Received power measurement

Hi all,

I need to know if there is any method to measure the received signal
power using USRP N210. I am using gr_probe_avg_mag_sqrd_x_0 to measure
the signal strength, however, the results are not accurate.


Is there any reason that this isn’t done at a few frequencies during
manufacturing and provided in the datasheet?
I’ve noticed this with a few SDR’s. Surely, it would provide a ballpark
figure for those who could not afford expensive test equipment.

On 03/26/2014 07:04 AM, Medhat Hamdy wrote:

Discuss-gnuradio mailing list
[email protected]
Discuss-gnuradio Info Page
To get accurate readings, you have to calibrate with an external, known,

Hash: SHA1

Hi Vanush,

good question, actually.
The point here (that I try to highlight whenever a question about
received signal strength / RSSI / RX power requirement comes up) is
that received signal strength is so much depending on what you are
trying to measure, and how you go about to do that.

Although I agree that something like a table saying “tuned to
1.284GHz, for the daughterboard YMCA in revision 42 in combination
with the onboard ADC of the ettus b430 USRP, sampling a expensively
generated noise that looks white for the complete ADC bandwidth, for
which the perfectly matched signal generator proclaims that the
accumulated power is -15dBm, should yield, given the following
combination of filter and sampling rate settings, an average magnitude
of 0.2-0.3, assuming that you have self-calibrated the USRP well
enough” can be useful - but only if your application looks similar
enough to the benchmark; which it usually won’t.

These tables will be big. And of limited helpfulness - in the end, the
user has to calibrate things himself; if there is no high-end
equipment at the user’s site, then there’s no way to measure e.g.
matching of antennas, quality of cabling…

However: Medhat, you just said the results were “inaccurate”; maybe if
he elaborated on that, we’d know what you were trying to measure.
In fact, the results of avg_mag_sqrd are accurate to the point of
numerical accuracy if you’re measuring the average numerical power of
the digital domain sample – but I do get the feeling that is not what
you’re trying to do.


On 26.03.2014 13:29, Vanush V. wrote:

_______________________________________________ Discuss-gnuradio

[email protected]://

To get accurate readings, you have to calibrate with an external, known,

_______________________________________________ Discuss-gnuradio
mailing list [email protected]
Discuss-gnuradio Info Page

Version: GnuPG v1
Comment: Using GnuPG with Thunderbird -