Dgps

[Jakobs team]

Hello Everyone,

Tonight I was out at one of my Caches and was so amazed that with my Etrex camo I was getting accuracy of 5 feet on 12 sat,s with 9 diff symbols, my question is how low will a gps go.

Jakobs team.

[+The Forester]

"Accuracy" is a misleading misnomer by Garmin.

 

In position-fixing Accuracy is defined as the degree of closeness to the true value. Your GPSr doesn't know the true value, so it can't tell you how accurate a fix is. If it knew the true value, then the "accuracy" number would always be zero.

 

The Figure of Merit which Garmin refers to as "accuracy" and some other manufacturers call "Estimated Position Error" is actually a measure of an estimate of the scatter of fixes.

 

As an analogy, consider the scatter of hits of a blunderbuss or short brrel shotgun being fired at the roundels of an archery target. Let's say a hundred pellets are fired. The co-ordinates of the centre of the hit is usually taken as being the mean position of all the hits. The radius of a circle which encloses 50 of the hits is called the Circular Error Probable (CEP). Note that the CEP does not tell you how far the centre of the cluster of hits is from the centre of the target, only the size of the cluster. Also note that a CEP (which Garmin chooses to call "accuracy") of let's say 10 feet is actually telling you that there is a 50% chance that any single fix is more than 10 feet from the indicated co-ordinates.

 

To observe the disconnect between CEP and accuracy for yourself, input the published co-ordinates of a trig pillar into your GPSr as a waypoint and plonk it on top of the trigpoint. Watch the difference between the FoM and the distance-to-go figure and notice that they are almost never the same. The distance-to-go figure really is an indication of the accuracy of the machine in this situation. As ever, the Figure of Merit is merely a description of the estimated size of the scatter of the fixes.

 

Cheers, The Forester

[+Fangsy]

Very good explanation that.

I've often found that even when my Garmin says it has an 'accuracy' of say 30 feet, when I am sitting on top of a cach the Distance to waypoint is often very low, under 4 feet, so even with a large apparent area of accuracy the GPS is pretty much spot on. you just wouln't rely on it in life or death situations

[JackiePenn]

<snip>

As an analogy, consider the scatter of hits of a blunderbuss or short brrel shotgun being fired at the roundels of an archery target. Let's say a hundred pellets are fired. The co-ordinates of the centre of the hit is usually taken as being the mean position of all the hits. The radius of a circle which encloses 50 of the hits is called the Circular Error Probable (CEP).

<snip>

I have been playing recently with with various pocketpc software for use with my ipaq hooked to a bluetooth gps.

 

The nature of a bluetooth gps is that your flying blind to values that one takes for granted as being present on a conventional hand-held gpsr screen.

 

The reliance is upon the software the IPAQ/pocketpc is running and what information software developer has decided to convey to the user. None of the software I have used states EPE values. But often there are references to either an overall DOP (Dilution of Precision) or states a list of individual DOPs. The definition of a DOP is said to be...

 

A value expressing the confidence factor in the accuracy of the position solution based on current satellite geometry. The lower the value the greater the confidence in the solution.

.

 

I am not total sure yet how to reference these values to real world situation.

 

Anyway... the reason I have quoted the analogy of 'The Forester' is that one of the softwares used recently does an exact representation of the analogy he has given in a rather cool graphical way. You can leave the software running and it places a mark on the target every 1 sec or so. After obtaining a number of marks you can obtain an averaged co-ordinate reading for the position.

 

 

That is a screenshot from the pocketpc software called gpstuner.

 

Going back to the pocketpc software... I also like to know if WAAS/EGNOS corrections are being used, on a conventional waas enabled unit the user will usualy be presented with a message on the GUI such as "WAAS" to signify this. I have yet to find any of the pocketpc gps or navigation software that states the status of WAAS. I have found that I can look at the raw NMEA GGA sentence and easily see if DGPS corrections (and of what type) are being used by the content of that sentence.

[+Simply B]

I have yet to find any of the pocketpc gps or navigation software that states the status of WAAS.

OziExplorer CE shows whether it's got a GPS or DGPS fix on its status line, and if the latter then it also shows the "DGPS last update time." I have no way to test it, but is that what you're looking for?

[JackiePenn]

I have yet to find any of the pocketpc gps or navigation software that states the status of WAAS.

OziExplorer CE shows whether it's got a GPS or DGPS fix on its status line, and if the latter then it also shows the "DGPS last update time." I have no way to test it, but is that what you're looking for?

yeah thanks, I had read that OziExplorer did have that feature but I was begrudged to install another GPS Mapping Software just for that feature, as I have tomtom 5 and memory map already.

 

I am looking at pocket pc development with embedded visual c and a small ultility app should be trival to make. I just would have thought there would be one already. It is not so important to me now as it was a few days ago, because I am now happy WAAS/EGNOS has been enabled and is working when ever it can on the GPS.

