pepe100
Newbie level 1
- Joined
- Sep 1, 2013
- Messages
- 1
- Helped
- 0
- Reputation
- 0
- Reaction score
- 0
- Trophy points
- 1
- Activity points
- 11
Hello,
I implemented complex FFT in MATLAB, as described in the book "The Scientist and Engineer's Guide to Digital Signal Processing" by S. Smith.
My implementation generally seems to work, however there are some problems I can't resolve.
I am testing my implementation it with a sinusoidal signal and compare the results with the MATLAB built-in function fft(), then I am using ifft() to reconstruct the signal. The effects I see are:
- using 4 Hz sinusoidal signal, it seems the reconstructed signal has some phase shift which I cannot explain;
- using 50 Hz and 300 Hz signals - the spectrum seems fine;
- when using 250 Hz signal the spectrum seems incorrect in a way I cannot explain.
Please, suggest where to search for the problem and how to track it... it could be something I am missing from the theory, or just something I didn't consider in using MATLAB... attached are reference images...
Thanks in advance
I implemented complex FFT in MATLAB, as described in the book "The Scientist and Engineer's Guide to Digital Signal Processing" by S. Smith.
My implementation generally seems to work, however there are some problems I can't resolve.
I am testing my implementation it with a sinusoidal signal and compare the results with the MATLAB built-in function fft(), then I am using ifft() to reconstruct the signal. The effects I see are:
- using 4 Hz sinusoidal signal, it seems the reconstructed signal has some phase shift which I cannot explain;
- using 50 Hz and 300 Hz signals - the spectrum seems fine;
- when using 250 Hz signal the spectrum seems incorrect in a way I cannot explain.
Please, suggest where to search for the problem and how to track it... it could be something I am missing from the theory, or just something I didn't consider in using MATLAB... attached are reference images...
Thanks in advance
