USRP2 "input spectrum data is of an incorrect type" error when using SMT Power in Band

[fucai1116]

I followed the Measuring-power-using-NI-USRP-and-Spectral-measurements-tool-kit to modify example niUSRP EX One Shot Rx-Modified.

But I got "input spectrum data is of an incorrect type" error when I tried with FFF Spectrum (Mag Phase) or FFF Power Spectrum.

I am using USRP2 and NI_USRP 1.3 driver.

Could you help me figure out the problem? See attached vi.

Thanks

[AirFrantz]

Hello,

I believe the error you are seeing happens when the power spectrum is not in the correct units for the 'Power in Band.vi'  I believe that the units default to dB and the Power in Band VI needs V^2.  Try to use the SMT Convert function to change the units and see if this gets rid of the error.

[fucai1116]

I modified and "input spectrum data is of an incorrect type" error is no longer occur. But there is a problem that measured power is always 0.00. Could anyone help me explain the reason? Thanks.

[Tunde_S · Accepted solution]

Hi fucai1116,

According to the Detailed Help documentation for the SMT Power In Band VI, the band specification specifies the properties of the band or channel in which you want to compute the power. As a result, I believe you need to set the band specification based on the center frequency, after the signal is downconverted by the USRP. For example, if the RF signal is centered at 100 MHz and you configure the USRP’s carrier frequency to be 100 MHz, the output from the USRP Fetch VI should be a signal centered around 0 Hz. The Magnitude Spectrum indicator in your VI should also confirm that the band is centered at 0 Hz. Therefore, you should set the band specification's center frequency to 0 Hz (not 100 MHz).

Regards,

[fucai1116]

Tunde S.,

Thank you. I changed the center freq to 0 Hz and I can get power in band value using FFT Spectrum + Power In Band.

However power in band value still is 0 at FFT Power Spectrum + Power In Band. Could you give me the reason?

Regards,

[xavvvv · Accepted solution]

Hi fucai1116,

What exactly are you measuring as far as input signal? The power in band2 might be showing 0.00, but I would recommend increasing the number of digits of precision or changing the display format to exponential (it might not be quite zero). If the incoming power spectrum has a peak at a very low amplitude (for example, 1E-5), you will not be able to see too much power in the band of interest. Are you actually measuring a signal at all? Please provide a screenshot of a graph indicator of the output of the Power Spectrum and PSD function.

[fucai1116]

Thank you for your suggestion.

The number is too small. I change the number digits of precision and I can see it.