What are the practical effects of automatic transmit/receive switching
on the USRP subdevices (in particular, the RFX2400). If I have a
transmit-only application, will having this option on hurt my
performance in any way? The reason I ask is that if I leave it off, then
after the application finishes the USRP continues to transmit
high-amplitude sine waves.



Dan H. wrote:


Discuss-gnuradio mailing list
[email protected]
Discuss-gnuradio Info Page

I replied hours ago from my normal email account, but don’t see it on
list. So I’ll post again from Nabble- forgive me if it shows up twice.

This sounds related to the problems I was seeing after a program ran,
the transmitter kept going. So I made that FPGA patch that zeroes out
the DACs (somewhat indirectly- it’s early in the TX path) whenever the
FIFO empties. The way I understand it (correct me if I’m wrong),
some/most of the applications just disable the transmitter rather than
waiting for the buffer to empty- and in those cases, the transmitter
keeps going. So in my tests, if I ctrl+c’d, the transmitter quit going,
but if I let the program terminate normally, it kept going.

Short of a complicated FPGA modification, the changes have to be made in
the various parts of the code to let the FIFO empty before disabling.

Perhaps Eric or Matt can comment further or tell you that I’m wrong.

View this message in context:
Sent from the GnuRadio mailing list archive at