| Versions |
  |
|
| Author |
 Topic  |
|
|
erupter
Italy
7 Posts |
 Posted - 03 févr. 2011 : 16:22:41
|
MTK (using 3329 chipsets) accepts RTCM, i guess SC104.
So i got an user/pass account with the NTRIP network.
Now from what i read SC104 is somewhat readable ( http://linux.die.net/man/5/rtcm-104 ).
But what i get from the NTRIP streams is total rubbish.
I can't see anything in there. All the terminal client programs crashed sooner or later after trying to watch an NTRIP streams.
Only streams not giving errors, but unreadable, are RTCM 2.1/2.3.
All the newer RTCM 3 are aunreadable.
Does anyone have any experience hooking up a receiver to an NTRIP stream? |
|
|
Ads
|
|
|
Hurston
30 Posts |
Posted - 08 févr. 2011 : 14:19:21
|
I gather data from ntrip servers is compressed, to aid the speedy transmission of the corrections. The official ntrip spec would probably help, but will cost you :
https://ssl29.pair.com/dmarkle/puborder.php?show=3
The alternative is to find some open source software that can already talk to ntrip servers, and see what they do with it.
How are you getting the RTCM into the MTK chip? I'm not aware of any receivers that have the RTCM input port wired up. Are you using some kind of OEM board? |
 |
|
|
erupter
Italy
7 Posts |
Posted - 08 févr. 2011 : 18:20:33
|
I work with these things.
Right now I have Fastrax IT500 modules, based on the ubiquitous MTK3329 chipset.
Nice modules, run up to 10Hz and support WAAS/EGNOS.
Do you know any program of the kind you refer to?
I was able to find programs that stream the data, but no program that deciphers it. |
|
|
|
Hurston
30 Posts |
Posted - 09 févr. 2011 : 13:39:01
|
Try http://lefebure.com/software/ntripclient.asp
If you happen to be developing a consumer level receiver that can take RTCM corrections via bluetooth, then I'd definitely buy one. I'm currently writing android software that (among other things) will be used to take an ntrip stream and either apply the corrections itself or forward the corrections to the chip. Unfortunately only the survey grade receivers will take the corrections. There are no consumer grade receivers that do. |
|
|
|
erupter
Italy
7 Posts |
Posted - 09 févr. 2011 : 17:06:26
|
Unfortunately for you I'm not in the commercial products development.
Anyway there is a reason if access to chipset control is mostly disabled: in many low-cost cases you could end up with an unusable device.
Just changing the baud rate on the internal serial port requires changing it also on the external one, otherwise you can't read anything.
This alone meant coding a command interpreter for the MTK commands able to switch speeds on the middleware.
So I guess you won't see any custom configurable consumer device any time soon.
As for the software thank you, I gave it a quick try but nothing happened.
Will keep trying. |
|
|
|
Hurston
30 Posts |
Posted - 09 févr. 2011 : 18:23:18
|
Let me guess, if you set RTCM to be received from port zero you will brick the device?
I always thought it was just a cost thing, with the extra line into port 1 being something that wouldn't be used by many people, so they didn't bother. |
|
|
|
erupter
Italy
7 Posts |
Posted - 09 févr. 2011 : 22:45:16
|
I don't know about all the gps modules out there.
The one i'm using has 2 serial ports:
-A is strictly NMEA output, doesn't receive anything
-B is input/output to receive commands (and still outputs NMEA)
As for real bricking i don't think it's possible: fact is that since in most cases there are other interfaces between the GPS module itself and the application using its data, the bricking would mean that something goes amiss in the intermediate points between app and gps. Not real bricking intended as something stopping working, rather that due to a misconfiguration the device intended as multiple modules linked together, stops working. But in 99% cases the single modules would still work by themselves only they would not be able to communicate between them thus rendering the system unfunctional. |
|
|
|
Hurston
30 Posts |
Posted - 10 févr. 2011 : 10:35:12
|
| If what you are saying about only 1 port receiving input is true, I am guessing that on the average GPS receiver, it is port B that is wired up, so that it can receive NMEA style commands? But I thought that RTCM was sent to a separate port to the port to which the NMEA commands are sent. If only 1 port accepts input, does that mean NMEA and RTCM go to the same port? Can I send RTCM to my cheapo bluetooth receiver and it would use it? I know there is an MTK NMEA command to set the port that receives RTCM. If you set that to be the same as the port that receives NMEA commands, would that cause problems with the chip? |
|
|
|
erupter
Italy
7 Posts |
Posted - 10 févr. 2011 : 11:03:51
|
On the average receiver it is PORT A that is wired up.
RTCM data goes to PORT B.
If you have an MTK chipset receiver, send $PMTK000*32.
If the receiver responds $PMTK001,0,3*30 then it accepts commands and you can try feeding it RTCM data.
But my guess is that PORT A is the one wired up, thus preventing you from controlling your receiver. |
Edited by - erupter on 10 févr. 2011 11:07:13 |
|
|
|
Hurston
30 Posts |
Posted - 10 févr. 2011 : 12:58:59
|
I've definitely sent NMEA commands to it, and it has worked. I think it is pretty standard to control consumer units in that way.
Looking at the MTK commands, it seems the one for setting the port to which the RTCM is sent is not one that appears on many devices, so I am guessing that is controlled by something outside of the chip. Either that or it is not generally published because of the problems it could cause. |
|
|
|
erupter
Italy
7 Posts |
Posted - 10 févr. 2011 : 13:02:46
|
Mmmm. I'm not familiar with what you're saying: maybe there are chips configured differently then mine (probably).
Mine on PORT A does not receive anything, not even NMEA commands.
And it does not accept NMEA commands, only MTK proprietary commands on PORT B.
So i guess things differ along with make. |
|
|
|
Martin87
1 Posts |
Posted - 25 oct. 2011 : 15:20:02
|
| I am also using Fastrax IT500 modules. I tried to feed it with Ntrip with several programs (GNSS Internet Radio, GNSS Surfer, BNC25), but the module didn't make use of the correction data. It was correctly configured (RTCM enabled). Did anyone manage to run the module in DGPS mode using Ntrip? |
Edited by - Martin87 on 25 oct. 2011 15:20:58 |
|
|
| |
Topic |
|
|
|
| This page was generated in 0,39 seconds. |
 |
|
- 
- 
MTK Gps and NTRIP streams
