Why should I use 2 fftshift?

8 views (last 30 days)
moose
moose on 31 Jul 2015
Commented: moose on 1 Aug 2015
Hello if I have signal x (in time domain).
I saw that sometimes people do fftshift(fft(ffshift(x))). Now, I understand the usage of fftshift to move the DC to the center. But what's the reasoning behind using fftshift twice?
Seems to me that fftshift(fft(x)), should be enough.
Am I missing something here?
Thanks

Answers (1)

Walter Roberson
Walter Roberson on 31 Jul 2015
You can test it yourself to see if the two are equal.
The magnitudes of the two are equal, but the real and imaginary components are different.
  1 Comment
moose
moose on 1 Aug 2015
thank you.
yes, I saw that the subtraction gives me zero.
so, what's the difference? or, what's the more right way to do it?
I can understand my way fftshift(fft(x)). you're shifting the resulting fft signal.
But, what does it mean, fftshift(fft(fftshift(x)))? it seems that you're first doing the shift on x which is in time domain???

Sign in to comment.

Tags

Community Treasure Hunt

Find the treasures in MATLAB Central and discover how the community can help you!

Start Hunting!