Hi, I have been just wondering... would it be possible to use your project as a base for DIY encrypted cellphone? My idea is that if I could get hold to output data after voice is encoded by GSM EFR codec, I might just easily encrypt this digital stream using AES128 and build trully encrypted cellphone.
Where does actual "analog voice from microphone to digital data" conversion happends? In a layer1 that runs in a chip, or layer23 that runs on a PC? If it would be in layer23, encrypting a data stream with AES 128 whould be doable, am I missing something?
Thanks Marek -- S pozdravem / Best regards Marek Stopka Kontakty / Contacts Mobil/Cell phone:+420 608 149 955 WEB: www.stopkaconsulting.eu
Hi,
I have been just wondering... would it be possible to use your project as a base for DIY encrypted cellphone? My idea is that if I could get hold to output data after voice is encoded by GSM EFR codec, I might just easily encrypt this digital stream using AES128 and build trully encrypted cellphone.
Yes, the idea has been raised many times before.
The problem is that in a lot of network the voice is going to be decompressed in the network, transferred as PCM and then recompressed.
Where does actual "analog voice from microphone to digital data" conversion happends? In a layer1 that runs in a chip, or layer23 that runs on a PC? If it would be in layer23, encrypting a data stream with AES 128 whould be doable, am I missing something?
It's actually done lower than layer1, inside the DSP itself. It is however possible to send custom data (taken from the PC microphone for eg.). That's what the LCR integration does.
Cheers,
Sylvain
Yes, the idea has been raised many times before.
The problem is that in a lot of network the voice is going to be decompressed in the network, transferred as PCM and then recompressed.
The crypt phones, which are used from our politicians today, establishing a CSD connection for voice crypt transmission.
Yes, and in most networks that CSD connection is a kind of beacon, screaming to the network operator, "Hey! Look at me! I'm using a cryptophone! Pay extra special attention to my movements and calling patterns!"
On Apr 10, 2012, at 12:04 AM, Jama Solomo wrote:
Yes, the idea has been raised many times before.
The problem is that in a lot of network the voice is going to be decompressed in the network, transferred as PCM and then recompressed.
The crypt phones, which are used from our politicians today, establishing a CSD connection for voice crypt transmission.
On 4/10/12 7:15 PM, David A. Burgess wrote:
Yes, and in most networks that CSD connection is a kind of beacon, screaming to the network operator, "Hey! Look at me! I'm using a cryptophone! Pay extra special attention to my movements and calling patterns!"
Yeah, CSD is the PITA of the inter-working now that most Mobile Carries does trunking via Packet Switched and Compressed link.
CSD is dead!
Evviva CSD over CS (Voice)! :P
-naif
i'm also still following this, but it seems to be a dead end.
off-topic: current smartphones already have the capability to run sms[1]/voice encryption software or voip with encryption [2].
[1] https://github.com/whispersystems/textsecure [2] https://play.google.com/store/apps/details?id=com.privategsm.beta
On 05/09/2012 11:21 AM, Aaron Zauner wrote:
i'm also still following this, but it seems to be a dead end.
off-topic: current smartphones already have the capability to run sms[1]/voice encryption software or voip with encryption [2].
[1] https://github.com/whispersystems/textsecure [2] https://play.google.com/store/apps/details?id=com.privategsm.beta
The Guardian Project is working on an open, interoperable secure VoIP stack and best practices as well:
Open Secure Telephony Network - public testbed: https://ostel.me/ - research wiki: https://guardianproject.info/wiki/OSTN
Encrypting GSM without IP is possible, but in my experience of real world use, it is incredibly difficult to make work between carriers, countries, and generally the CDPD or second digital channel for something like Cryptophone to work on, is near impossible to active on most operators today.
+n
csipsimple supports zrtp
On Wed, May 9, 2012 at 4:21 PM, Aaron Zauner azet@azet.org wrote:
i'm also still following this, but it seems to be a dead end.
off-topic: current smartphones already have the capability to run sms[1]/voice encryption software or voip with encryption [2].
[1] https://github.com/whispersystems/textsecure [2] https://play.google.com/store/apps/details?id=com.privategsm.beta
On Wed, 9 May 2012 17:24:42 +0100 Francisco Guerreiro francisg@fnop.net wrote:
csipsimple
on top of android. which is kinda moot.
off-topic: i had various problems with csipsimple while using 2g or 3g. the built-in android 2.3+ voip implementation does not support voip via telco network, only wifi.
On Wed, May 9, 2012 at 7:52 PM, Michael Horn nibbler@ccc.de wrote:
On Wed, 9 May 2012 17:24:42 +0100 Francisco Guerreiro francisg@fnop.net wrote:
csipsimple
on top of android. which is kinda moot.
On 4/9/12 11:18 PM, Marek Stopka wrote:
Hi, I have been just wondering... would it be possible to use your project as a base for DIY encrypted cellphone? My idea is that if I could get hold to output data after voice is encoded by GSM EFR codec, I might just easily encrypt this digital stream using AES128 and build trully encrypted cellphone.
Where does actual "analog voice from microphone to digital data" conversion happends? In a layer1 that runs in a chip, or layer23 that runs on a PC? If it would be in layer23, encrypting a data stream with AES 128 whould be doable, am I missing something?
Yeah, it has been discussed in this thread: http://lists.osmocom.org/pipermail/baseband-devel/2011-January/000991.html
It would be really nice if you would make some experiment and post the patch and result.
Especially it would be nice to know the experience when there is transcoding and when there's no-transcoding between various operator.
It would be very nice to have a 13kbit/s raw channel to be used to transport encrypted AMR 4.75 audio samples :-)
-naif
baseband-devel@lists.osmocom.org
-
Aaron Zauner -
David A. Burgess -
Fabio Pietrosanti (naif) -
Francisco Guerreiro -
Jama Solomo -
Marek Stopka -
Michael Horn -
Nathan Freitas -
Sylvain Munaut