[USRP-users] Timed transmission after U/L errors

Dario Fertonani dario.fertonani at gmail.com
Fri Apr 10 14:18:31 EDT 2015


We are seeing a weird behavior that hints at a problem in recovering from
U/L transmit errors. I wonder if we are using the timed send API and
tx_metadata_t properly.

We call send within a loop, so that every time the transmission is timed
and that the interval between two consecutive transmissions is equal to the
desired chunk duration. Essentially, all is done with
    txMetaData.time_spec += ChunkDuration_s;
with the loop.

All goes well until an error U/L occurs. Soon after that, a few chunks
later (1-3 ms), the transmit stream appears to be misaligned. Still nice
and strong, but not aligned with the original timing.
All we want is that, after a packet doesn't get transmitted properly (U/L
errors), the stream can go on as if nothing happened. Any special care that
should be taken to achieve this behavior?

Thanks,
Dario
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20150410/2711d5bf/attachment-0002.html>


More information about the USRP-users mailing list