GSM SoftPhone (resend - original not sent to mailing list)

GNU Radio
1

Hi All,

Appologies about the re-post, but the original post wasn’t submitted to
the
maililing list.

I have been looking at the USRB for a work project in implementing a
softphone. What are other peoples opinions of using the USRB for
developing
a software based GSM cell phone. The reason I ask is this.

Obviously the USRB can handle the channel bandwidth of 200Khz, but these
physical channels then use TDMA to create another 8 channels within the
one
physical RF channel. Now GSM has a few timebased counters for
synchronization to the base station, the lowest timebased counter that
are
involved for synchronization (QN i belive in the GSM spec) is
incremented
every 12/13 of a microsecond. Would an OS like Linux or Window be able
to
keep up with this level of accuracy, or would a daughterboard need to be
developed to fulfill this requirment. In other words, could an OS like
Windows acuretly send out burst within the requested timeslot within the
TDMA frame that the base station desires during MS setup.

Hope that made sense.
Regards

–
View this message in context:
http://www.nabble.com/GSM-SoftPhone-(resend---original-not-sent-to-mailing-list)-tf2235198.html#a6197180
Sent from the GnuRadio forum at Nabble.com.

2

On Thu, Sep 07, 2006 at 12:52:23PM -0700, JoePub wrote:

Obviously the USRB can handle the channel bandwidth of 200Khz, but these
Hope that made sense.
Regards

I don’t know about Windows, but Linux is capable of this when the
ultimate timebase is implemented in the FPGA. This level of timing
resolution is currently in developement and will be implemented in
combination with the upcoming “message blocks”. [Check the archives
for more info on “m-block” and/or “message block”]

Eric