Jump to content

Accurracy with WAAS only +- 20 feet?


rjhollan

Recommended Posts

I've got a Garmin 76CSX and even with WAAS enabled, standing out in a field with clear blue sky all around me in all directions, I can only get accurrate readings to within +-10 feet. I've seen several micro geocache listings where the poster claims 5' or 6' accuracy with their eTrex Legend....

 

Does anyone have any tips for getting my unit to be more accurate? I've tried with WAAS both enabled and disabled, and get pretty much the same result -- not really noticible one way or the other.

 

Thanks!

Rich

Link to comment

Does the little "D" appear on the bars of the satellite page? If the D's arent there then even though WAAS is enabled the WAAS correction data hasn't been applied/received.

 

On my Venture CX with WAAS enabled (and with the Ds displayed), my unit regularly displays an accuracy of +-1meter (3feet)

Edited by passdump
Link to comment

Want deadly, pin-point accuracy in your GPSr?

Simply enlist in the the U.S. Armed Forces.

 

....and pray you are lucky enough to actually get issued a receiver. If you do, then pray it is not outdated and actually in working order. The overwhelming majority of people I served with provided their own commercial units. Don't even get me started on body armor.

 

Seriously, though, 10 ft is phenomenal. Be happy.

Edited by astraltourist
Link to comment

Want deadly, pin-point accuracy in your GPSr?

Simply enlist in the the U.S. Armed Forces.

 

Actually, I have a PLGR unit... my understanding was that the only difference between these and the commercial units are that the PLGR is (1) ruggedized, (2) able to get 3m accuracy even with SA enabled, and (3) able to operate in a jamming environment.

 

I got my Garmin because it's smaller & lighter than the PLGR, SA is disabled, and if someone wants to Jam my GPS signal so I don't find my cache, well I guess that's okay. :(

 

That does make me think though... maybe I should take it out with my Garmin side by side and do a comparison....

 

I fat-fingered my original post -- I meant 20 feet, not 10. I'll walk up to a cache with my GPSR saying "2 feet" and pointing me at something, then flip over to the satellite page to see the accuracy at "+-20" or higher. I just figured I had something misconfigured. I'll look for the "D" next to the satellites next time I'm out...

Link to comment

From what I've learned here and from browsing the web, the civilian GPS system, even with high-sensitivity and WAAS, has a maximum accuracy of 15' - 20'. Sometimes, I'll find a cache spot-on to the posted coordinates. I consider that to be more luck than anything else, because most of the time, I'm within 20', and occasionally, I'm only within 40'. So, given two different GPSr's (the one used to place the cache, and mine, used to find it) with 20' accuracy, that makes sense. Statistically, it's the sort of distribution you'd expect to see.

 

 

I originally had an eTrex Vista that regularly reported accuracy of 6'. But with that unit, under a clear sky, I usually got within 20' or so--just like with my 60 CSx, which rarely reports accuracy better than 15'. I figure the 60 CSx is simply more accurate in its reporting of...accuracy.

Link to comment

The GPSr displayed accuracy is just one estimate based on the algorithm used in your GPSr. Comparing one unit's measure to another is probably pointless unless you know exactly how it is calculated. Even then, the displayed accuracy still may not be very useful to you.

 

A more useful (in my opinion) measure of the accuracy can be obtained by simply plotting tracks over calibrated maps. Don't forget to configure your GPSr to gather track points however you feel is most useful to you. If your GPSr is not moving, the points should be very tightly packed together with no spurious points away from your location. Aerial photo maps with known benchmarks are also handy for checking real world accuracy.

 

I think the best accuracy any of my GPSrs has displayed was +/-7 or 8 feet. I have no idea if Garmin displays the 95% accuracy or something else. Both my Vista C & Vista HCx will display the same accuracy, but the HCx collects track points that are quite a bit more accurate when not moving. When moving they are very similar except that the HCx does better in difficult terrain (as advertised). Both have been worse with WAAS enabled lately.

