sell a cm level Geo at close to the price of a ProXR we would do
it, the profit margins would be huge.
Obviously all the evidence of this is confidential so you'll either have
to take my word for it or continue to think we cap it. For that matter
you only have my word that I even know what an XT looks like let alone
know what it can do.

As for keeping up with 1992 technology, the GPS portion of the XT is
only a fraction of it's size, the bulk of the space is taken up with the
Windows CE front end and the battery. If you take one apart you will
find that the GPS reciever is all on one board with a diameter of around
3 inches.

Andy
____________________________________________
reply address : newspost at . org
talon . uk
____________________________________________
I'm not paranoid about spam.
Well ok, maybe a little bit.
 

Members online

Total: 41 (members: 3, guests: 38)
Back
Top