View Single Post
  #16   Report Post  
James
 
Posts: n/a
Default


Tom

this may well be one of the best posts I have reac on this group!

Thanks for sharing info so clearly.

Jimmy


Tom Cade wrote:
"anchorlt" wrote in message
om...

I keep reading about GPS for boats and how innacurate it can be.

Why is the same not innacurate in cars?

My car GPS is so accurate it tells me I am on the white line at a
traffic signal stop light or in my driveway.

Why not the same for boats?



I'm a land surveyor (and boat owner) and I've been working with GPS for
about 11 years now. Accuracy of GPS depends on several factors, but here's
some general info...

In the early days of GPS, SA (Selective Availability) was active. SA was a
method by which the Defense Department (DoD) purposely degraded the quality
of the signal and data to presumably prevent a potential enemy from using
the system for targeting purposes against US. The accuracy of an autonomous
(stand alone, no corrections, post-processing, etc...) GPS position during
that period was about +/- 100 m horizontally, about 150 m vertically at
two-sigma (95% of the time).

During that period, the marine community (and others) obviously had a need
for better accuracy than that. This is when the USCG began installing the
GPS beacons. These beacons consisted of a GPS receiver being placed over a
precisely surveyed point. The receiver would compute its autonomous position
based upon the satellite signals. The resulting GPS position would be
compared to the known position, and a correction computed. That computed
correction data was then broadcast over the airwaves, and was freely
accessible to anyone who had a "beacon receiver". Some GPS receivers had
beacon capabilities built-in, other set-ups used a seperate beacon receiver
linked to the GPS unit via a NEMA or RTCM interface. With beacon corrections
applied, the accuracy of the GPS was on the order of 10 to 15 m
horizontally.

Towards the end of the Clinton era, the White House ordered the DoD to come
up with some other method of safeguarding us from hostile use of GPS. There
were too many developing commercial uses for GPS to limit the accuracy,
vehicle navigation being one of them. So in May 1998 (I believe), SA was
turned off. The result was that just about any commercially available GPS
receiver could now determine positions to about 10 - 15 m without the aid of
the beacons.

During this time, the FAA began developing WAAS (Wide Area Augmentation
System) as the aviation community's answer to GPS accuracy. WAAS makes use
of corrections computed from a network of ground-based GPS receivers all
over the country. The correction data is computed for various regions (the
corrections will vary over distance and location) and uploaded to various
geo-stationary satellites which each have an "area" of coverage on the face
of the earth. From these satellites, the corrections are broadcast over
their respective coverage areas. The WAAS correction is broadcast on the
same frequency as the GPS L1 signal, somewhere around 1.2 GHz. Since it's on
the same frequency as that used by most consumer GPS receivers (there are
currently two GPS frequencies, with plans to add three more), GPS
manufacturers simply had to add the capability to use the data. Both the FAA
and the NGS (National Geodetic Survey) have performed extensive testing on
the accuracy of WAAS enabled GPS. The results have shown accuracies of
better than 5 m horizontally, at 95%. My personal experience with a Garmin
GPS76S handheld has shown typical accuracies of 3-5 m (I do high accuracy
geodetic control, so have lots of chances to compare). Not all GPS units are
WAAS capable, so if you're in the market, make sure it says so. Also, as a
CYA, the FAA has not officially sanctioned WAAS for use in mission-critical
applications, where lives and property are at stake. The system currently
broadcasts a "health" bit in the data stream saying the system is not
functional. Most WAAS capable receivers have an option to use the data even
though the health bit has been set. You may need to change the default
settings on the receiver.

So, the accuracy of the GPS, be it a vehicle, vessel or aircraft is
essentially the same. You will most likely see better performance on a boat
since you should typically have an un-obstructed view of the sky. As others
have mentioned, the most likely cause of any discrepancies you might see
when watching the map display is the accuracy of the base map data. Most of
the mapping and charting that has been done in the US was originally based
upon the North American Datum of 1927 (NAD27). Recent mapping may have been
done on NAD83, a more recent datum. The GPS is based upon the World Geodetic
System of 1984 (WGS84). Confused yet? WGS84 and NAD83 are very nearly
identical (To the geodesists, yes, I know they are not identical...) and for
most purposes they may be treated as the same. NAD27 is a different story.
If you plot a NAD83/WGS84 position on a NAD27 map or chart, it will be in
error somewhere on the order of 30 m in the northing and 100 m in the
easting (Lat and Lon). The exact error is dependent upon your latitude. The
numbers cited are good at around 45 degrees north.

Therefore, it is critical that for navigation, both your GPS and the
maps/charts are on the same datum. Most GPS receivers can be set to work on
just about any datum you want, but it needs to be the same as your maps.

Because most of the mapping was NAD27, and the GPS uses WGS84, many mapping
packages have been "transformed" from the old system to the new.
Unfortunately, there is no exact mathematical translation from NAD27 to
NAD83/WGS84. Some map makers have done a better job of transforming their
products than others. Some have been transformed using the best models
available from the NGS and others, but some have merely been
"rubber-banded". It's advisable to do your homework and make sure that both
your base maps and GPS are capable of working in the same datum.

Sorry for the length, but hope it helps...

Tom




x-- 100 Proof News - http://www.100ProofNews.com
x-- 3,500+ Binary NewsGroups, and over 90,000 other groups
x-- Access to over 1 Terabyte per Day - $8.95/Month
x-- UNLIMITED DOWNLOAD