USRP N210 Receiving distored signal even with RF cable

[ounety]

I am trying to send a very basic signal (generated on matlab and is basically random integers ranging from 0 to 1), and trying to receive it on the USRP. I am using a single USRP N210 with two daughterboards, the LFTX and the LFRX connected via cable. The program is written as an m.file and a simulink file, the transmitted signal is generated on the m.file, then, the transmitted signal is loaded from workspace into a simulink file that contains the transmitting and receiving blocks of the USRP. The received signal is then fed back to workspace for processing. Attached are the photos of the hardware and the received vs transmitted signal. The program is divided into m.file and simulink basically because I am using it to detect the problem of a more complicated program. When i tried sending a sinusoidal signal and not integers, i received an attenuated and shifted version but not distorted (which was perfectly fine). However, with the random integers the signal is distorted badly. Any help please?

[AirFrantz]

Hello,

Could you describe your signal in more detail as well as give us an overall summary of your application?  It may be difficult for me to help since you are using m files and simulink and not NI software, but a better description of the waveforms you are sending and your intention could help myself and others understand what you would like to accomplish.

[ounety]

I simply want to transmit real valued samples over the USRP at base band and receive them, using two daughterboards and a single USRP and a single PC. Previously, i thought the problem was from the receiving side. However, after further inspection, when i connected the transmitter directly to the Oscilloscope, the signal was badly distorted. To test the transmitting board, i tried sending a sinusoidal signal and i received it (a little attenuated and shifted but OK), however if the signal is saw tooth or contains rapid transitions, the transmitting board distorts it (and i made sure of that using the oscilloscope). Attached is an example of signal i generated, and how the oscilloscope sees it. Changing the interpolation rate, the frequency of the signal didn't solve the problem. The signals i generate have amplitudes limited between (-1 to 1). I hope i answered your question accurately. 

[AirFrantz]

I can see that the signal shows some oscillation near the discontinuity in the ramp signal in your screenshot.  It is likely that signal conditioning in the LFTX daughterboard is causing this behavior.  Unfortunately, I do not have access to documentation on the LFTX daughterboard to say whether this distortion is to be expected or not, as this USRP configuration is not supported by us here at NI.  You may consider reaching out to Ettus at support@ettus.com.

If you remove the discontinuity from your signal, say by outputting a triangle wave, do you still see the distortion?  What happens when you use high-impedance input mode on your scope to view the signal?