[USRP-users] Waveform Corruption

Steve Peters stevenpeters at gmail.com
Mon Feb 6 18:15:00 EST 2012


Hi all,

In testing my own code I've noticed that my preamble, which should be
constant amplitude, is somehow being corrupted so that its amplitude is
varying widely.

To perform a sanity check, I've run

tx_waveforms --wave-type SINE --wave-freq=40e3 --freq=1750e6 --rate=250e3
--ampl=0.3

and at the receiver run an FFT (using the GNU Radio companion) and checked
out the time-domain samples using my own recording & plotting tools.  I've
also recorded the samples using rx_samples_to_file, and the result is
similar.

The results are attached.  As you can see, there are harmonics every 40
kHz, which make the time-domain samples look pretty bad.  I've found that
this effect gets better as the sinusoid frequency increases.  For example,
at 100e3, the harmonics are present but are about 15 dB below the main
sinusoid.  At 1e6, I get a pretty clean sinusoid.

I've tried this in both directions (i.e., had both USRPs act as TX or RX),
and it's the same either way.  I'm really not sure what could be causing
this.

I'm running USRP N210 with RFX1800 daughtercards.  Communication is over an
SMA cable with 66 dB attenuation, and RX gain is set to 45 dB.

Regards,
Steve
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20120206/2397aeee/attachment-0002.html>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: fft.png
Type: image/png
Size: 49318 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20120206/2397aeee/attachment.png>
-------------- next part --------------
A non-text attachment was scrubbed...
Name: time_domain.png
Type: image/png
Size: 38581 bytes
Desc: not available
URL: <http://lists.ettus.com/pipermail/usrp-users_lists.ettus.com/attachments/20120206/2397aeee/attachment-0001.png>


More information about the USRP-users mailing list