that's a pretty good application of the spectrum, I agree. One could certainly modify the freq_sink to do that, however, as it is now, the PSD calculation (based on the fft result) is done in a single VOLK kernel, 32fc_s32f_x2_power_spectral_density_32, which probably has some performance advantages, so changing that would mean to either abandon that benefit or introduce a new "processing path" inside the frequency sink.
I'm a bit confused, though: The DFT is a linear operation. So averaging k FFT vectors (linear operation) before or after the DFT wouldn't make a difference, because
, with 
being our
DFT length-sized input sample vectors.You should be able to do
with a stream to vector->add block combination in front
of the normal frequency sink.Cheers,
Marcus
On 05.11.2015 20:26, madengr wrote:
Could this be modified to, optionally, do the (complex) average after the FFT. That would give a coherent average assuming no samples are dropped prior to FFT. Nice for digging signals out of the noise compared to a normal spectrum analyzer type average. Lou Marcus Müller-3 wroteBecause I find that highly cool, here's what's happening inside the Qt frequency sink, even if that might be unrelated to your problem: * takes in a vector of N samples * Computes a discrete Fourier transform (using FFTw's fast fourier transform algorithms) of length N * computes the square of the magnitude, the logarithm of that, normalizes each of the N complex result values * optionally: averages each bin individually * regularly updates the plot with the N-vector of PSD-- View this message in context: http://gnuradio.4.n7.nabble.com/Frequency-discriminator-using-a-frequency-to-voltage-converter-tp56739p56775.html Sent from the GnuRadio mailing list archive at Nabble.com. _______________________________________________ Discuss-gnuradio mailing list address@hidden https://lists.gnu.org/mailman/listinfo/discuss-gnuradio