Technical discussion about Matlab and issues related to Digital Signal Processing.
i want to compute the ifft of the following data data=[0 1 1 0] in matlab it should return me a single value but it is giving me a column of values can anybody explain this to me and secondly how matlab carries out the ifft , how can it be done by hand to verify the result thanks
Your question is a bit confuse. It is better that you analyze how the fft/ifft works. You can use Google e.g. This is a numerical implement*ation of the Fourier Transform. To work correctly, you have to respect the Nyquist Theorem and the input vector needs a number of points (typically a power of 2 like 10 , 20) to define the wave form. In output the fft produce a vector of length equal to that in input Bye * On Sun, Jun 29, 2008 at 8:05 AM, edu_forme <e...@yahoo.com> wrote: > i want to compute the ifft of the following data > > data=[0 1 1 0] > > in matlab it should return me a single value but it is giving me a > column of values can anybody explain this to me and secondly how matlab > carries out the ifft , how can it be done by hand to verify the result > thanks > > >
Hi,
Your assumption that the ifft should give u only 1 value, is wrong. You will 4 values; remember
IFFT/FFT is a transform operation you will get as many ifft points as fft points provided you
do not zero-pad or clip your data. [0.5000, -0.2500 + 0.2500i, 0, -0.2500 - 0.2500i] is the
correct IFFT of your data points. Maybe if I know some more details on what you are trying to
do, I can provide more information.
Amit Pathania
--- On Sun, 6/29/08, arkkimede <a...@gmail.com> wrote:
From: arkkimede <a...@gmail.com>
Subject: Re: [matlab] ifft in matlab
To: "edu_forme" <e...@yahoo.com>
Cc: m...@yahoogroups.com
Date: Sunday, June 29, 2008, 10:55 PM
Your question is a bit confuse. It is better that you analyze how the fft/ifft
works.
You can use Google e.g. This is a numerical implementation of the Fourier Transform.
To work correctly, you have to respect the Nyquist Theorem and the input vector needs
a number of points (typically a power of 2 like 10 , 20) to define the wave form.
In output the fft produce a vector of length equal to that in input
Bye
On Sun, Jun 29, 2008 at 8:05 AM, edu_forme <edu_forme@yahoo. com> wrote:
i want to compute the ifft of the following data
data=[0 1 1 0]
in matlab it should return me a single value but it is giving me a
column of values can anybody explain this to me and secondly how matlab
carries out the ifft , how can it be done by hand to verify the result
thanks