Link to comment

There's an interesting in depth article on GPSr accuracy here.

 

Good article; I found this one helpful as well. I was able to get to +-13 feet last night while paying attention to where I pointed the antenna with WAAS enabled; I think that's reporting a bit high though, because the GPSR put me within 4 feet of the cache while it said +- 13. Maybe it's a "glass half empty" kinda algorithm. :mad:

 

Thanks for the suggestions everyone!

Link to comment

If you want to check out the accuracy of a unit, one easy way to do so is to turn track log on and walk some trails. Do this several times over several days. You will see that each time you travel a trail, a track log is written and they will not be identical. There will be a dispersion of tracks that varies, depending on time of day (the satellites are not stationary), tree cover, and other variables.

 

I have had three 60Cx units, and they all have performed similar to each other. On same days my tracks will be within a few feet of reality, but on others they may be 50' or more off. That's just the way it is. Don't trust the accuracy estimate: find out for yourself.

Link to comment

In all honesty, I think +/- 10 ft is pretty good. I've used a variety of survey grade Trimble and Thales units and on the fly, they really don't do much better than a significantly cheaper Garmin. With our Trimbles, under ideal circumstances we can manage to pull off 3 - 5 ft accuracy, which generally post processes to about 30 to 50 cm.

 

In the next 5 - 10 years, it will be interesting to see how the addition of the L5 to GPS, as well as fully operational GLONAS and Galileo constellations will impact the consumer market. It wont be long until survey grade GNSS receivers that can perform real time differential correction will be able to provide sub meter accuracy on the fly.

 

The big question will be how long will it take for Garmin to provide a $300 GNSS unit that can actually take advantage of the different systems (esp. for modeling ionospheric distortion). Even if real time < 1 m accuracy is a ways away for the consumer market, at the very least it will be nice to take advantage of a potential constellation of 100+ satellites.

 

In the mean time, an experiment I've been wanting to do for awhile is to take the same point with the various GPS units I have laying around, and then compare the accuracy against post-processed data taken with our GeoXH at work.

Link to comment

Want deadly, pin-point accuracy in your GPSr?

Simply enlist in the the U.S. Armed Forces.

 

Actually, I have a PLGR unit... my understanding was that the only difference between these and the commercial units are that the PLGR is (1) ruggedized, (2) able to get 3m accuracy even with SA enabled, and (3) able to operate in a jamming environment.

 

I got my Garmin because it's smaller & lighter than the PLGR, SA is disabled, and if someone wants to Jam my GPS signal so I don't find my cache, well I guess that's okay. :grin:

 

That does make me think though... maybe I should take it out with my Garmin side by side and do a comparison....

 

I fat-fingered my original post -- I meant 20 feet, not 10. I'll walk up to a cache with my GPSR saying "2 feet" and pointing me at something, then flip over to the satellite page to see the accuracy at "+-20" or higher. I just figured I had something misconfigured. I'll look for the "D" next to the satellites next time I'm out...

 

You have the PLGR down pat. The added accuracy of the mil units was from not being subject to SA. Once that was turned off, all GPS receivers are on equal footing... at least until the next generation of mil GPS comes on line :^)

 

I have always thought it was interesting that most mil users go with the Garmin. I expect it is just a matter of using what the guys around you are using. Going with something different is of no point when the standard works just fine. I actually worked for Thales (not in the GPS division) and did some work with GPS. We took a commercial GPS receiver module an put it into a remote control of a piece of gear. Remote as in on a cable, not on an RF link. We put the GPS receiver in the remote so it could get a clear view of the sky when worn on the chest or the shoulder. It worked pretty well and as long as SA is not turned on the only disadvantage compared to the PLGR is the lack of anti-spoof and anti-jam.

Link to comment

If you want to check out the accuracy of a unit, one easy way to do so is to turn track log on and walk some trails. Do this several times over several days. You will see that each time you travel a trail, a track log is written and they will not be identical. There will be a dispersion of tracks that varies, depending on time of day (the satellites are not stationary), tree cover, and other variables.

