Google
  Web www.gpspassion.com


GpsPasSion LIVE!
www.flickr.com
This is a Flickr badge showing public photos from GpsPasSion Live !. Make your own badge here.

www.NaviBlog.com



Versions

Links/Liens




Portal/Portail
Rechercher

- -

Polls/Sondages
Sondage
Pour vous guider sur la Route :
GPS Mobile (SEM)
GPS Intégré
Smartphone
Autre
Voter  -  Résultat des votes
Votes : 2265




Club GpsPasSion
Soutenez le site!

USA: (US$)
EUROPE: (€)
Guide Paypal


GpsPasSion Forums
Home | Profile | Register/Enregist. | Active Topics | Search/Recherche | FAQ
Username:
Password:
Save Password
Forgot your Password?

 All Forums
 Advanced Topics
 General Technical Discussions
 Using GPS systems with >1Hz refresh
 New Topic  Reply/Répondre
 Printer Friendly
Next Page
Author Previous Topic Topic Next Topic
Page: of 4

gpspassion

93993 Posts

Posted - 12 janv. 2007 :  15:13:28  Show Profile  Visit gpspassion's Homepage  Reply with Quote

Test details below

Since this topic has come up a few times, here is thread where we can discuss the systems offering high refresh rates (the "standard" is 1hz or 1/s) and what type of applications they can be used for.

Systems :
- uBlox based : 4Hz - no consumer system
- MTK based : 5Hz (optional) - iBlue 737 Bluetooth (special version)
- Garmin : GPS18 USB GPSr

Applications :
Probably not much point in having more than 1Hz for road guidance, but typically high refresh rates would be useful to measure acceleration or to improve accuracy (by averaging) for precision use like surveying of farming possibly.

Measurements :
1. Side by side testing of 1Hz and 5Hz systems, see below

Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?

Ads


robbiex

USA
218 Posts

Posted - 13 janv. 2007 :  02:03:49  Show Profile  Reply with Quote
There are an increasing number of GPS receivers that have update rates in excess of 1Hz. In addition to those listed above, other examples are: various Thales receivers (5Hz), the Laipac TF50MCX (20Hz) and the OXTS RT3000 series (100Hz).
Rates in excess of 1Hz are required by a variety of industrial and military applications. One interesting example is the measurement of automotive performance as has been developed by MaxQData. See:
www.maxqdata.com.
An update rate faster than 1Hz is not a necessity for automotive navigation, but it does have some advantages. A higher rate helps to decrease the annoying display lag time and it helps to improve tracking and distance measurements especially if one is traveling on a curvy route.
The only disadvantage is that a faster data update results in an increased computer processing load. Rates of 5-10Hz should, however, be well within the capability of current generation personal computers.
The issue of whether or not a GPS navigation program is compatible with update rates other than 1Hz depends on the synchronous or asynchronous nature of the data transfers and the software being able to adjust any time dependent measurements as is appropriate for the different sample rates. It is unknown which, if any, of the current GPS navigation programs can handle data update rates other than 1Hz.



Go to Top of Page

robbiex

USA
218 Posts

Posted - 19 janv. 2007 :  03:31:30  Show Profile  Reply with Quote
(This was moved from the 737 forum)
We now have an iBlue 737 that updates at a confirmed 5Hz. Unfortunately, as expected, the MapPoint software apparently expects a 1Hz data rate rather than making use of the highly accurate time information that comes with each group of GPS information sentences. In any case, the current MapPoint software cannot accept GPS data at 5Hz. I don't know what the situation may be with other GPS navigation software.
I am not happy about the misleading 737 specification sheet that lists the update capability as 1-5Hz. 5Hz is only possible in a special order version. It is currently only available in the USA from MaxQData (www.maxqdata.com) for what I think is a very high price. Nevertheless, this unit operates correctly at 1 or 5Hz update rates. The GPS works well with our navigation software.
Go to Top of Page

robbiex

USA
218 Posts

Posted - 20 janv. 2007 :  02:26:15  Show Profile  Reply with Quote
I am familiar with the specification data for the MTK chip, but don't know what Transystem actually does to make a difference between the 1Hz and the maximum 5Hz update 737s. One would suspect that the only difference is a minor change in the Flash or EEPROM stored code, but it may be more involved than this. I do know that the NMEA sentence that sets the special order unit to 5Hz will not work on the "standard" unit. It is possible to set the 5Hz unit to 1Hz, but no information is available on whether or not you can use a NMEA sentence to set rates of 2,3 or 4Hz.
We find that all of our GPS test programs work with 5Hz as does our program that displays navigation parameters. As noted, MapPoint does not work at 5Hz. I do not know if any of the consumer level GPS navigation programs will work with 5Hz updates. We have DeLorme, Deluo, Ostia, TeleType, and TomTom and could test these packages, but perhaps someone else has already done this and can answer the question about 5Hz compatibility.
I do think that it won't be long before the major vendors support the 5Hz updates.
Go to Top of Page

