0
$\begingroup$

I trying to measure the frequency drift between two N210 USRPs using GNURadio Companion. I used one USRP as a TX and the other one as an RX. The transmitted signal is a complex sinusoidal.

The issue that I am facing is how to get the modulated signal from the TX before the transmission through the antenna. In other words, I am trying to connect the UHD:USRP Sink to a frequency sink to see the FFT of upconverted/modulated signal to be compared with the signal received at the RX. However, the UHD:USRP Sink cannot be connected to the Frequency sink in GNURadio.

So, is there anyway to check the modulated signal from the TX?

Can I connect the USRP to a real oscilloscope?

Improve this question
$\endgroup$

1 Answer 1

Reset to default
1
$\begingroup$

The issue that I am facing is how to get the modulated signal from the TX before the transmission through the antenna.

Since that's already an analog RF signal: You would need a receiver.

So, you'd need to attach a receiver instead of an antenna.

In other words, I am trying to connect the UHD:USRP Sink to a frequency sink to see the FFT of upconverted/modulated signal to be compared with the signal received at the RX.

That's not how that works. The signal, after being upconverted, is too fast to be sampled, that's why you synthesize it in baseband.

Can I connect the USRP to a real oscilloscope?

Sure, but it'd need to be an oscilloscope whose bandwidth is larger than the highest frequency.

It's not clear why you think you need to do this. Just connect the transmitting USRP to the receiving USRP, and observe the frequency difference in baseband.

Improve this answer
$\endgroup$
6
  • $\begingroup$ Do you mean connecting the TX and Rx using a coaxial cable? $\endgroup$
    – wesam amiri
    Commented Nov 2, 2022 at 11:49
  • $\begingroup$ yes, and please also include an attenuator. You can damage a sensitive receiver with the full output power of a USRP. $\endgroup$
    – Marcus Müller
    Commented Nov 2, 2022 at 11:50
  • $\begingroup$ Okay. I will try this method. But my idea was also to see the effect of the environment on the frequency drift. Thanks! $\endgroup$
    – wesam amiri
    Commented Nov 2, 2022 at 11:54
  • $\begingroup$ there's none, unless you have a moving reflector,( or surprisingly high transmitter power, multiple tones, and very nonlinear media in your environment (you do not.)). $\endgroup$
    – Marcus Müller
    Commented Nov 2, 2022 at 11:56
  • 1
    $\begingroup$ Any bistatic radar that needs to estimate the frequency shift caused by a moving reflector has to deal with the frequency error between TX and RX, anyway. So, you need to have a (typically, stationary) reference path to estimate that frequency offset first, or you need to take some properties of the possible trajectories of physical reflectors into account to do that offset removal. Mathematically, there's zero difference between a frequency offset of RX relative to TX and the frequency shift caused by a constant relative motion. $\endgroup$
    – Marcus Müller
    Commented Nov 2, 2022 at 12:00

Your Answer

By clicking “Post Your Answer”, you agree to our terms of service and acknowledge you have read our privacy policy.

Not the answer you're looking for? Browse other questions tagged or ask your own question.