Ok, great. This simplifies things; I was afraid I was going to have to
create
a modulating circuit to modulate those low-frequency timing signals so
they
could be captured along with the 2 MHz am-modulated signals using a
single
tuning frequency.
I have some questions about the deinterleaver. Firstly, how does it
know how
many channels are interleaved? Does it determine this information from
the Mux
settings, or from the number and type of subdevices detected, or
something
else?
Secondly, in terms of computational efficiency, is there any advantage
to
performing the deinterleave operation earlier in the flow-graph
sequence, or
later? For example, I’m doing AM demodulation. So, currently my flow
graph
consists of the usrp block connected to a low-pass filter, connected to
a mag
block, connected to a high-pass filter (to remove dc), connected to a
block
that corrects for gain, connected to a block that factors in a
“calibration
constant” for the data path, and finally the file-write. I also branch
off to
some fft sinks. Is it necessary for me to introduce a deinterleave
block
immediately after the usrp block or can I put it anywhere?
Thirdly, if all I was doing was capturing to file, is the deinterleaver
strictly necessary? I could write a program to deinterleave the data in
the
file in post-processing, correct?
thanks!
eric