gpspassion

93993 Posts

Posted - 20 janv. 2007 :  17:52:55  Show Profile  Visit gpspassion's Homepage  Reply with Quote
That might have high CPU requirements on mobile platforms, I wonder if some type of averaging might not be more useful. I've noticed that the NMEA data is similar at 5hz, except that there 5x more of it. The time stamp is broken down in .20, .40, .60, .80, .00 and it seems you have to connect at 128,000bps for all the data to come through.

I've logged data at 5hz and need to analyze it to see how the track compares to that of a 1Hz MTK receiver.

Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?
Go to Top of Page

robbiex

USA
218 Posts

Posted - 21 janv. 2007 :  19:42:44  Show Profile  Reply with Quote
I will certainly look forward to the results of your 5Hz tests.
Some areas of specific interest are reducing display lag (which may require a combination of the 5Hz updates and some sophisticated look ahead mathematics along the lines of that used by TomTom), improved distinction between mobile and immobile situations, and reduction of navigation errors on curvy routes. I agree that some form of averaging will be helpful especially with speed and other navigation parameters.
We have now had a little more experience with the iBlue 737 5Hz GPS receiver and MapPoint 2006.
GPSGate, which we use to create multiple virtual ports which access the data from a single GPS receiver, works fine with a 5Hz update and a serial transfer rate of 57.6Kbps. The various functions of GPSTools also work properly with these data rates. It would be desirable to use the 737's maximum 115.2Kbps serial data rate, but GPSGate, GPSTools, AGPS and MapPoint do not have a selection for this rate.
MapPoint does not track one's position along a route at 5Hz, but, interesting, the turn directions and elapsed distances do display properly albeitly with some jerkiness. The latter is undoubtedly due to MapPoint responding to some, but not all, of the 5Hz GPS data.
We tried AGPS( Advanced GPS, not Assisted GPS). It has some problems with the 5Hz data, principally that the time of day display is jerky whereas the intervals are properly separated with GPSTools.
It does display one's location along a MapPoint route with position updates occurring at close to or actually at 5Hz. Unfortunately, the intersection lag is the same or worse as occurs with MapPoint's regular display with 1Hz. Nevertheless, the trials do demonstrate that there is more than enough processing time available for concurrently running the combination of GPSGate, GPSTools, AGPS and MapPoint and that a current day laptop has no problems supporting 5Hz map updates.


Go to Top of Page

gpspassion

93993 Posts

Posted - 22 janv. 2007 :  01:33:09  Show Profile  Visit gpspassion's Homepage  Reply with Quote
Spent a few hours tonight analyzing these tracks and the 5hz 737 track is really stunning alhtough oddly enough it comes out as 3Hz, possibly because the Bluetooth 115.2kbps rate was not enough. Other systems in the comparison : MTK 1hz Qstarz 818, Globalsat BT-359 SS3 v3.2.2, Garmin GPS10x SS3.

I "slalomed" between high-rise buildings, a typically problematic environment for GPS receivers, but as we've seen before, the MTK show very little multipath "bouncing" unlike the SiRFstarIII receivers that are a good 30+ meters away on the 4th segment. The extra Hz allow for some nice rounding of the curves and generally better accuracy.


Overlaid on an aerial photo


Side by side comparisons


Zooming in to the first turn, the grid is 20 by 30 meters



Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?
Go to Top of Page

robbiex

USA
218 Posts

Posted - 22 janv. 2007 :  02:19:34  Show Profile  Reply with Quote
Thank you very much for the impressive tracks.
57.6Kbps doesn't result in any data loss with GPSTools based on checking the 200ms changes in the time information and some experimentation with MaxQData's automotive performance program. The situation is different with map routing and I don't doubt that at least a 115.2Kbps rate is important. You can set the types of GPS data that are transmitted and it might be helpful to eliminate those sentences that are not necessary for routing. I haven't counted the characters in the current repetitive GPS sentences.
Go to Top of Page

robbiex

USA
218 Posts

Posted - 22 janv. 2007 :  19:08:36  Show Profile  Reply with Quote
Still trying to figure out the apparent 3Hz sample rate.
I ran MaxQData's "flight" with recording of GPS seconds after midnight (updated every 200ms), latitude and longitude. I didn't see any data loss. All the measurements were present at a 5Hz rate.
Go to Top of Page

gpspassion

93993 Posts

Posted - 22 janv. 2007 :  22:51:19  Show Profile  Visit gpspassion's Homepage  Reply with Quote
Yes, the GGA messages seem to have that 200ms timing, maybe my NMEA->PLT track conversion is losing some in the process, odd though. Still the graphs above show what >1hz brings in terms of increased accuracy.

