I am using GNURadio with a USRP1 and would like some clarity on how the
USRP1 works because I’m new to both. Because of the limited speeds of
USB transmission, the USRP1 cant efficiently perform a spectrum
analysis. So to analyse a large range of frequencies, GNURadio measures
8 MHz chunks by changing the center frequency and stitches the data
together before performing an analysis. But constantly changing the
center frequency causes a time delay before the correct samples from the
ADC can be seen by the computer. Are there any better ways to do this
with a USRP1? Is it possible to apply a sub-Nyquist sampling rate in
hardware before transmission through USB to create a more efficient
spectrum analysis? From what I understand, a sub-Nyquist sampling rate
can somehow allow the reconstruction of a signal using a rate below
Nyquist. If possible, this would mean more information could be
transmitted through the same USB cable and fewer changes in the center
frequency. Or is GNURadio’s implementation about as good as its going to
get on the USRP1?
I already asked ETTUS about directly controlling the hardware, and they
suggested I ask the GNURadio community.