0
$\begingroup$

So, I have 2 types of signals - Dirac: $\left[ 1,0,0,0,0..NumberOfSamples \right]$ and Gauss (it's sinus * window_gausian).

I need to do some operation in Frequency Domain and use the function: impulseFreqDomain = np.fft.fft(Dirac)

Results in one constant line with 1s. Ok, good. With Gauss impulse, it works too 2 frequency in a different part. Good.

Next Operation is multiplication with some filter $\left[1, 1, 1, 1, 0.99, 0.98,...NumberOfSamples\right]$

It's easy: I use my impulseFreqDomain and Vector of value my filter in frequency domain, then multiply them.

And next step is inverse Fourier transform. I use ifft, but the result is not what I am expecting. I tried with fftshift and np.real\np.abs and this seems not what I want (it must be small than the original signal with time-shift).

And if I try to increase accuracy of my result by the way of increasing sampling frequency, it has an influence on an amplitude of the output signal.

$\endgroup$

Your Answer

By clicking "Post Your Answer", you acknowledge that you have read our updated terms of service, privacy policy and cookie policy, and that your continued use of the website is subject to these policies.

Browse other questions tagged or ask your own question.