Hi all-
I got a simple form of frequency hopping working without too much
trouble,
by having a free-running flowgraph in thread A, while thread B
periodically
retunes the flowgraph’s USRP output. This seems to work fine.
Now, what are some techniques to use if I want to introduce dead time to
the
transmission? In other words, I want something like:
Transmit at Freq 1 for 50ms
silent for 70ms
Transmit at Freq 2 for 50ms
silent for 70ms
Transmit at Freq 3 for 50ms
…
etc.
Should I make my flowgraph be “single-shot”, and just restart it every
transmit?
If I’m using a vector_source_b to provide input to the system, do I need
to
call rewind() prior to each flowgraph.start()?
If I’m using a file_source_b to provide input to the system, do I need
to
call seek(0,0) prior to each flowgraph.start()?
Or should I make the flowgraph continuous, but try to prevent any data
from
reaching the USRP during the dead time?
Should I use usrp_out.set_auto_tr(True), or should I manually toggle
usrp_out.set_enable(True) and set_enable(False)?
How do I avoid usrp underruns (“uU”), if I am cutting off data from the
USRP
periodically?
How much tuning range do I have with the DDC alone? Is it +/- 32MHz? IE
after I tune my daughtercard LO to 700MHz, I can tune from x to y solely
with the DDC…what are x and y?
Thanks for any wisdom!
-Steven