| Versions |
  |
|
| Author |
 Topic  |
|
beechfarm
United Kingdom
15 Posts |
 Posted - 13 mars 2006 : 09:43:41
|
I am testing my new Holux GPSlim 236 (SiRFStar III + GSW 3.1.1) linked to SiRFDemo (v3.81) by
Bluetooth.
I can "see" the 3 EGNOS satellites (PRNs 120, 124 and 126) with signal strengths in the range 28 to 38 db-Hz and I specify each of them in turn. My excitement rises as I watch the dark blue icon on the 'Radar View' turn to light blue. This must mean my receiver is reading intelligible signals.
My excitement turns to disappointment when I open the 'DGPS View' dialog box. The only field containing new information tells me that DGPS source is SBAS. There is no new information in any of the 12 records of the 'GPS corrections' record set. The 5 fields of the 'SBAS' record are blank (So are the fields of the Internal Beacon record).
My spirits drop lower as I watch the scatter pattern develop in 'Map View', showing the apparent wanderings of my stationary receiver. These finally overflow a 15m radius circle, indicating no improvement over the non-SBAS situation.
I check the EGNOS website. The Dutch ground station is currently using PRN 126 for SBAS correction and its scatter pattern for the day would easily fit in 5m dia circle. I try PRN 126 again but nothing changes and I still have a light blue icon.
Is there a hardware problem or am I misunderstanding the software?
Can the Gurus of GpsPasSion help please or should I start a dialogue with the receiver makers? |
|
|
Ads
|
|
|
gpspassion
93389 Posts |
Posted - 13 mars 2006 : 11:26:21
|
1. Yes that light blue color means you have an SBAS fix
2. Yes, SiRFdemo (at least the current versions) don't show any activity in the DGPS screens with SiRFIII like it does with SiRFII
3. As for the actual impact of EGNOS, what you're seeing is what I've been seeing for the past 3+ years. I've done side by side testing, EGNOS/NO-EGNOS and have seen zero impact over 12+ hours.
The bottom line is that SBAS was not really designed to improve accuracy see here http://www.gpspassion.com/fr/articles.asp?id=143&page=2 and while it does (maybe 1/1.5 meters) I've come to the conclusion that it is not measurable with consumer equipment.
_________________________________________________________________________
Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer? |
 |
|
|
beechfarm
United Kingdom
15 Posts |
Posted - 13 mars 2006 : 21:58:10
|
Thank you for this information which is very interesting but disappointing!
SiRF is misleading potential customers if their device only accesses the SBAS data but does not apply the correction when calculating the receiver position. (Their website says that the SiRFStar III provides 'SBAS (WAAS and EGNOS) support'.)
The Holux website (http://en.holux.com.cn/product/search.htm?filename=gpsreceiver_bluetooth_gpslim236_gg_wzhjzhd.htm&target=gpsreceiver00&level=grandsonson) describes the the GPSlim 236's accuracy as follows: -
Non DGPS (Differential GPS)
@ Position : 5 - 25 m CEP without SA
@ Velocity : 0.1m / sec
@ Time : 1 ƒÊsec sync GPS time
EGNOS/WAAS /Beacon
@ Position : < 2.2 m, horizontal 95 % of time
@ < 5 m, vertical 95 % of time
Your observations (and mine) suggest to me that Holux is creating unobtainable expectations in the minds of potential customers. I certainly feel mislead.
If I have misunderstood the situation I would welcome clarification from SiRF or Holux. |
|
|
|
gpspassion
93389 Posts |
Posted - 14 mars 2006 : 01:53:52
|
Non don't get me wrong, SiRFstarIII DOES apply SBAS corrections, hence the change in colour and the 2 flag in the GGA message, it's just that SiRFdemo does not show them anymore as it did with SiRFstarII, not sure why. My observervations of the lack of impact of EGNOS were made both with SiRFII and SiRFIII receivers.
As for the Holux numbers, maybe they can be achieved in an ideal environment with no obstruction at at all, I haven't been able to replicate that. What is confusing I think is that they should show the same range as for non-GPS, SBAS offering a 1.5/2 meter improvement, it doesn't magically give a 2.2m accuracy regardless of the environment!
_________________________________________________________________________
Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer? |
|
|
|
saimhe
Lithuania
134 Posts |
Posted - 14 mars 2006 : 13:47:04
|
quote:
Originally posted by gpspassion
SiRFstarIII DOES apply SBAS corrections, hence the change in colour and the 2 flag in the GGA message, it's just that SiRFdemo does not show them anymore as it did with SiRFstarII, not sure why.
