> But whether you use a USRP2, a USD 20,000 military SDR or a small custom
> cheap board will not change the fact that somebody still needs to write good
> demodulaton/decoding software. And any work spent on new hardware development
> is not going to bring any progress to the project.
That's the point I was trying to raise when I suggested that efforts
would be better spend exploiting to the full extend the HW we have
rather than building new one.
A single USRP with 2 db should be able to capture at least 3 MHz of
uplink and downlink synchronized. If the ARFCN of your hopping
sequence fit into those 3 MHz, then you're good.
As mentionned OpenBTS laurent's decomposition demod seems to be way
better than the current one (from the limited testing I did). Another
benefit is that you can exploit CUDA _a_lot_ for the first stage of a
multi ARFCN receiver. (when you do the math you'll see that things fit
together nicely and you end up with a bunch of complex MACs and at the
output you have N channels of I/Q samples pre-multiplied for
laurent's).
But yeah ... need time to implement all that :(
> Next problem is: If you ever want to tune into the uplink, you need custom
> hardware as the filters for the frequencies are reversed. Plus, it is
> impossible
> to find any SAW filters for the reverse bands in the same or similar
> mechanical
> dimensions - not even speaking of the same impedance and
> balanced/unbalanced-ness
> as needed. Thus, it is impossible to simply replace the filters on existing
> hardware.
Well, on that point I don't agree ...
1) You can just ignore the filter: Works fine as long as the handset
is close enough, or if you're using a good antenna. Those filter are
just not that good.
2) You can do it 'brutally' and put a pre-amp in front of the antenna.
Definitely not the cleanest solution but it'll work. And if you want
to bring the uplink and the downlink 'in the same ballpark', just add
a filter to attenuate downlink as well.
3) There are simple SMD 'baluns' that will convert the unbalanced
50ohm to balanced 50ohm with a pinout close enough that soldering them
in place of the other could works. They (that one I haven't tried yet,
but I'll order a couple and see what I can work out when I have some
spare time)
The hardware is definitely limited but for demonstration purposes it
can do the trick IMHO. Workaround 1 is good for that use-case.
Workaround 2 is an option if you already have the parts laying around
from other GSM experiments. The option 3 ... well not tested yet so
can't really say anything.
OTOH, as you said creating a brand new board based on those chips
would be a loss of time.
Cheers,
Sylvain
_______________________________________________
A51 mailing list
A51@lists.reflextor.com
http://lists.lists.reflextor.com/cgi-bin/mailman/listinfo/a51
