[Top][All Lists]

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: [Discuss-gnuradio] [USRP-users] B205mini half-duplex? (GRC)

From: Marcus D. Leech
Subject: Re: [Discuss-gnuradio] [USRP-users] B205mini half-duplex? (GRC)
Date: Fri, 05 Jul 2019 10:53:42 -0400
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.7.0

On 07/05/2019 09:36 AM, Tom McDermott via USRP-users wrote:
Hi Marcus - thanks for the link to the documentation.  Messages don't have
the required functionality, so I have tried to insert tx_sob and tx_eob tags.
This unfortunately has not changed anything.  This is what the tag debug shows,
are these formatted correctly for UHD USRP sink?

Tag Debug: 
Input Stream: 00
  Offset: 139250  Source: burst_tagger4     Key: tx_sob   Value: #t

Tag Debug: 
Input Stream: 00
  Offset: 196587  Source: burst_tagger4     Key: tx_eob   Value: #t

I have to stop streaming samples to the USRP Sink to get it to stop transmitting and
switch TX/RX over to RX - the tags aren't enough.  However the next time back to Tx,
the problem with transmitter underruns and weird oscillating T/R behavior resumes.

Still looking for advice as to how to T/R switch the unit without the TX going nuts?

-- Tom, N5EG

Indeed, if there are ANY samples after a tx_eob, the USRP will go back into transmitting mode.

I'm copying the discuss-gnuradio mailing list, since this is really an issue of how to make tagged streams behave
  as-expected with GRC-derived flow-graphs.

I've never implemented a half-duplex TX scheme out of GRC myself, but there are folks on the discuss-gnuradio mailing list
  who have.

On 07/04/2019 11:14 PM, Tom McDermott via USRP-users wrote:
> I am trying to use a B205mini in half-duplex mode through the TX/RX 
> connector
> from a GRC flowgraph.  Gnuradio  Very slow switching, 
> manually implemented.
> The flowgraph implements a simple half-duplex transceiver. In order to 
> switch
> the TX/RX antenna switch, I am starting and stopping samples to the 
> USRP sink
> using the Copy block.  When the Copy block is enabled it passes 
> samples to the USRP sink,
> and when disabled it does not copy samples to the USRP sink.
> If the Copy block is statically enabled, then the USRP transmits (RED 
> led 'On' for TX/RX)
> continuously without any underruns.
> If the Copy is statically disabled, then USRP goes into receive mode 
> (GREEN led 'On' for TX/RX)
> and stays in that mode.
> So far so good.
> However if I toggle the enable/disable on the copy block to disable 
> the TX/RX switches as it should
> from red to green, but after I re-enable the Copy block and leave it 
> enabled, the TX/RX  LED goes red
> for awhile, then rapidly flashes red/green/red etc. while printing 
> UUUUUUU  on the GRC console, then
> goes RED for about 5 seconds without underruns, then flashes RED/GREEN 
> rapidly for 5 or 10 seconds
> and underruns, back and forth ad infinitum.
> So Copy appears not to be a good way to implement T/R switching.
> There is a message port on the USRP sink block.  Can this be used to 
> implement T/R switching?
> Is there some place that the messages that this port accepts defined?  
> I cannot seem to find
> the syntax for the message definitions in the USRP documentation, so I 
> must be looking in the wrong place.
> -- Tom, N5EG
There's some documentation here:


USRP-users mailing list

reply via email to

[Prev in Thread] Current Thread [Next in Thread]