Re: [Discuss-gnuradio] Signal power w.r.t the amplitude of sample sent to USRP?

[Marcus D. Leech]

On 07/31/2011 12:23 PM, Marcus M wrote:
> Hi,
> How can we determine the transmitted signal power with respect to the 
> amplitude value of the sample sent to the USRP and also how can we 
> measure the power of the received signal at the USRP?
>
> Thanks
In general, at least under UHD, an envelope value of 1.0 delivers 
maximum baseband power to the transmit up-converter, whose actual
  power output is dependant on the TX gain setting, and the delivered 
output power of the amplifier at any given frequency at that
  gain setting.  Actual measured RMS power on a precision RF power 
measurement instrument will also depend very much on the
  specifics of the waveform being transmitted.

There can be batch-to-batch variability in RF power amplifiers of 1dB or 
more, and RF power amplifiers also have frequency-dependant
  gain.

So, it's hard to give you a straightforward numerical answer.  The only 
thing you can do is measure and calibrate, that is also true in the
  receive direction as well.

Received signal power is *proportional* to AVG((I**2) + (Q**2)).  By 
injecting calibrated signal sources at various gain settings, you can
  determine what the proportionality will be for the specific device 
under test.  Due, again, to batch-to-batch variability in the gain of
  various elements in the analog portion of the system, you can't get 
exact values that are reproducible across multiple instances of
  the same device.  This is why laboratory instruments for measuring RF 
power are calibrated at the factory, and must be recalibrated
  on a regular basis thereafter (typically every couple of years).



--
Marcus Leech
Principal Investigator
Shirleys Bay Radio Astronomy Consortium
http://www.sbrac.org