I ran MaxQdata Flight on my 6515 last night, pretty interesting, albeit a bit confusing, the HP measurements or the "triggers". Too bad I don't a have a car that's up to the challenge ;-)

Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?
Go to Top of Page

robbiex

USA
218 Posts

Posted - 23 janv. 2007 :  00:34:17  Show Profile  Reply with Quote
I don't think that my diesel VW Beatle is up to the high performance of the racing cars that MaxQData usually evaluates.
I did learn a little more information. Johan Franson wrote me that GPSGate should work fine at a serial data rate of 115.2Kbps. You have to set the value manually because there is no tab for this data rate. It does work properly.
MaxQData has tabs to set the update rate to: 1,2,3,3.3,4 and 5Hz. I tried 2Hz, but it was still sending updates at 5Hz. MaxQData's site indicates only two possible rates: 1Hz and 5Hz so the tabs probably don't work.
As mentioned, AGPS tracking works with MapPoint, but MapPoint itself does not do any tracking with a 5Hz update rate. Interestingly, it completely ignores 115.2Kbps data which makes me wonder if part of the limitation is the serial data rate. MapPoint normally uses 4.8Kbps which is definitely too slow. There is apparently no way to change this parameter.
Go to Top of Page

igalan

Spain
358 Posts

Posted - 24 janv. 2007 :  09:27:37  Show Profile  Reply with Quote
I always wondered why all GPS where 1 Hz and no one was innovative enough to build one that worked at a higher frequency. Now is this GPS suitable for general use or regular navigation programs will get confused by the increased frequency? It would be nice to have the extra data on turn arounds

Samsung Galaxy SIII | TomTom Carminat
Go to Top of Page

Gu

654 Posts

Posted - 28 févr. 2007 :  18:11:32  Show Profile  Reply with Quote
quote:
Originally posted by gpspassion
I've logged data at 5hz and need to analyze it to see how the track compares to that of a 1Hz MTK receiver.


Hi gpspassion,

I recently modified ChronoGPS to manage (in theory !) 5Hz GPS (decoding the specific time stamp). If you have a 5hz GPS can you test if the program is really able to manage it and send me the xxxx.log.txt file (even 100 m tracking ! just to see if everything is ok). Thank you very much !
Go to Top of Page

gpspassion

93993 Posts

Posted - 01 mars 2007 :  12:08:46  Show Profile  Visit gpspassion's Homepage  Reply with Quote
Ok, I'll download the latest version and give it a try in the next few days ;-)

Update - noticed the max baudrate was 38400, can you increase it to 115200, 5hz needs all the bandwidth it can get !

Discounts and Assistance/Réductions et Assistance (Club GpsPasSion) / Où commencer?
Go to Top of Page

Gu

654 Posts

Posted - 08 mars 2007 :  14:19:54  Show Profile  Reply with Quote
OK. You can enter this speed directly with virtual kerboard. I'll add it in next release.
Did you make any test with ChronoGPS ? Any problem ?
Go to Top of Page

sooby77

87 Posts

Posted - 13 mars 2007 :  20:21:38  Show Profile  Reply with Quote
gpspassion, I've talked to Semsons and they said that they have tested the 5hz versions and it seems that Street and Trips is the one that doesn't work with the 5Hz edition of the i-Blue! That's great news.

In addition, this thread from aximsite supports this as well. From what I read on the thread and also from the person I spoke to from Semsons, the software won't choke, but will be more accurate. The software display refresh is still going to be 1Hz, but it will just display the data (from among the 5Hz) as it is received! This is great since this will take care of the lag problem that is so common with the 1Hz GPS receivers!
------------------------------------------
Go to Top of Page
Page: of 4 Previous Topic Topic Next Topic  
Next Page
 New Topic  Reply/Répondre
 Printer Friendly
Jump To:
GpsPasSion Forums © 2002-2014_GpsPasSion/Manzanite Go To Top Of Page
This page was generated in 0,77 seconds. Powered By: Snitz Forums 2000 Version 3.4.05