Hi,
I’m new to radio and I have a lot to learn. I have a question about
oscilloscopes and the usrp_oscope.py.
I believe the following. Please correct me where I am out to lunch.
There is a strong 103.3Mhz FM station in my area. Connect an antenna to
a band pass filter that passes 103.3Mhz ± a little bit. Plug the
output of the filter to an oscilloscope (a real one) and you will see a
103.3Mhz wave. Note: you do not tune the oscilloscope.
My questions about usrp_oscope.py:
How do I see the carrier wave at 103.3Mhz?
What are the units of the Y axis?
Why does it want to tune?
What is usrp_oscope.py showing me?
Thanks!
Chris
On Mon, Dec 04, 2006 at 10:03:25PM -0800, Chris S. wrote:
103.3Mhz wave. Note: you do not tune the oscilloscope.
My questions about usrp_oscope.py:
How do I see the carrier wave at 103.3Mhz?
Assuming you’ve got a TV_RX daughterboard on side A:
$ usrp_oscope.py -R a -f 103.3M -d 256
This will plot the signal that is received in a 250kHz wide window
centered at 103.3 MHz. (64e6 / 256 = 250e3)
What are the units of the Y axis?
They are the output of the digital down converter.
They are not calibrated to any particular voltage level.
The levels depend on the gain setting too.
Why does it want to tune?
Because it wants to know what portion of the RF spectrum you are
interested in.
What is usrp_oscope.py showing me?
The real and imaginary parts of the complex baseband signal coming
from the output of the digital downconverter in the USRP.
Since it sounds like you’re not too familiar with what’s going on, you
may want to play with usrp_fft.py first. It’ll show you the same
data, but in the frequency domain instead of the time domain.
If this isn’t making much sense, try spending some time with Richard
Lyon’s “Understanding Digital Signal Processing”.
Eric
Eric B. wrote:
If this isn’t making much sense, try spending some time with Richard
Lyon’s “Understanding Digital Signal Processing”.
Eric, all:
First, thanks for your answers. I’m going to check out the book you
recommended. I have “Telecommunication Breakdown” and of course
wikipedia which are helpful.
Second, I tuned the USRP oscope to 103.3MHz. I believe the USRP
downconverted 97.55MHz to 0Hz and 103.3MHz to 5.75MHz. It clearly
showed a sine wave which I’m sure is the 103.3 MHz carrier frequency.
I used the graph (see attached) to calculate the frequency to make sure
it agreed with my 5.75MHz belief. I counted 10 cycles over 3 divisions:
10 cycles / 3 div
50 us / div = 50e-6 s/div (assuming u is mu)
10 cycles / 3*50e-6 s
10 cycles / 150e-6 s
10/150e-6 Hz
10/150 MHz
1/15 MHz
0.066 MHz
Two orders of magnitude off.
Did I do something wrong? I’m using TVRX on RXA.
Thank you!
Chris
On Tuesday 05 December 2006 01:26, Eric B. wrote:
On Mon, Dec 04, 2006 at 10:03:25PM -0800, Chris S. wrote:
103.3Mhz wave. Note: you do not tune the oscilloscope.
$ usrp_oscope.py -R a -f 103.3M -d 256
This will plot the signal that is received in a 250kHz wide window
centered at 103.3 MHz. (64e6 / 256 = 250e3)
What would be useful is for oscope is to have the X axis scaled
according to
the actual viewed frequency, not the frequency after downconversion.
This
would be the same as the fft program displaying in -X to X MHz instead
of
showing the acutal tuned frequency on the axis.
In other words, make oscope look like a real oscilloscope display rather
than
the downconverted display now. I have some other ideas; I need to look
at
QtDSO (which works with the Velleman PCS64i) and see what would be
required
to do the same with the USRP code.
Lamar Owen
Director of Information Technology
Pisgah Astronomical Research Institute
1 PARI Drive
Rosman, NC 28772
(828)862-5554
www.pari.edu
Lamar Owen wrote:
What would be useful is for oscope is to have the X axis scaled according to
the actual viewed frequency, not the frequency after downconversion.
You can do this with the fft scope by adding this call inside set_freq:
self.scope.set_baseband_freq(target_freq)
BTW Eric, I bought Richard Lyons’s book, thanks for the referral.
Chris