EECE 301 Signals & Systems Prof.

Mark Fowler
Note Set #23
D-T Signals: DTFT Details Reading Assignment: Section 4.1 of Kamen and Heck

1/17

Course Flow Diagram

The arrows here show conceptual flow between ideas. Note the parallel structure between the pink blocks (C-T Freq. Analysis) and the blue blocks (D-T Freq. Analysis). New Signal Models
Ch. 1 Intro
C-T Signal Model Functions on Real Line System Properties LTI Causal Etc D-T Signal Model Functions on Integers

Ch. 3: CT Fourier Signal Models

Fourier Series Periodic Signals Fourier Transform (CTFT) Non-Periodic Signals

Ch. 5: CT Fourier System Models

Frequency Response Based on Fourier Transform

Ch. 6 & 8: Laplace Models for CT Signals & Systems

Transfer Function

New System Model

New System Model

Ch. 2 Diff Eqs

C-T System Model Differential Equations D-T Signal Model Difference Equations Zero-State Response Zero-Input Response Characteristic Eq.

Ch. 2 Convolution
C-T System Model Convolution Integral D-T System Model Convolution Sum

New System Model

New Signal Model Powerful Analysis Tool

Ch. 4: DT Fourier Signal Models

DTFT (for Hand Analysis) DFT & FFT (for Computer Analysis)

Ch. 5: DT Fourier System Models

Freq. Response for DT Based on DTFT

Ch. 7: Z Trans. Models for DT Signals & Systems

Transfer Function

New System Model

New System Model2/17

Sect 4.1 continued: The Details

Define the DTFT:

X () =

n =

x[n]e jn
radians

In rad/sample

Compare to CTFT:

X ( ) = x(t )e

jt

dt

In rad/sec

radians

Very similar structure so we should expect similar properties!!!

3/17

Example of Analytically Computing the DTFT

x[n]
n -3 -2 -1 1 2 3 4 5 6 q=3

With your brain, not a computer

0, n < 0 x[n] = a n , 0 n q 0, n > q

If a < 1, x[n ] decays If a > 1, x[n ] " explodes" If a < 0, x[n ] oscillates

Given this signal model, find the DTFT. By definition: X () =

n =

x[n]e

jn

= a e
n =0

n jn

= (ae j ) n
n =0

1 (ae ) X () = 1 ae j

j q +1

General Form for Geometric Sum:

r q1 r q2 +1 r = 1 r n = q1
q2 n

4/17

Characteristics of DTFT
1.Periodicity of X() X() is a periodic function of with period of 2

X ( + 2 ) = X ()

Recall pictures in notes of DTFT Intro: Note: the CTFT does not have this property

|X()| is periodic with period 2

X() is periodic with period 2

2. X() is complex valued (in general)

X () = x[n]e jn
n

complex

Usually think of X() in polar form:

X () = X () e jX ( )
magnitude

phase

Same as CTFT
5/17

3. Symmetry If x[n] is real-valued, then:

X ( ) = X ()
X () = X ()

(even symmetry) (odd symmetry)

Same as CTFT Inverse DTFT Q: Given X() can we find the corresponding x[n]? A: Yes!!

1 x[n] = 2

X ()e jn d
We can integrate instead over any interval of length 2 because the DTFT is periodic with period 2

6/17

Generalized DTFT Periodic D-T signals have DTFTs that contain delta functions Example:

x[n ] = 1, n

2 (), < < X () = periodic, elsewhere

With a period of 2

2
3 2

X()

Main part

Another way of writing this is:

X () = 2

k =

( k 2 )
7/17

How do we derive the result? Work backwards!

1 x[n] = 2
1 = 2

X ()e jn d
2 ()e jn d

Sifting property

=e

jn0

=1

8/17

Transform Pairs: Like for the CTFT, there is a table of common pairs (See Web) Be familiar with them
Table 3.2 Table 4.1

Compare and contrast them with the table Of common CTFTs

Careful here the books table doesnt have this subscript see next slide.

9/17

DTFT of a Rectangular Pulse (Ex. 4.3)

1, Define : D - T pulse as pq [n ] = 0,

n = q, ,1, 0, 1, , q otherwise
Use DTFT Tables on my Website Use Geometric Sum Result see Eq. (4.5)

Book doesnt use this subscript!

So, by DTFT definition:

Pq () =

n= q

e jn

e jq e j ( q +1) sin{( q + 1 / 2)} Pq () = = j 1 e sin{ / 2}

See book for details
10/17

Properties of the DTFT (See table on my website)

Like for the CTFT, there are many properties for the DTFT. Most are identical to those for the CTFT!! But Note: Summation Property replaces Integration There is no Differentiation Property Most important ones: -Time shift -Multiplication by sinusoid Three flavors -Convolution in the time domain -Parsevals Theorem Compare and contrast these with the table of CTFT properties

11/17

Table 3.1

Table 4.2

This one has no equivalent on CTFT Properties Table See next example

Use the Tables on my Web Site!!!

It provides a way to use a CTFT table to find DTFT pairs here is an example
12/17

Example 4.7: Finding a DTFT pair from a CTFT pair X() -2 - -B 2

Books picture is not quite correct the B is in the wrong place

Say we are given this DTFT and want to invert it The four steps for using Relationship to Inverse CTFT property are: 1. Truncate the DTFT X() to the - to range and set it to zero elsewhere 2. Then treat the resulting function as a function of call this ()

() = X()p2()
() = X()p2() -2 - -B B 2
13/17

3. Find the inverse CTFT of () from a CTFT table, call it (t) From CTFT table:
(t ) =
B sinc t B

4. Get the x[n] by replacing t by n in (t)

x[n ] = (t ) t =n =

B sinc n B

14/17

Example of DTFT of sinusoid

x[n ] = cos(0n )
Note that:

X () = ?
So use the mult. by sinusoid property

x[n] = 1 cos( 0 n)
From DTFT Table

= y[ n ] = 1

Y () = 2

k =

( 2k )
k=2 term

Y()
k = -2 term k = -1 term k=0 term k=1 term

-4

-3 -2

2 (), < < Another way of writing this: Y () = 2 periodic elsewhere

15/17

Recall:

x[n] = 1 cos( 0 n)

so we can use the mult. by sinusoid result

1 X () = [Y ( + 0 ) + Y ( 0 )] 2
Using the second form for Y() gives:

[ ( + 0 ) + ( 0 )], X () = 2 periodic elsewhere

Orusing the first form for Y() gives:

< <

mult. by sinusoid property says we shift up & down by 0

Y () =

k =

[ ( +

2k ) + ( 0 2k )]
16/17

To see this graphically: Y()

2 (), < < Y () = 2 periodic elsewhere

-4 -3 -2 - 2 3 4

[ ( + 0 ) + ( 0 )], X () = 2 periodic elsewhere

< <

X()

Red comes from Up-shifted Y() Blue comes from Down-shifted Y()

-4 -3 -2 - -0 0 2 3 4
17/17