 

I have had three 60Cx units, and they all have performed similar to each other. On same days my tracks will be within a few feet of reality, but on others they may be 50' or more off. That's just the way it is. Don't trust the accuracy estimate: find out for yourself.

 

Better yet, put the receiver into NMEA mode and record the readings over a long period. I have done that with my bluetooth receiver and found that the raw, un-averaged data wanders within (if you are lucky) a 50 m circle. Of course the 95% circle is much smaller than that, but it shows that even under static conditions, a receiver will get large errors at some time. That is why a hand held receiver averages the raw data to display to the user. You will see this when you walk or drive and it takes the cursor a second or two to catch up when you stop. In fact, I expect that is the source of the "overshoot" that the Magellans are so famous for.

 

I have found that most of the larger error readings happen in a "burst" where the position wanders away from the current spot for up to 10 seconds and then returns to the smaller error. I can't say if this is something external or internal to the unit being tested. I have not tested a lot of units.

 

Also, a note about WAAS. While trying to get diff GPS mode turned on in my bluetooth unit, I had a number of conversations on one of the GPS forums. One was with a representative of one of the bluetooth GPS makers and he indicated that WAAS does *not* give a significantly more accurate position. The improvement is on the order of 1 to 2 meters. In the spec sheet this can be up to 50% of the non WAAS accuracy. But that is under very ideal conditions only seen in the lab. In the real world (even standing in an open field with 12 SVs in perfect locations) there are significant sources of error and you will only get 10 to 15 meters regardless of what the receiver tells you.

 

The EPE reported by a GPS receiver is the "estimated" position error and is calculated from the location and number of SVs. ***There is NO WAY the receiver can KNOW the actual error*** If that were possible, it could be subtracted out of the calculation and corrected. This number is just an indication of how the constellation will degrade the accuracy and tells you nothing about the other sources of error.

Link to comment

Also, a note about WAAS.

 

I almost forgot, the purpose of WAAS is to provide an indication of whether your calculated position is within an appropriate error range or not. This is used in aircraft so that an erroneous reading is not used to guide the aircraft on landing where even an excessive error can cause a problem. I don't know exactly how WAAS provides this safety check, but that is its purpose, not greatly improved accuracy.

Link to comment

While some small handheld gps units might report they're getting 3ft or less, most likely they're just mis-reporting that information. The only real way to have that good of an accuracy is to have either an array of gps receivers or a differential gps setup. Both of these would be too costly and large for geocaching. 10ft is just fine to find any cache.

Link to comment

"While some small handheld gps units might report they're getting 3ft or less, most likely they're just mis-reporting that information."

 

I would correct that by striking out the words "most likely". Any consumer grade GPS reporting 3' accuracy is mis-reporting. Period.

Link to comment

Turn of Waas and save battery.

 

There are no use for Waas, since it does not give you any more accuracy on the ground.

Waas was primary intended for use with airplane.

 

Reflection from trees, buildings and other objects does even make more problems.

 

I have use Garmin GPS for 5 years and Waas have not given my anything.

It’s a marketing hype.

 

Do a search for waas here on the forum.

 

"accurate navigation system developed for civil aviation by the Federal Aviation Administration (FAA),"

http://en.wikipedia.org/wiki/Wide_Area_Augmentation_System

Link to comment

Turn of Waas and save battery.

 

There are no use for Waas, since it does not give you any more accuracy on the ground.

Waas was primary intended for use with airplane.

 

Reflection from trees, buildings and other objects does even make more problems.

 

I have use Garmin GPS for 5 years and Waas have not given my anything.

It’s a marketing hype.

 

I would say it is more user hype rather than marketing hype. I have never seen a manufacturer claim that it makes a significant difference. The data sheets all show between 1 and 2 meters improvement which, as you say, in the midst of larger, more obvious errors, is not an improvement at all really.

