[USRP-users] Removing the whole LO signal when transmitting
derek.kozel at ettus.com
Mon Jul 17 06:28:35 EDT 2017
The gain values for USRPs are indexed from 0 dB gain being minimum gain up
to the device's maximum. Usually this means that 0 dB of gain is actually a
large amount of attenuation or loss through the RX chain. The same is true
for TX. It doesn't surprise me that you are seeing those values.
Ideally you will experiment to find an operating point on RX to maximize
the signal quality (minimize Noise Figure and capture the maximum dynamic
range) of your digitized signal. The performance data for the
daughterboards can help with this. There are similar tradeoffs on the TX
Also be sure to run the self-calibration routines if your daughterboard
supports them. This can improve the signal quality significantly. Marcus
has also mentioned UHD's ability to do LO offsetting to move it outside of
your passband, definitely worth investigating.
On Mon, Jul 17, 2017 at 11:47 AM, Brais Ares via USRP-users <
usrp-users at lists.ettus.com> wrote:
> Ok, I understand!
> Also I have to configure a rx_gain of 60 dB, and a tx_gain of 10 dB If I
> want to reconstruct the transmitted signal to the very same power level I
> received it.
> The scheme would be:
> ~~~ signal_generator (QPSK at -60 dBm) -> cable(0.3 dB attenuation) ->
> rx_usrp -> tx_usrp -> cable(0.3 dB attenuation) -> Signal analyzer (I want
> to obtain original power here, -60 dBm).
> *Is it normal that I have to apply roughly 70 dB of gain to replicate the
> incoming signal?*
> @Karan, did this happen to you?
> Appreciate all the help, guys
> USRP-users mailing list
> USRP-users at lists.ettus.com
-------------- next part --------------
An HTML attachment was scrubbed...
More information about the USRP-users