On Thu, Feb 14, 2013 at 2:44 PM, Pablo Belzarena <[email protected]> wrote:
> Hi everyone, I am newbie on gnuradio and I have a question about ofdm
> benchmark. If I use benchmark_tx.py and benchmark_rx.py I have a good
> performance, (an error rate below 10%).
> However, if I only modify the benchmark_tx.py and I add only a
> time.sleep(1) between send_pkt(payload) like is shown in the
> following code of the benchmark_tx.py:
>
> while n < nbytes:
> if options.from_file is None:
> data = (pkt_size - 2) * chr(pktno & 0xff)
> else:
> data = source_file.read(pkt_size - 2)
> if data == '':
> break;
> payload = struct.pack('!H', pktno & 0xffff) + data
> send_pkt(payload)
> n += len(payload)
> print (n)
> if options.discontinuous and pktno % 5 == 4:
> time.sleep(1)
> pktno += 1
> time.sleep(1) # this is the line added
> send_pkt(eof=True)
>
> In this case, the packets error rate is above 50% and there are some
> underruns on the transmitting screen.
>
> Could someone explain me where is the problem and how can I solve it?
>
> Thanks in advance,
>
> Pablo Belzarena.
>
This code must not be setting the burst conditions for the UHD driver
correctly. There's some ongoing work that's overhauling the OFDM code for a
lot of the major advancements in GNU Radio since these benchmark scripts
were first written. I think this is the best place to look at that code
development:
git://github.com/benreynwar/gnuradio.git
Using the branch 'ofdm'. Not sure how usable it is right now, though.
Tom
_______________________________________________
Discuss-gnuradio mailing list
[email protected]
https://lists.gnu.org/mailman/listinfo/discuss-gnuradio