Link to comment

Turn of Waas and save battery.

 

There are no use for Waas, since it does not give you any more accuracy on the ground.

Waas was primary intended for use with airplane.

 

Reflection from trees, buildings and other objects does even make more problems.

 

I have use Garmin GPS for 5 years and Waas have not given my anything.

It’s a marketing hype.

 

I would say it is more user hype rather than marketing hype. I have never seen a manufacturer claim that it makes a significant difference. The data sheets all show between 1 and 2 meters improvement which, as you say, in the midst of larger, more obvious errors, is not an improvement at all really.

 

From Q&A on faa.gov:

 

Q. Can I use WAAS in my vehicle?

 

A. WAAS is extremely useful for vehicle navigation, as it increases accuracy from 10-12 meters with GPS alone, to 1-2 meters horizontal. This increase in accuracy can mean knowing which side of the highway a vehicle is on, lane determination,specific vehicle location, or where an exit is early enough so that you can make the turn before the exit is passed.

Link to comment

 

The EPE reported by a GPS receiver is the "estimated" position error and is calculated from the location and number of SVs. ***There is NO WAY the receiver can KNOW the actual error*** If that were possible, it could be subtracted out of the calculation and corrected. This number is just an indication of how the constellation will degrade the accuracy and tells you nothing about the other sources of error.

 

On this you are absolutely 100% correct!

Link to comment

Also, a note about WAAS.

 

I almost forgot, the purpose of WAAS is to provide an indication of whether your calculated position is within an appropriate error range or not. This is used in aircraft so that an erroneous reading is not used to guide the aircraft on landing where even an excessive error can cause a problem. I don't know exactly how WAAS provides this safety check, but that is its purpose, not greatly improved accuracy.

 

Incorrect. It is both.

Link to comment

From Q&A on faa.gov:

 

Q. Can I use WAAS in my vehicle?

 

A. WAAS is extremely useful for vehicle navigation, as it increases accuracy from 10-12 meters with GPS alone, to 1-2 meters horizontal. This increase in accuracy can mean knowing which side of the highway a vehicle is on, lane determination,specific vehicle location, or where an exit is early enough so that you can make the turn before the exit is passed.

 

This was some years ago.

 

1. Most SiRF and other new chipset do get around 3m without WAAS

 

2. You have to be with in the areas that are covered by WAAS or Egnos. (only a fraction of the Earth)

 

3. The more away from a calibrated ground station (that do send the correction data to the satellites) the less accuracy.

 

4. The more to the north you stay, the less accurate WAAS becomes, since the WAAS birds are located above equator

 

Conclusion

Turn it off and save battery.

 

WAAS reference stations:

waas-coverage-dale.gif

Link to comment

2. You have to be with in the areas that are covered by WAAS or Egnos. (only a fraction of the Earth)

 

WAAS reference stations:

waas-coverage-dale.gif

 

So if you live in the US turn WAAS on, I have never had any issues with receiving waas signals with any of my gpsr anywhere in the US. There is definitely an improvement in location stability and accuracy when receiving a WAAS signal. Suprise!!! That is exactly what WAAS is suppose to do.

I will agree that receivers with a sirfIII chipset do get very good claimed accuracy without WAAS, but curiously as soon as the receiver gets the WAAS signals the units claimed accuracy actually goes up a bit. I assume this is because the unit was falsely reporting it's position error, and test I've done with and without WAAS at known benchmarks, this seems to be the case.

 

The battery savings you get when turning WAAS off is insignificant, a couple hours at best(if that, its probably less than an hour). you would be better off using less back lighting if battery power is an issue.

Link to comment

 

So if you live in the US turn WAAS on, I have never had any issues with receiving waas ......

 

 

I agree. Also, the ' fraction of the Earth' thats covered by WAAS/EGNOS covers about 700 million people, possibly more (Europe & US). And I would imagine that's where quite a chunk of the worlds geochachers reside.

