Scribd
Upload
3 views

DFT

The document discusses the Discrete Fourier Transform (DFT), which transforms discrete data between time and frequency domains, allowing for the extraction of signal information. It covers the computation of DFT and its inverse (IDFT), examples of their application, and properties such as linearity, time shifting, circular convolution, and cross-correlation. Additionally, it presents matrix representations for DFT and IDFT computations.

Uploaded by

2105019
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
3 views

DFT

The document discusses the Discrete Fourier Transform (DFT), which transforms discrete data between time and frequency domains, allowing for the extraction of signal information. It covers the computation of DFT and its inverse (IDFT), examples of their application, and properties such as linearity, time shifting, circular convolution, and cross-correlation. Additionally, it presents matrix representations for DFT and IDFT computations.

Uploaded by

2105019
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 23

Lecture 9

ELE 301: Signals and Systems

Discrete Fourier Transform

Cu↵ (Lecture 9) ELE 301: Signals and Systems Fall 2011-12 1 / 16


Signals & Systems Lecture Eight

Discrete Fourier Transform


The Discrete Fourier Transform (DFT) is the equivalent of the continuous Fourier
Transform for signals known only at N instants separated by sample times T (i.e.
a finite sequence of data). The transformation of discrete data between the time and
frequency domain is quite useful in extracting information from the signal.
The DFT is denoted by X(k) and given as,

…………………… (1)

Here X(k) is the DFT and it is computed at k=0,1,2,…. N-1. "N" discrete points. Thus
DFT X(k) is the sequence of N samples. The sequence x(n) can be obtained back
from X(k) by taking Inverse Discrete Fourier Transform, i.e. IDFT is given as,
…………………. (2)

Here x(n) is sequence of N samples . thus X(k) and x(n) both contains N number of
samples.
Let us define,
This is called twiddle factor. Hence DFT and IDFT equation can be written as,
………………… (3)

And
…………………… (4)
Signals & Systems Lecture Eight

Example One

Calculate the four-point DFT of the aperiodic sequence x[k] of length N = 4,

which is defined as follows:

Solution
Signals & Systems Lecture Eight

Fig. 1. (a) DT sequence x[k];(b) magnitude spectrum and


(c) phase spectrum for example one

Example Two
Signals & Systems Lecture Eight

‫ث‬
Signals & Systems Lecture Eight

Example Three
Calculate the IDFT of
Solution
Signals & Systems Lecture Eight

Matrix Multiplication

An alternative representation for computing the DFT is matrix multiplication.


The above DFT and IDFT are obtained by putting in equation (1)

and equation (2). Let us represents sequence x(n) as


vector of N samples

And X(k) can be represented as a vector of N samples

The values of can be represented as a matrix of size N × N as follows:

…………………. (5)

Here the individual elements are written as with "k" rows and "n" columns. Then N –
point DFT of equation (3) can be represented as

…………………. (6)

Similarly IDFT of equation (4) can be expressed in matrix form as,


Signals & Systems Lecture Eight

Or in other expression matrix vector format are given by:

Similary, the expression for IDFT given by:

Periodicity property of
Let us see the values of for N=8.
We know that is given as,

With N=8 above equation becomes

Table below shows values of


Signals & Systems Lecture Eight

The values of these phasors are observe that,

Example Four
Calculate the four-point DFT of the aperiodic signal x[k] considered in Example one.
Solution
Signals & Systems Lecture Eight

Example Five
Calculate the inverse DFT of X[r] considered in Example two.

Solution
Arranging the values of the DFT coefficients in the DFT vector X, we obtain
X = [5 3 − j2 −3 3+ j2].

The above values for the DT sequence x[k] are the same as the ones obtained
in Example two.

Properties of the DFT


 Linearity
If x1[k] and x2[k] are two DT sequences with the following M-point DFT pairs:
x1[k] X1[r ] and x2[k] X2[r ],

then the linearity property states that


a1x1[k] + a2x2[k] a1X1[r ] + a2X2[r .

for any arbitrary constants a1 and a2.


 Time shifting
Signals & Systems Lecture Eight

If x[k] X[r ], then

x[k−k0] X[r]
for an M-point DFT and any arbitrary integer k0.

Circular convolution
If x1[k] and x2[k] are two DT sequences with the following M-point DFT pairs:
x1[k] X1[r ] and x2[k] X2[r ],

then the circular convolution property states that


x1[k] x2[k] X1[r ]X2[r ] (12.27)

and
x1[k]x2[k] [X1[r] X2[r ]],

where denotes the circular convolution operation. Note that the two sequences
must have the same length in order to compute the circular convolution.
Find the circular convolution between
x[n]=[1,2,3,4]
x[n]=[4,3,2,1]
Signals & Systems Lecture Eight

Cross Correlation Property

Cross-correlation between two discrete signals measures the similarity between


them as a function of the time shift applied to one of the signals. It helps identify how
well one signal matches with another when one of them is shifted by various
amounts. If x(n) and y(n) ae two discrete signals, cross coreelation can be
calculated :
r(n)= IDFT(X(k)Y*(k))

In r(n) , find the index that has the highest positive amplitude. That index represents
the shift amount. In time domain,

r(n) is the circular convolution of x(n) and y*(-n)

You might also like