Thread Subject: fft and windows

Hi all,

I've got two questions for you:
1. Using the fft(x) function, if you don't indicate the N-point, what will MATLAB assume the N point is?

2. Using the window(@wname, N), is it possible to shift it up to the points of interest you are choosing? It looks like by default, it starts with 0 and ends with N. I would just like to window a portion of my fft.

Thanks for your time!!

On 31 Okt, 00:46, "Judy " <sauwen...@gmail.com> wrote:
> Hi all,
>
> I've got two questions for you:
> 1. =A0Using the fft(x) function, if you don't indicate the N-point, what =
will MATLAB assume the N point is?

I was going to reply that "This is mentioned in the FFT documentation"
but - surprisingly - it isn't (R2006a).

When you call FFT with an N-length vector x as argument,

X =3D fft(x);

the function returns the N-point FFT.

> 2. =A0Using the window(@wname, N), is it possible to shift it up to the p=
oints of interest you are choosing? =A0It looks like by default, it starts =
with 0 and ends with N. =A0I would just like to window a portion of my fft.

You window the argument to the FFT. If you want
to compute the N-point DFT of a small section of
signal, you do something like

w =3D ... % some N-point window
X =3D fft(x(100:100+N).*w);

Rune

"Rune Allnor" <allnor@tele.ntnu.no> wrote in message
news:beecad17-3bd3-4128-b76d-ea3d62f4413d@v16g2000prc.googlegroups.com...
> On 31 Okt, 00:46, "Judy " <sauwen...@gmail.com> wrote:
> > Hi all,
> >
> > I've got two questions for you:
> > 1. Using the fft(x) function, if you don't indicate the N-point, what
> > will MATLAB assume the N point is?
>
> I was going to reply that "This is mentioned in the FFT documentation"
> but - surprisingly - it isn't (R2006a).
>
> When you call FFT with an N-length vector x as argument,
>
> X = fft(x);
>
> the function returns the N-point FFT.

You're right, Rune. The documentation doesn't mention this explicitly, but
it does imply that N is the size of the input in the appropriate dimension
(for a vector, it's the length; for an array it's the size in the first
nonsingleton dimension.) I'll put in a request for that to be clarified in
the documentation.

--
Steve Lord
slord@mathworks.com

Coool. Thanks for clarifying!

Judy