 

I made the post on my blog, mainly to help me in the future after I forget what I did. I'll past here just for reference.

 

 

 

How to tell if the Holux GPSlim236 Wireless GPS Receiver with SiRF-Star-III chipset is using WAAS / EGNOS ?

 

After using the utility (from holux website) GpsViewer 1.5.exe to turn on WAAS, the only way I have found to see if the unit is actually using WAAS/EGNOS is to view the NMEA-0183 sentences ASCII output in a utility such a the SIRFstar Demo (also available from the Holux webby).

 

Before continuing, it seems to me that the GPSViewer does not turn on or off the WAAS setting in the GPS unit. I later used the features of the SIRFstar Demo tool to first switch over to the SIRF protocol to allow control over the GPS unit. Then from the Navigation drop down menu set the DGPS mode to AUTOMATIC and sent this to the GPS. Then set the DGPS source to the WAAS channel. To make sure these settings where applied to the unit.. from the POLL menu select Navigation Parameters the info GPS replies with should confirm the DGPS src is:WAAS and that it is in auto mode. After these steps I return the unit to NMEA protocol. It is important to set the correct NMEA messages for this unit. The only changes to what is set in the SIRFdemo.exe is that VTG need changing from 0 to 1 and the baud rate from 4800 to the correct 38400. Leave the rest of values as is and click send. The GPS will resume output in NMEA protocol.

 

As for checking if we are now using WAAS...

 

The GGA sentence is the one to watch (breakdown of the NMEA protocol can be found elsewhere on the interweb but a key to the GGA sentence is pasted below).

 

This an example GGA output message when not using WAAS:

$GPGGA,235205.000,5118.8304,N,00212.8319,W,1,09,0.9,60.1,M,48.5,M,,0000*73

 

That message was logged after WAAS was turned on, but shortly after a hot re-start of the GPS, and the EGNOS bird (PRN 120 in this case) had yet to kick in.

 

When WAAS/EGNOS kicks in the NMEA GGA output changes to...

 

$GPGGA,235411.000,5118.8306,N,00212.8312,W,2,09,0.9,60.7,M,48.5,M,0.8,0000*5A

 

The difference between the two messages... a "1" changes to a "2" and an number (currently at 0.8 in my example) is now included in the message.

 

The "2" is the "GPS quality indicator" part of the GGA sentence. A 2 means a differential fix (i.e WAAS) as opposed to a 1 which is a plain jane GPS fix.

 

The other value is the "Age in seconds since last update from differential reference station" in other words how long it been since the WAAS bird did its thang.

 

The "Diff. reference station ID#" part of the sentence will not change as this is reserved for the Earth based radio Differential Correction Beacons and not WAAS/EGNOS sats.

 

    $GPGGA,hhmmss.ss,llll.ll,a,yyyyy.yy,a,x,xx,x.x,x.x,M,x.x,M,x.x,xxxx*hh

 

    GGA = Global Positioning System Fix Data

 

    1 = UTC of Position

    2 = Latitude

    3 = N or S

    4 = Longitude

    5 = E or W

    6 = GPS quality indicator (0=invalid; 1=GPS fix; 2=Diff. GPS fix)

    7 = Number of satellites in use [not those in view]

    8 = Horizontal dilution of position

    9 = Antenna altitude above/below mean sea level (geoid)

    10 = Meters (Antenna height unit)

    11 = Geoidal separation (Diff. between WGS-84 earth ellipsoid and

    mean sea level. -=geoid is below WGS-84 ellipsoid)

    12 = Meters (Units of geoidal separation)

    13 = Age in seconds since last update from diff. reference station

    14 = Diff. reference station ID#

    15 = Checksum

 

Edited July 12, 2005 by stonefisk

[Jakobs team]

Thanks for all the info, I'm still getting used to the GPS and certainly enjoying it, I love geocaching I just wish I lived in England has there are not enough caches down here allthough I'll be going to Bretton in the autum so I hope to find a few there.

Jakobs team.

[JackiePenn]

Going back to the pocketpc software... I also like to know if WAAS/EGNOS corrections are being used, on a conventional waas enabled unit the user will usualy be presented with a message on the GUI such as "WAAS" to signify this. I have yet to find any of the pocketpc gps or navigation software that states the status of WAAS.

Doh!

 

My bad.

 

BeelineGPS the software I have installed as alternative to gpstuner does in fact declare this info. A previous unnoticed status bar (I had it turn off) states what mode the GPS is in, changing from "GPS SPS MODE" to "DIFFERENTIAL GPS" also the status bar includes a graphic of a small light which changes to blue when using WAAS/EGNOS.

 

Edited July 13, 2005 by stonefisk