[Discuss-gnuradio] Software mobile phone

2010-07-23 Thread Abdalaleem Andy James Potter

Dear All,

Do we think it is possible to create a software mobile phone using the  
USRP, with the OpenBTS code or something else?


I mean everything would be in software, plus the USRP?


Abdalaleem

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


Re: [Discuss-gnuradio] Software mobile phone

2010-07-23 Thread John Gilmore
 Dear All,
 
 Do we think it is possible to create a software mobile phone using the  
 USRP, with the OpenBTS code or something else?
 
 I mean everything would be in software, plus the USRP?

It is absolutely possible.  So far I don't know anyone who has
tried to do it.  The OpenBTS code would give you a big head start.

I also think it would be interesting to port the resulting code into a
mobile phone.  Generally the GSM protocols in a phone are run in a
baseband processor separate from the user interface processor.
Every phone I know of uses secret, proprietary code running in the
baseband processor, even when the user interface is largely free
software.  Once you had working code running in GNU Radio on a Linux
machine, the challenge would be finding a well-documented baseband
chip (in which the manufacturer tells you where to find the radio I/O
gear on the chip, and how it works, etc).  Porting clean GSM code to
run in that chip in realtime would require some adaptation to exploit
unusual on-chip DSP hardware, mastering an embedded debugging
environment, and perhaps shrinking the memory consumption of the GNU
Radio-based code.

I think it's not only doable, but well worth doing.  It should be
worth a couple of PhDs at least.  You would certainly know the GSM
protocols inside and out by the time you were done!

John

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


Re: [Discuss-gnuradio] Software mobile phone

2010-07-23 Thread Sylvain Munaut
Hi,

 The OpenBTS code would give you a big head start.

The tranceiver part sure. But you'd still need pretty big modification
to have it act as a TDMA clock slave and not TDMA clock master (that's
quite different).

Once you have done the L1 inside the USRP somehow, you can re-use the
osmocom-bb code (which is an opensource baseband implementation project)
for the upper layers.

Sylvain Munaut

___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio


Re: [Openbts-discuss] [Discuss-gnuradio] Software mobile phone

2010-07-23 Thread Abdalaleem Andy James Potter

Fantastic stuff!


On 23 Jul 2010, at 22:26, Joshua Lackey wrote:


And leveraging the work from the osmocombb project
(http://bb.osmocom.org/trac/) will get you a far ways towards the  
goal.



Quoting John Gilmore (g...@toad.com):

Dear All,

Do we think it is possible to create a software mobile phone using  
the

USRP, with the OpenBTS code or something else?

I mean everything would be in software, plus the USRP?


It is absolutely possible.  So far I don't know anyone who has
tried to do it.  The OpenBTS code would give you a big head start.

I also think it would be interesting to port the resulting code  
into a

mobile phone.  Generally the GSM protocols in a phone are run in a
baseband processor separate from the user interface processor.
Every phone I know of uses secret, proprietary code running in the
baseband processor, even when the user interface is largely free
software.  Once you had working code running in GNU Radio on a Linux
machine, the challenge would be finding a well-documented baseband
chip (in which the manufacturer tells you where to find the radio I/O
gear on the chip, and how it works, etc).  Porting clean GSM code to
run in that chip in realtime would require some adaptation to exploit
unusual on-chip DSP hardware, mastering an embedded debugging
environment, and perhaps shrinking the memory consumption of the GNU
Radio-based code.

I think it's not only doable, but well worth doing.  It should be
worth a couple of PhDs at least.  You would certainly know the GSM
protocols inside and out by the time you were done!

John

--
This SF.net email is sponsored by Sprint
What will you do first with EVO, the first 4G phone?
Visit sprint.com/first -- http://p.sf.net/sfu/sprint-com-first
___
Openbts-discuss mailing list
openbts-disc...@lists.sourceforge.net
https://lists.sourceforge.net/lists/listinfo/openbts-discuss



___
Discuss-gnuradio mailing list
Discuss-gnuradio@gnu.org
http://lists.gnu.org/mailman/listinfo/discuss-gnuradio