Therefore, the Radar View shows uncorrected position? But position in the GGA message perhaps shows some improvement -- or also does not? |
|
|
|
gpspassion
93389 Posts |
Posted - 14 mars 2006 : 14:12:00
|
No, I would think the radar view shows corrected positions, I don't think there is a simultaneous dual output, SBAS or NON-SBAS, if it's on all the data coming out of the GPS is corrected.
_________________________________________________________________________
Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer? |
|
|
|
Carl@SiRF
USA
158 Posts |
Posted - 20 mars 2006 : 17:47:16
|
Hope I can shed some light here for people. When SiRFstarIII receivers say they are applying SBAS corrections (when the navigation mode reports DGPS used, either in binary message 2 or NMEA GGA message) it is using the corrections. However, unlike SiRFstarII receivers, it does not output the actual corrections.
The impact of using the corrections is rather small. If you could connect your receiver to an antenna located on top of one of the tallest buildings around (to reduce multipath), and if this was in a location with minimal jamming signals, and if you collected data over a 24-hour period (to average out the variations due to the changing satellite constellation), you would normally see a wander pattern with an average, or 50% probability) of below 2.5 to 3 m. Repeat that test and you would probably see 50% probability drop to maybe 1 or 1.5 m, for a 1 to 2 m improvement. The improvement is due to better satellite clock corrections, and ionospheric correction improvement. Not a lot to write home about, but that isn't SBAS system's primary goal -- integrity monitoring is where SBAS systems shine. An aircraft on final approach will be warned much sooner if one of the satellites he is using should go bad, so that that satellite can be dropped from the solution before the error causes a position fix that would make the pilot correct for an non-existent altitude change.
Now, take your same receiver, and instead of an ideal antenna location, put it in a car driving around among all sorts of reflectors and jammers, or carry it on your person in the same world, and that average error blossoms out to 10 or 20 m due to local issues in your environment. Turn on SBAS and you still get 1.5 to 2 m improvement, and among the noise of real-life situations, you probably won't see it. Remember, any type of DGPS can only correct for errors that are common between the reference receiver and the receiver using the corrections. It cannot correct for factors unique to the remote receiver.
Carl - SiRF Customer Support |
|
|
|
paulkbiba
USA
5064 Posts |
Posted - 20 mars 2006 : 19:10:52
|
Excellent explanation. Thank you very much. Maybe this will help put some of the "WAAS Fever" to bed!
Moderator
Don't forget the GPSPassion Club! |
|
|
|
gpspassion
93389 Posts |
Posted - 21 mars 2006 : 14:39:43
|
Yes indeed and I will link it from the current write-up here and I'm planning a summary article with real case scenarios.
Just to close the loop on the original question, any reason why SiRFstarIII is not outputting the corrections or would a new version of SiRFdemo be required to view them ?
_________________________________________________________________________
Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer? |
|
|
|
beechfarm
United Kingdom
15 Posts |
Posted - 29 mars 2006 : 12:26:29
|
Thank you, Carl, for your excelent explanation. You are absolutely right. When my GPS receiver receiver is fitted with an antenna and hauled to the top of the flagpole the 'wander' pattern is much smaller and all the points logged during the battery life lie within a 10m dia circle. (I have not done the analysis but I guess that 95% of the points lie in 5m dia circle.)
I have been looking for a copy of SirfView to help me with the analysis. Is it available for download or is it restricted to EVK subscribers?
Thanks also to GPSPassion |
|
|
|
saimhe
Lithuania
134 Posts |
Posted - 29 mars 2006 : 23:30:20
|
| Carl, I believe you are the one who at last can give a sound confirmation whether SBAS PRNs are reported in NMEA GSV message, and under which conditions. For example, does it [accidentally] depend on hardware, on firmware (except obvious cases like XTrac), must SBAS be enabled or not, etc. |
|
|
|
musky
USA
226 Posts |
Posted - 13 avr. 2007 : 14:25:41
|
SirfIII units are being marketed left and right with WAAS support. If you go to any of the retailers, they all advertise in big letters WAAS ENABLED. However, it is a crap shoot if you will actually get WAAS support. In my opinion this is called false advertising. The explanation that the manufacturer can just "compile" the code out seems really lame to me.
