Article 53210 of sci.geo.satellite-nav:
Path: matra.meer.net!news1.best.com!news.maxwell.syr.edu!ais.net!news.rain.org!not-for-mail
From: Jeff Valine <jvaline@pleasedontspamme-illgen.com>
Newsgroups: sci.geo.satellite-nav
Subject: Re: Professional GPS Usage
Date: Thu, 18 Dec 1997 12:10:22 -0800
Organization: Regional Alliance for Information Networking (RAIN)
Lines: 55
Message-ID: <3499832E.90850BEB@pleasedontspamme-illgen.com>
References: <34973B74.22DEDC10@NOSPAM.ipass.net> <67blvl$s97$1@qvarsx.er.usgs.gov>
NNTP-Posting-Host: 198.68.152.27
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Transfer-Encoding: 7bit
X-Mailer: Mozilla 4.04 [en] (Win95; I)
Xref: matra.meer.net sci.geo.satellite-nav:53210

Close... RTK is typically used by surveyors to do relative positioning,
for example, when a site is staked-out for construction.  You place the
base unit on a mark, and then you can walk about with the rover.  Often
the waypoints are preloaded into the rover, and the rover will then
guide you to the next point, to within a few centimeters.  Typically,
for this application, the surveyor uses a local coordinate system, and
is not concerned with world-wide coordinates, just relative coordinates
to the base.  You are right that they need to see the same constellation
to correlate the carrier phase-based pseudoranges between the two
receivers.  The data is transmitted from base to rover via either a FM
or spread spectrum radio modem. 

Phase tracking is not a function of the antenna you use, except that the
lower the noise the better.  Tracking the phase of the carrier signal
(L1=1575.42 Mhz, L2=1227.60 Mhz) is a function of the receivers ability
to resolve how many carrier cycles are in the delay of the signal
received from each satellite.  The C/A code is 1023 bits long, repeats
every 1 millisecond, and always sends a particular bit-pattern every six
seconds, so it is fairly easy to align C/A codes received from each
satellite.  The carrier cycles all look alike, so resolving the integer
number of cycles in the delay from the satellite is much harder.  It's
worth it though, if you assume a 0.1% resolution in the code-based
pseudorange, than you get about 29 cm for C/A, but 0.1% resolution in
carrier resolution is only 0.2 mm.  This is not the overall accuracy of
the units, just how well they can resolve the pseudorange.


Sorry, for the long-winded response, I guess I'm one of the folks who
wants to put a little more "sci" in sci.geo.satellite-nav.

Jeff


Joseph L. Jones wrote:
> 
> Kevin, I'll try to fill in some of the gaps left by the other responders,
> but there are some details I'll have to arm wave about as I have heard
> the explanations but was to dumb to 'get it'.
> 
> You are thinking of RTK, 'Real-time kinematic' gps. These use 2 very
> expensive ($30k ballpark) units both with phase-discriminating
> antennas. These antennas, well, um, they, well, they can tell
> the phase of the radio signal coming from the satelites, and this
> allows them to resolve position down to centimeters. That's
> precision, though, not accuracy. The accuracy part comes in as
> follows. One unit is placed on a point which is know to at least
> the accuracy you desire from the remote unit. The 2 recievers
> are in radio contact with each other, and 'agree' on the suite
> of satelites to use for position computation--they must be 'looking'
> at the same birds. Since both have phase discriminating antennas,
> are looking at the same birds, and one's position is know accurately
> and precisely, the software can determine the position of the
> remote unit to within a centimeter under ideal conditions. No
> post processing required, just bucks, and lots of them. Joseph
>