Hi all,
i'm testing gprsdecode with osmocomBB and using the samples provided from Nohl works well.
Using a Blackberry 8300 and my Wind's sim, i tried to dump some my internet sessions. Unfortunately even if i'm locking the MS to a fixed ARFCN, it is "upgrading" from GRPS to EDGE very quickly and i cannot dump an entire session and then decode it on gprsdecode.
The only datas i could decode and view on wireshark are few "malformed packets".
As i suppose the problem is that i'm not able to fix the MS to work only on gprs.
Which phone did Luca use to make his tests? (just a simple one that was not EDGE capable? )
Which was the successful-rate of (third-party) decoded sessions?
I'm asking it, because i'm not even able to "see" other's phones sessions... ok, could be possible that everyone else is using only EDGE... but seems strange.
Have you some hints for that?
Thank you for attention.
Regards,
Luca
Hi Sylvain,
as i understood, Luca made some tests with real networks in Italy, infact as he stated during the talk, Wind [222-88] and Tim[222-01] (two providers) are using GEA/0.
Assuming it, i wanted to make some tests, trying to dump my own sessions while using a Wind's sim.
Unfortunately i'm having some issues due gprs is continuously upgrading to EDGE and not having an old "gprs-only" phone, the only solution that i'm trying is to force it to remain in gprs-mode by reducing the gain and creating noise with the usrp.
Btw... what i was curious to know is... also Luca, while testing its own sessions, was not able to see other third-party gprs sessions on the air?
I know that almost all real users are using EDGE or UMTS, but how is possible that no one gprs session is available?! I'm not interested on third-party sessions, i'm just trying to understand if it is working well or it is having a strange behaviour and there is something to patch.
I think the nanoBTS serving the network was just configured to do GPRS and that's it.
P.S: Of course i would like to test it with my own test set, but at the moment i cannot.
Regards,
Luca
On Sat, Aug 27, 2011 at 03:10:35PM +0200, Luca Bongiorni wrote:
Unfortunately i'm having some issues due gprs is continuously upgrading to EDGE and not having an old "gprs-only" phone, the only solution that i'm trying is to force it to remain in gprs-mode by reducing the gain and creating noise with the usrp.
I think this sounds like a lot of effort and unreliable. Why not just ues an old phone that is not EDGE capable? you can attach such a phone to the serial port of your PC and then generate GPRS traffic through it. Or even get a GPRS modem instead of a phone.
There are several mobile phone databases on the web that show you if a phone supports GPRS and/or EDGE.
Also, the Ti Calypso cannot do EDGE, so any phone that uses this chipset and has GPRS support would probably do the trick. The C155 should fit that description (but you cannot attach it to a PC and have to generate traffic from the built-in WAP browser). The OpenMoko GTA01 and GTA02 also fit that description.
Regards, Harald
Hi Harald,
Also, the Ti Calypso cannot do EDGE, so any phone that uses this chipset and has GPRS support would probably do the trick. The C155 should fit that description (but you cannot attach it to a PC and have to generate traffic from the built-in WAP browser). The OpenMoko GTA01 and GTA02 also fit that description.
thank you for your suggestions.
Actually i could use some old phones that supports only GPRS (e.g. t68i, t39m, etc..), but these phones are not enough useful and i will explain why:
- i would like to use a phone that has field-test engineering; - i would like to set the phone to camp only in a fixed ARFCN.
I was also trying a Pirelli DP-L10, but unfortunately i can only fix the band and not the arfcn.
Btw... after a while i finally found how to force my Blackberry to work on GPRS on a fixed arfcn:
- Select an arfcn; - Lock to this cell; [ http://tinyurl.com/3dcc6vm ] - Disable Uplink 8-PSK; [ http://tinyurl.com/3dxp85d ]
Finally the phone is ready to start the tests. [ http://tinyurl.com/3rhldrn ]
As you can see, that engineering field-test is very useful because is showing which Coding Scheme and TS are used.
Unfortunately at this point... even if i tried a lot of times, i could see imm-ass, the TS, but when i'm trying to decode the dumps... nothing is decoded and i cannot even decode third-party sessions that obviously i could find on the same arfcn that i'm using.
So, at this state of art, I could decode only samples released from Nohl. Someone else is having same problems?
Regards,
Luca
baseband-devel@lists.osmocom.org
-
Harald Welte -
Luca Bongiorni -
Sylvain Munaut