 

If you live in Europe switch it on too. :o

Link to comment

 

3. The more away from a calibrated ground station (that do send the correction data to the satellites) the less accuracy.

 

 

gpsinformation.net:

"Ionospheric Corrections:

The IONO information transmitted by the WAAS system is much more accurate than the basic GPS IONO model. Also, the WAAS system will generally be more accurate than beacon based DGPS because of the way the corrections are rendered by the WAAS system and applied by the GPS receiver. The primary factor is spatial decorrelation, which is the degradation of corrections due to separation from the reference station. RTCM based DGPS corrections suffer from spatial decorrelation, but WAAS corrections do not.

 

This Iono data (and other corrections) are constantly uploaded to the Geo Sats for re-transmission to GPS navigation receivers. There is no interpolation between ground stations by the receiver. This is because the WAAS master system computes a "grid of Iono corrections" which are location dependent based on the user's position. There is an interpolation/extrapolation process to determine the iono correction, but it is not specifically related to the location of ground stations that collect the information. The Iono-corrections grid offered by WAAS are interpolated and applied by the receiver.

 

GPS receivers must then apply the data for corrections appropriate at their location."

 

A model is built from the data. That grid no longer has anything to do with ground stations nor there locale.

Edited by EraSeek
Link to comment

Think of it this way. The stations collect info on delay of signal from an active (or less active) ionosphere. That data is sent to a master station that produces a modeled grid. It fills in the blanks. That grid of corrections is sent via the WAAS sats to your GPS. Your GPS know where it is on the grid and applies corrections based of the closest grid points. There are time and places where little corrections are needed, and times when more correction is needed. Later in the day generally, and further south, where the sun (solar wind) hits most directly. Think of the following graph in two ways. First it is a model of the ionosphere though a normal day. Secondly you may also view it as the grid of corrections, and your GPS is near one of the grid crossover points. Those are the corrections your GPS knows to use.

 

iono-day.gif

Link to comment

1. Most SiRF and other new chipset do get around 3m without WAAS

 

 

Specs on Garmin 60cx with SiRF chip:

GPS accuracy <10 meters (33 feet) 95% typical

DGPS (waas accuracy in north america) 3-5 meters (10-16 feet) 95% typical

 

Garmin Map76 NON-SiRF

GPS accuracy: <15 meters (49feet) RMS 95% typical

DGPS (USCG)3-5 meters (10-16 feet) 95% typical

DGPS (WAAS) 3 meters (10ft) 95% typical with DGPS corrections

 

The SiRF chip does some interesting things. It is not necessarily more accurate. It grabs every availible signal, rejecting none as others will, and computes the heck out of it for the best position. This is why you can get a fix in the woods. At times, such as on start-up, it will make assumptions to provide you with a quick fix and position, as I understand it.

Link to comment

The accuracy on my GPSMAP 60csx runs in the +/-20s most of the time.

 

My 60cx does as well. It almost always stays in that area. As stated above the accuracy reading is not all that accurate. I liked my old Map76 because the accuracy reading moved around more giving me a better relative idea of actual accuracy. The reading on the 60cx is never very helpful.

Link to comment

EraSeek has got it all right. If you read the thread above, forget what jotne said. One thing I will add from personal experience (several Garmins & Magellans & the new Delorme PN-20): If you have been out in the open getting good WAAS the day before, and then today you are in canyons (real or man-made), and can't get WAAS (they can easily be screened, especially if you are a bit north), BE CAREFUL! Best turn off the WAAS if you want accuracy (like for placing a cache). Your unit likely will use the ionospheric & other corrections from the day before, and they could be quite different than today. I have asked Garmin & Magellan tech support about this, with no good answer. WAAS corrections should be set to expire, but do not seem to be in consumer units, in my experience.

Link to comment

Great information.

 

When I place a cache I take a reading using my Legend, WAAS enabled. I'll do a few GOTO readings and if it seems on target that's what I'll submit.

