discuss-gnuradio
[Top][All Lists]
Advanced

[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Re: First tutorial


From: Fabian Schwartau
Subject: Re: First tutorial
Date: Wed, 30 Mar 2022 13:45:03 +0200
User-agent: Mozilla/5.0 (X11; Linux x86_64; rv:91.0) Gecko/20100101 Thunderbird/91.7.0

Hi Alexander,

that is obviously a "problem" in gnu radio ;)
It depends on the window chosen and if it is normalized or not. The problem with the window functions in an FFT is, that you spread the power of the signal over multiple bins (which also happens when you have spectral leakage). You can either normalize the window so that the peak power will be correct (what you would like to see) or that the total mount of power/energy in the complete spectrum is preserved (what gnu radio is probably doing). You can choose a rectangular window, make the fft length an exact multiple of the signal period and you will get a perfect peak at a single bin with the correct amplitude. But rectangular windows suffer from a lot of spectral leakage.

Best,
Fabian

PS: Please always (also) respond to the mailing list, so that others can find the discussion as well :)

Am 30.03.22 um 09:47 schrieb Alexander Plotnikov:
Greetings Fabian!

Thank you very much for your response.
I was talking about the main harmonic - the one I am generating with sin source.
I have attached a screen to be more exact.

What I am expecting is to get the spectrum of a pure sin with the Qt Freq Sink, For now the problem seems to be due to Qt Freq Sink small number of FFT points.

Maybe you can advice me with any other Gui spectrum block? As you have probably noticed - I am very new to GNU Radio.

Thanks in advance,

Alexander

On 30.03.2022 10:19, Fabian Schwartau wrote:
Hi Alexander,

I see exactly what is shown in the tutorial and I don't see any harmonics. There are a two "shoulders" left and right to the signal, which are caused by the window function, which is applied by the Frequency sink to reduce spectral leakage. When I increase the lower boundary of the Frequency sink I can also see a lot of "noise" around -180dB, which is caused by numerical instability. Everything as expected... or did I get your question wrong?

Best,
Fabian

Am 30.03.22 um 08:19 schrieb Alexander Plotnikov:
Greetings everyone!

Can anybody please explain to me why in this example (https://wiki.gnuradio.org/index.php?title=Your_First_Flowgraph) the spectrum graph shows a harmonic far below 0dB?

Thanks everyone in advance!
Alexander








reply via email to

[Prev in Thread] Current Thread [Next in Thread]