So let's do a little thought experiment here, OK? I have a GPS company and I get SirfIII chips for my unit, and the firmware source code, with the option to "compile out" WAAS functionality. I go, "gee, hmmm, I can turn WAAS on or off, I think I'd like it better with WAAS off, BUT, I'm going to lie to my marketing department and say that WAAS is really on". Or how about this, I go, "gee, hmmm, I can turn WAAS on or off, I think I'd like it better with WAAS off, BUT, I'm not going to tell my marketing department". Or I turn WAAS off and tell the marketing dept, but they intentionally lie in the ads. You pick the scenario, they are all absurd in my opinion.
Then when people find out WAAS isn't really enabled, I say, "Well you really don't need WAAS anyway!"
How about an analogy here. I agree to buy a car that is advertised with power windows. I pay for the car, take delivery, and realize the car doesn't have power windows, it has hand cranks. So I go back to dealer and say, "Hey, you told me I had power windows, and I don't have them". Dealer tells me, "yeah, but you don't need them, hand crank works just fine". And so I go, "Oh cool, thanks for the explanation!" |
|
|
|
nemogps
USA
31 Posts |
Posted - 10 mai 2007 : 20:29:55
|
In response to Carl's post about the amount of improvement you will see with SBAS, that does not fit my personal experience. I have used handheld GPS receivers and found them to be much less accurate if DGPS is not operating. When I initially turn on my Magellan, it shows an EPE of about 30 feet (10 meters) after about 30 seconds. It only drops to 10 to 15 feet once the WAAS satellite has been received and incorporated in the calculations.
On the other hand, I have checked the data sheets for GPS chip sets from other manufacturers (I can't check the SiRF chip sets since they don't provide them to the public). They provide numbers similar to what Carl indicated.
Even so, the choice of using or not using WAAS is up to the end user. Right now there are a lot of GPS receivers being advertised as using the SiRF III chipsets and supporting WAAS DGPS, but no one seems to be able to get WAAS to work. That is not at all the same as finding out that WAAS DGPS only provides a small improvement.
BTW, how does DGPS help with "integrity monitoring"? I am not sure I see how these "small" corrections would provide a faster detection of a bad satellite.
|
www.gnuarm.com |
|
|
|
musky
USA
226 Posts |
Posted - 10 mai 2007 : 23:41:03
|
Unfortunately, since WAAS doesn't work on SirfIII, I can't choose if I have it or not!
I have been trying to evaluate WAAS vs no-WAAS with my MTK 32 channel, but unfortunately MTK has static navigation, which really messes up any kind of precision study you are trying to do!
I just think Carl's post about "you don't need WAAS" is a copout. There is something else going on here and I wonder what it is. Maybe it's just a case of WAAS not really helping that much for "low end" units (not Trimble for instance). Maybe there is some problem with WAAS kicking in and out and causing jumps in position when it does so. When you turn WAAS on, it requires more power, but that should not be an issue with USB "mouse" devices. Maybe the vendors are all stuck selling the "WAAS is better fantasy" and are afraid to say they don't have it.
I started another thread here, simply asking if ANYBODY ON THE PLANET can direct me to a USB mouse SirfIII unit that really has working WAAS, and still I haven't got an answer. Somebody, give me the model and I will go out and buy a dang 3rd SirfIII unit in my lifetime search for this elusive item. |
|
|
|
nemogps
USA
31 Posts |
Posted - 11 mai 2007 : 05:53:48
|
I finally got WAAS to "work" on my Holux 240 receiver. I had it enabled by using the GPS NMEA Monitor program, but I could not detect that WAAS was working. I got an email from Dennis Gröning that explained in great detail how to do the same thing using SirfDemo. This program displayed all the info from binary mode and I was able to reorient the receiver to pick up a WAAS satellite. It now reports that it is in 3D+DGPS mode although I don't see any specific correction data, but the SiRF rep said this is a change in the rev 3.2 software, right?
|
www.gnuarm.com |
|
|
|
musky
USA
226 Posts |
Posted - 11 mai 2007 : 12:59:27
|
| It may appear to be in WAAS mode (SirfDemo will report it's enabled), but chances are it's not really making any corrections. I call it "Broken WAAS enabled". |
|
|
|
Topic |
|
|
|
| This page was generated in 0,53 seconds. |
 |
|
- 
- 
SS III receives SBAS data but makes no corrections