 

What you have just done is to verify that the same receiver on the same day will find the point it just measured. That is not the same thing as getting the actual coordinates more accurately. If you want to measure the location more accurately, you need to measure it many times over multiple days. There are a lot of different sources of error in a GPS measurement and to try to measure around them you need to average a lot of data over a long period of time.

 

For the purpose of Geocaching, there is not much point. It is better to just accept that the error can be larger than you might like and live with that. But then maybe that is what you are saying...

Link to comment

1. Most SiRF and other new chipset do get around 3m without WAAS

 

 

Specs on Garmin 60cx with SiRF chip:

GPS accuracy <10 meters (33 feet) 95% typical

DGPS (waas accuracy in north america) 3-5 meters (10-16 feet) 95% typical

 

Garmin Map76 NON-SiRF

GPS accuracy: <15 meters (49feet) RMS 95% typical

DGPS (USCG)3-5 meters (10-16 feet) 95% typical

DGPS (WAAS) 3 meters (10ft) 95% typical with DGPS corrections

 

The SiRF chip does some interesting things. It is not necessarily more accurate. It grabs every availible signal, rejecting none as others will, and computes the heck out of it for the best position. This is why you can get a fix in the woods. At times, such as on start-up, it will make assumptions to provide you with a quick fix and position, as I understand it.

 

I wonder why the 60cx is spec'd to perform so much more poorly than SIRF rates their own products. Their GSC3f/LP eval kit spec sheet gives an accuracy rating of...

 

Horizontal Position Accuracy

Autonomous <2.5 m

SBAS <2.0 m

 

Of course the devil is in the details. There are a lot of conditions that have to be spec'd in order to state an accuracy. I would be curious as to how they perform in the real world. Like I have said in other posts, if you just record all the readings from a unit over a period of time, regardless of whether SBAS is on or off, the measured values range over a 20 to 30 m (or larger) circle with the 95% radius being much smaller, say 10 m. But I have never seen a unit get anywhere near the 3 meters (or less) that they are rated.

 

BTW, how do you know how the SIRF chip works internally that is different from other GPS chips?

Link to comment

they are rated.

 

BTW, how do you know how the SIRF chip works internally that is different from other GPS chips?

 

Gosh, let me see. I know I read a discription of it on the SiRf site in a letter by the founder, Kanwar Chadha, when the chips first came out in Garmins. He spoke about the gazillion calculations they do. There is an artilce on CNET that speaks about it in an interview with him. gpspassion forums has a link to it. GPS world has had an interview with him as well as speaking about the chips. From what I gather the chips are 20 channel chips being used in 12 channel recievers. There were other places but I can't recall right now where I've read about it.

Link to comment

EraSeek has got it all right. If you read the thread above, forget what jotne said. One thing I will add from personal experience (several Garmins & Magellans & the new Delorme PN-20): If you have been out in the open getting good WAAS the day before, and then today you are in canyons (real or man-made), and can't get WAAS (they can easily be screened, especially if you are a bit north), BE CAREFUL! Best turn off the WAAS if you want accuracy (like for placing a cache). Your unit likely will use the ionospheric & other corrections from the day before, and they could be quite different than today. I have asked Garmin & Magellan tech support about this, with no good answer. WAAS corrections should be set to expire, but do not seem to be in consumer units, in my experience.

 

Well I don't think that is quite right, but you may have something. The errors which are corrected through WAAS are sat position errors, clock errors, and seperately Ionospheric errors. A couple of these are somewhat long term, and the clock errors I think are very short term. The Ionospheric errors in a Garmin are used for up to 2 mins after the WAAS sat lock is lost, but Magellens keep them a fair amont longer. But I don't think this could be the problem. In my old map76 I noticed that there seemed to be a benefit to turning off WAAS when I knew I could not get it such as in the woods. My 60cx simply drops it automatically when no locked on to waas. Here is the deal. With some units if you have WAAS on, but have no WAAS reception, then it is possible that your unit which has some basic preinstalled assumed corrections, is now not using those; niether WAAS nor the default corrections, and thus there are not even any minimal corrections, thus a lousey position fix.

