Forum: GNU Radio Frequency Counter‏

Announcement (2017-05-07): is now read-only since I unfortunately do not have the time to support and maintain the forum any more. Please see and for other Rails- und Ruby-related community platforms.
Dimitris Siafarikas (Guest)
on 2014-05-05 10:27
(Received via mailing list)
Hi list, Has anyone been able to implement a frequency counter yet?
Any ideas on how I should try it?
330ebbec4dc7177866ead8cde767d6f8?d=identicon&s=25 Louis B. (madengr)
on 2014-05-05 17:42
(Received via mailing list)
jimsiaf91 wrote
> Hi list, Has anyone been able to implement a frequency counter yet?
> Any ideas on how I should try it?
> _______________________________________________
> Discuss-gnuradio mailing list

> Discuss-gnuradio@


I have been pulling FFT vectors out of the flow into python using a
queue, then finding the frequency of the maximum bin.  I suppose you
do this, retune the USRP or channelizer, then increase the decimation.
doing this to narrow down the spectrum.

I suppose a gated type counter would work too, where you just could the
cycles, and may work better for modulated signals.


View this message in context:
Sent from the GnuRadio mailing list archive at
Mike Jameson (Guest)
on 2014-05-05 17:56
(Received via mailing list)
The 'spectrum sense mode' in scanoo locks on to the strongest signal
the rx bandwidth.  Just set the squelch and then enable spectrum sense
the spectrum sense tab:


Mike Jameson M0MIK BSc MIET
Marcus Müller (Guest)
on 2014-05-05 17:58
(Received via mailing list)
Hi Dimitris,

a frequency counter counter is a very analog-world concept, I'd say.
As DSP person I'd assume you'd want a frequency estimator.
You can implement a simple one by using an FFT block, abs, max arg
chain, or just get your result from looking at any FFT plot.

Anyway, for spectrum estimation, I recommend the gr-specest toolbox,


On Mon, May 5, 2014 at 10:25 AM, Dimitris Siafarikas
Marcus Müller (Guest)
on 2014-05-06 12:54
(Received via mailing list)
Ah sorry, I didn't mean to put it that way: It's not an analog device;
the thing I wanted to point out
that there being a device that counts frequency --to me-- belongs to a
very "analog" thinking (i.e. it's not math on a computer, but an actual
thing with an HF connector). Actually, you're absolutely right, counting
things in discrete steps in defined time intervals is the most digital
thing one could do.

Thanks for your mail!

4c6bdde8daef67c94285dac9a01e8c19?d=identicon&s=25 Activecat K. (activecat_k)
on 2014-06-03 14:06
(Received via mailing list)
On Mon, May 5, 2014 at 4:25 PM, Dimitris Siafarikas

> Hi list,
> Has anyone been able to implement a frequency counter yet?
> Any ideas on how I should try it?

Hi Dimitris,

I have recently created a block for frequency counting, refer

Try to run this flowgraph:  gr-test2/examples/flowgraph_04.grc
amazingly it shows quite an accurate result.

Let's test it with various frequencies by adjusting the GUI Slider.

I created this block when I need to observe the frequency stability of
signal from USRP, as in flowgraph_05.
Hope this block is useful to you too.
4c6bdde8daef67c94285dac9a01e8c19?d=identicon&s=25 Activecat K. (activecat_k)
on 2014-06-06 04:46
(Received via mailing list)
On Tue, Jun 3, 2014 at 11:41 PM, Dimitris Siafarikas

> Exceptional work! Looking forward to try this with my usrp tomorrow ;)


How is your test result with USRP, any updates ..?
4c6bdde8daef67c94285dac9a01e8c19?d=identicon&s=25 Activecat K. (activecat_k)
on 2014-06-08 02:36
(Received via mailing list)
First of all, please stay on the list.
Refer inline reply below.

On Fri, Jun 6, 2014 at 11:12 PM, jimsiaf91 [via GnuRadio] <> wrote:

> Well,
> I tried it with my usrp N210. The first time it showed something on the
> Scope sink. After that it did not.

Your setup could be wrong.

> Only when I run first the flowgraph_6.grc then the 5th ( the one with the
> USRP source) showed something. I do not understand why this is happening.

I see no correlation on this.

> Besides that, I tried to measure at 2Ghz with and external frequency
> generator (sin) and also attached a number sink after the block you wrote.
> The number sink did not show 2Ghz (in units), instead jittered at about
> 1000 units.

If the Sample Rate (samp_rate) of the UHD (USRP) source is only 100kHz,
cannot measure 2GHz external frequency.
Refer Nyquist sampling theory.
This topic is locked and can not be replied to.