GPS accuracy

[poter]

I understand that the new GPS's are now far more accurate as they are using some data from the european system alongside the USA sats. is this correct?

My interest is not just Marine but from a survey location point of view.


poter

<hr width=100% size=1>

 

[Talbot]

new gps are more accurate than older as they (mostly) have a 12 channel receiver rather than the older 6. Thus inherently they have an improved accuracy. In addition a lot have one or more of the different systems that are being developed for even greater accuracy derived from differential GPS or the use of european or russian sattelites in addition to the GPS. Whilst these may be of some use if you are trying to bring a 100,000 ton freighter up a very narrow channel, or navigate through a minefield, to the average yachties they offer nothing of any great advantage. In all cases the fact that USA has disabled the Selective Ability form the GPS signals so everyone gets the same accuracy, its the most important step. How many people on this forum remember the wonderful increase in yacht navigation when we were able to get Decca for the first time!

<hr width=100% size=1>

 

[tome]

GPS receiver technology hasn't moved on much in terms of accuracy for some years. Very early receivers had limited channel counts which made them rather inaccurate. Recent improvements include augmentation systems from geostationary satellites which provide integrity checking and differential corrections over a limited footprint. The North American WAAS system was recently declared operational, but the European EGNOS system is still in a trial stage with transmissions being intermittent. The main improvements in receiver technology over the past few years have been reduced power consumption.

If you're looking for survey level accuracy then the only way to go is via differential corrections. There are various commercial services available, or you can set up a local reference station (as we do). Another way is to record raw data at a control point and at your survey points and to process this data post-event. This will give about 0.5m 1 sigma horizontal accuracy. Baselines up to hundreds of miles are possible.

If you want to go better then you have to invest in dual-frequency receivers with RTK capability: this will give you 2cm 1 sigma accuracy (bit of an overkill for marine). Baselines are limited to around 10km with this method.

<hr width=100% size=1>

 

[Cornishman]

The September Journal of Navigation (The Royal Institute of Navigation) has several articles connected with GPS which should be of interest to you, including Indoor Navigation System Using Asynchronous Pseudolites as a means of using GPS in obstructed environments, Effective Cycle Slip Detection and Identification for High Precision GPS/INS Integrated Systems, and Marine DGNSS Availability and Continuity.
The Journal is included in the Cambridge Journals Online service http://journals.cambridge.org

<hr width=100% size=1>

 

[Oldhand]

Rubbish! More channels doesn't necessarily mean more accuracy. If the software is, like I suspect on most cheap receivers, designed to employ every obtainable satellite in the fix solution, then the fix accuracy may well be degraded by "bad angle". Think about your cocked hats from a 3 way bearing or radio direction finding fix. Elongation of the triangle due to a low angle between 2 of the bearings results in poor accuracy. Unless possible low angles subtended from "pairs" of satellites in view to the receiver are illiminated from the fix, degredation will occur. Accuracy will also depend on how the software estimates/calculates corrections for the "radio range" due to refraction through the earth's atmosphere. This error (greatest for low elevation satlellites) is mostly illiminated by using a differential GPS system.

For best GPS accuracy available from commonly available GPS receivers you need to look at Leica products.<P ID="edit"><FONT SIZE=-1>Edited by Oldhand on 30/09/2003 19:57 (server time).</FONT></P>

 

[pvb]

You\'re right...

Yes, GPS is getting more accurate. The US has WAAS, and Europe has EGNOS - both are systems designed to improve GPS accuracy from around 10-15m to a staggering 1-2m. EGNOS is due to be operational around mid-2004, but there's some sort of testing system called ESTB which has been transmitting since 2000, although I don't know what sort of coverage it has.

Most current generation GPS units are WAAS/EGNOS compatible, and most use 12-channel receivers, so accuracy is good. I recently bought a Navman Tracker 5600 plotter, and I'm amazed by its performance. It's somewhat eerie to arrive in a strange marina and find that the plotter is showing the actual finger pontoon you're tied to!

<hr width=100% size=1>

 

[tome]

Oldhand

Satellite pseudoranges are weighted in the fix solution based on geometry (eg elevation and angles of cut) and signal levels. This is done to avoid the degradation you describe. It is universal practice to use an over-determined solution and all available SVs are used down to preset thresholds for elevation and signal levels. SV geometry dictates that solving height is the most inaccurate part of a GPS fix.

Leica, Trimble, Novatel and Thales (incorporating Ashtec and Magellan) all provide similiar levels of accuracy with Trimble the clear market leader in terms of survey equipment. In some parts of the world where sunspot activities are high (equitorial) it is common practice to use dual frequency tracking to overcome ionospheric effects which can cause severe position jumps.

Tom

<hr width=100% size=1>

 

[longjohnsilver]

Well I've got an old eagle GPS and an up to date Garmin, both equally accurate, will put me on the same diving spot time after time, always within a few feet. The switching off of SA made all the difference.

<hr width=100% size=1>

 

[bob_tyler]

Suggest that you have a look here.

http:// [url]http://esamultimedia.esa.int/docs/egnos/estb/schedule.htm [/url]

Mind you, if you can understand the technology, you are a better man than I am, Gungha Din.

If you can translate into normal English and post here even better. (Not into Jimi's Scottish please)

<hr width=100% size=1>

 

[tome]

Try this link:

<A target="_blank" HREF=http://esamultimedia.esa.int/docs/egnos/estb/schedule.htm>http://esamultimedia.esa.int/docs/egnos/estb/schedule.htm</A>

EGNOS is a European version of the Satellite Based Augmentation System. In the states they have WAAS (Wide Area Augmentation System) which is now operational. Japan has MSAS. This is a series of geo-stationary satellites which provide corrections to normal GPS satellites (which are in polar orbits) to improve accuracy. Since the satellites are staionary they can only cover a limited earth footprint, hence the need for regional solutions. They also transmit integrity signals which allow the user to see how accurate their position is, the idea being that the augmentation will allow GPS to be used in safety critical applications.

The SBAS satellites use the same frequency as the GPS sats but have a different data format (known as Signal in Space or SIS). This allows them to be received by GPS receivers, although they need additional software to decode and apply the data.

ESTB is the Egnos System Test Bed, based on Inmarsat satellites and currently running in test mode. It doesn't provide integrity information yet, and coverage is poor at present. The ESTB program is scheduled to end in April 2004 shortly after which EGNOS should be operational.

EGNOS will improve GPS accuracy from around 15-20m at present to potentially 1-2 metres by use of Wide Area Differential (WAD) corrections. It will also give integrity warnings within 6 seconds of any problems with either GPS or EGNOS.

So there's something to look forward to next season! Note that in order to receive EGNOS your GPS receiver must be RTCA/DO-229C MOPS compliant. This will allow it to use any of the regional SBAS solutions.

If you have a compatible receiver, you will be able to see some experimental data after 10th October when the Artemis satellite is activated. At present there is not much coverage in our area so reception is unlikely.

Regards
Tom

<hr width=100% size=1>

 

[duncan]

FWIW and in relation to normal retail GPS units - in practical terms I get up to a 30 ft variation on a standard GPS over the period of a few hours which reduces to between 5 - 10ft utilising a differential beacon.
At this level the charting is probably more in doubt than the position - but, as LJS says,youcan return to the same spot within the above variation over and over again as a fisherman or diver.

<hr width=100% size=1>

 

[tome]

Duncan

This sounds about right, using RDF differentials. It can be improved to better than a metre using suitable receivers and better reference stations. Errors tend to be slightly greater in the N-S axis as this is the direction of travel for the satellites.

Tom

<hr width=100% size=1>