 

This is not gospel, just something I recall reading at some point.

Link to comment

Error source Total DGPS

------------------------------------------------------------

Ephemeris data 2.1 0.0

Satellite clock 2.1 0.0

Ionosphere 4.0 0.4

Troposphere 0.7 0.2

Multipath 1.4 1.4

Receiver measurement 0.5 0.5

 

These are the standard error models. There is no way an earthbound chip can correct this or get a better fix without the asistance of being able to analyze signals passing through space as with WAAS or dual band. I suppose you can average the heck out of it for long periods of time and conclude where the most probable position is, but to say you can get a real time fix with less then the above total error, I don't know if that is at all possible.

 

No you don't need WAAS for what we do, but why not use the technology? My trials on benchmarks usually get me about 12' - 15' off the mark without WAAS and 0'- 5' with waas. Just a general recording of the 4 or 5 times of have done it on different days.

 

WAAS does come in handy when you are on the water in pea soup, or in a white out on Mt Rainier. This I know from experience. And again, the corrections needed are based on how active the ionosphere is. Some days little correction is needed, other times more is.

Edited by EraSeek
Link to comment

Error source Total DGPS

------------------------------------------------------------

Ephemeris data 2.1 0.0

Satellite clock 2.1 0.0

Ionosphere 4.0 0.4

Troposphere 0.7 0.2

Multipath 1.4 1.4

Receiver measurement 0.5 0.5

 

These are the standard error models.

 

I assume the source of your info is Sam Wormley's page

 

Sam Wormley's GPS Errors & Estimating Your Receiver's Accuracy

 

The table data is shown here and I apologize for not being able to align the columns. Is there a feature in this forum to do that properly?

 

One-sigma error, m

Error source Bias Random Total DGPS

------------------------------------------------------------

Ephemeris data 2.1 0.0 2.1 0.0

Satellite clock 2.0 0.7 2.1 0.0

Ionosphere 4.0 0.5 4.0 0.4

Troposphere 0.5 0.5 0.7 0.2

Multipath 1.0 1.0 1.4 1.4

Receiver measurement 0.5 0.2 0.5 0.5

------------------------------------------------------------

User equivalent range

error (UERE), rms* 5.1 1.4 5.3 1.6

Filtered UERE, rms 5.1 0.4 5.1 1.5

 

However you omitted the Bias and Random columns which are very important. Sam's page describes these two columns here, "Each error is described as a bias (persistence of minutes or more) and a random effect that is, in effect "white" noise and exhibits little correlation between samples of range." This says to me that the "random" component should not be correctable by a non-real time differential method such as SBAS. However, the Satellite Clock error is shown as being eliminated completely.

 

There are other methods of differential correction than SBAS or WAAS. The table does not specifically talk about SBAS or WAAS, but rather DGPS. I don't see how SBAS can achieve this level of corrections as it is not truly real time in terms of providing unique corrections for each sample. To compensate for "random" errors which do not correlate between different samples, you would have to have a real time correction such as a direct radio link to a differential ground station.

 

Am I misinterpreting the data above?

 

I will say that my measurements have not shown a significant improvement in measurement accuracy when using WAAS, but like yours, they are fairly limited.

Link to comment

Error source Total DGPS

------------------------------------------------------------

Ephemeris data 2.1 0.0

Satellite clock 2.1 0.0

Ionosphere 4.0 0.4

Troposphere 0.7 0.2

Multipath 1.4 1.4

Receiver measurement 0.5 0.5

 

These are the standard error models.

 

I assume the source of your info is Sam Wormley's page

 

Sam Wormley's GPS Errors & Estimating Your Receiver's Accuracy

 

The table data is shown here and I apologize for not being able to align the columns. Is there a feature in this forum to do that properly?

 

One-sigma error, m

Error source Bias Random Total DGPS

------------------------------------------------------------

Ephemeris data 2.1 0.0 2.1 0.0

Satellite clock 2.0 0.7 2.1 0.0

Ionosphere 4.0 0.5 4.0 0.4

Troposphere 0.5 0.5 0.7 0.2

Multipath 1.0 1.0 1.4 1.4

Receiver measurement 0.5 0.2 0.5 0.5

------------------------------------------------------------

User equivalent range

error (UERE), rms* 5.1 1.4 5.3 1.6

Filtered UERE, rms 5.1 0.4 5.1 1.5

 

However you omitted the Bias and Random columns which are very important. Sam's page describes these two columns here, "Each error is described as a bias (persistence of minutes or more) and a random effect that is, in effect "white" noise and exhibits little correlation between samples of range." This says to me that the "random" component should not be correctable by a non-real time differential method such as SBAS. However, the Satellite Clock error is shown as being eliminated completely.

 

There are other methods of differential correction than SBAS or WAAS. The table does not specifically talk about SBAS or WAAS, but rather DGPS. I don't see how SBAS can achieve this level of corrections as it is not truly real time in terms of providing unique corrections for each sample. To compensate for "random" errors which do not correlate between different samples, you would have to have a real time correction such as a direct radio link to a differential ground station.

 

Am I misinterpreting the data above?

 

I will say that my measurements have not shown a significant improvement in measurement accuracy when using WAAS, but like yours, they are fairly limited.

 

Yeah, I tried to clean up the tables because the forums weren't letting me line things up, just like you.

As far as your questions about random and bias, I really don't know. I've never delved into that, and don't understand it much yet, but I'm willing to learn.

Link to comment

So, lots of fuss about a small improvement.

 

I do only see WAAS as a sales argument, ok to have, but not needed.

 

On Nüvi and other car GPS (that do support WAAS) you cannot see if you get it or not.

There is no "D" in bar like 60 series.

 

Early on WAAS was discribed as buoy bumping accuracy. That is estentially the best way to discribe it. When in thick fog on the water, with WAAS you'll find that bouy the first time, without WAAS it may take 4 or 5 passes. A lot of fuss about a bit of accuracy, but sometimes that bit of accuracy can save a life. But you have my permission not to use it being so dead set against it :)

 

WAASaccuracy2.gif

Edited by EraSeek
Link to comment

So, lots of fuss about a small improvement.

 

I do only see WAAS as a sales argument, ok to have, but not needed.

 

On Nüvi and other car GPS (that do support WAAS) you cannot see if you get it or not.

There is no "D" in bar like 60 series.

 

Early on WAAS was described as buoy bumping accuracy. That is essentially the best way to describe it. When in thick fog on the water, with WAAS you'll find that buoy the first time, without WAAS it may take 4 or 5 passes. A lot of fuss about a bit of accuracy, but sometimes that bit of accuracy can save a life. But you have my permission not to use it being so dead set against it :D

 

WAASaccuracy2.gif

 

The only trouble with what you say is that it is not really quantitative. Your description is essentially "sales" talk as Jotne is saying. The diagram you provide is not factually accurate. I can assure you that you do not get a 5x improvement in accuracy as is shown (15 m >>> 3 m). I get 3 meter accuracy (50% CEP) without WAAS and I don't see any measurable improvement with WAAS working. Where did you get this diagram?

 

Certainly there is little down side to using WAAS, but it is silly to expect it to make a difference in anything you do in normal usage. If I were trying to "bump a buoy" in dense fog and my life depended on it, well, I wouldn't want to be counting on a commercial GPS receiver anyway!!!

 

I'm curious, why would you need to "bump" a buoy to save your life? Why would you need to be that accurate if you were in open water? If you were navigating, I don't think you need that sort of accuracy and if you were trying to maintain a position, why would you need the buoy rather than just the GPS?

Link to comment

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.
Note: Your post will require moderator approval before it will be visible.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...