CAN207 Signals and Systems

Part 2 – Discrete-Time Signals and Systems

Lecture-D4
Discrete-Time Fourier Transform (DTFT)

Zhao Wang
Zhao.wang@xjtlu.edu.cn
Room EE322

1
Content
• Discrete-Time Fourier Transform (DTFT)
– 0. Background
– 1. Definition and importance of DTFT
– 2. DTFT pairs
– 3. Important theorems
– 4. DTFT properties
– 5. Inverse DTFT
– 6. CCLDE represented LTI system in FD

2
0.1 Why introduce the FD analyses?
• Time domain operation are often not very informative
and/or efficient in signal processing.
• An alternative representation and characterization of signals and
systems can be made in transform domain
• Much more can be said, much more information can be extracted
from a signal in the transform / frequency domain.
• Many operations that are complicated in time domain become
rather simple algebraic expressions in transform domain
• Most signal processing algorithms and operations become more
intuitive in frequency domain, once the basic concepts of the
frequency domain are understood.

3
 0.1 Signals in Time Domain

Filter out noises in Analyze female and Analyze the vocal

the signal male voices from different singer

• Time domain analyses are not informative enough in these

examples.
4
0.1 Importance in Physics
• In physics, a signal in transform domain will exhibit more
properties, which facilitates us to implement signal
processing
– Eg: speech signals, image signals, etc.

Speech signal in time domain Speech signal in frequency domain

5
0.1 Importance in Mathematics
• In mathematics, signal processing can be simplified in
transform domain
– Eg: s-transform, z-transform, Fourier transform
– S-transform can make differential equation (of continuous signals) become to
algebraic equation.

– Z-transform can make difference equation (of discrete signals) become to

algebraic equation.

6
 0.1 Frequency domain representation
• The frequency domain
representation of a signal
can be obtained through
Fourier transforms.
sin(2 50t )

sin(2 50t )  sin(2 75t )

sin(2 20t )  4 cos(2 50t )  2 sin(2 100t )

Spectrum: A compact representation of the frequency content

of a signal that is composed of sinusoids
7
0.2 Fourier’s family
• A prism can be used to break up
white light (sunlight) into the
colors of the rainbow.

• Fourier transform is used to

break up signals into the
frequency components.
– Continuous-Time VS Discrete-Time
– Periodic VS Non-periodic
CT Signals DT Signals
Periodic Fourier Series Discrete FS
Non-periodic CTFT DTFT
8
0.2 Fourier WHO?

“An arbitrary function, continuous or with

discontinuities, defined in a finite interval by
an arbitrarily capricious graph can always
be expressed as a sum of sinusoids”
J.B.J. Fourier
December 21, 1807

9
Dirichlet conditions
• Dirichlet conditions: the sufficient conditions for
the existence of Fourier representations of signals
– The signal must have finite number of discontinuities
– The signal must have finite number of extremum points within
its period
– The signal must be absolutely integrable within its period

All periodic signals of practical interest satisfy these conditions

10
• Fourier Series (FS)
– Fourier’s original work: A periodic function can be represented as a finite,
weighted sum of sinusoids that are integer multiples of the fundamental
frequency Ω0 of the signal. These frequencies are said to be harmonically
related, or simply harmonics.
• Continuous Time Fourier Transform (CTFT)
– Extension of Fourier series to non-periodic functions: Any continuous
aperiodic function can be represented as an infinite sum (integral) of
sinusoids. The sinusoids are no longer integer multiples of a specific
frequency.
• Discrete Time Fourier Transform (DTFT)
– Extension of FT to discrete sequences. Any discrete function can also be
represented as an infinite sum (integral) of sinusoids. While time domain is
discretized, frequency domain is still continuous.
• Discrete Fourier Transform (DFT)
– Because DTFT is defined as an infinite sum, the frequency representation is
not discrete. An extension to DTFT is DFT, where the frequency variable is
also discretized.
• Fast Fourier Transform (FFT)
– Mathematically identical to DFT, however a significantly more efficient
implementation. FFT is what signal processing made possible today! 11
 1.1 From CTFT to DTFT
• From CT Signals to DT Signals xa(t) Xa(Ω)
– a DT signal can be
obtained from a CT
signal through the t Ω
process of sampling: p(t) P(Ω)

= ( − )
-Ts 0 Ts t -Ωs 0 Ω
Ωs
xp(t) Xp(Ω)
=

= ( ) = 0 Ts 2Ts 3Ts … t -Ωs 0 Ωs Ω

x[n] X(ω)
= [ ]

= 0 1 2 3 … n -2π 0 12ω
2π
1.2 DTFT Definition
• The discrete-time Fourier transform (DTFT) X(ejω) of a
sequence x[n] is defined by:

– DTFT X(ejω) of a sequence x[n] is a continuous function of ω

• Inverse Discrete-Time Fourier Transform - the Fourier

coefficients {x[n]} can be computed from X(ejω) using

13
1.2 DTFT Definition
• X(ejω) is a complex function with the real variable ω

or

Fourier Magnitude Phase

spectrum spectrum spectrum

• It is usually assumed that the phase function θ(ω) is

restricted to (− , ), but since it’s periodic, it can be
extended to (−∞, ∞)

14
2.1 Important DTFT Pairs (1) Impulse signals

• 1. Impulse Function

ℱ = = 0 =1

• 2. Delayed Impulse Function

ℱ − = −

= − =

15
2.1 Important DTFT Pairs (1) Impulse signals
• 1. The DTFT of the impulse function is “1” over the entire frequency
band.

• 2. Constant Function
ℱ 2 ( −2 )
=ℱ =
1
=2 ( −2 )
– Note that x[n]=1 is not absolute summable; 2
= =1
– But its DTFT still exists: = 2 δ( );

16
2.3 Important DTFT Pairs (3) Complex Exponential
• 3. The complex exponential

– We are only interested in [-π, π] range, where there is only one

spectral component.
– Hence, the spectrum of a single complex exponential at a specific
frequency is an impulse at that frequency.

17
2.4 Important DTFT Pairs (4) Sinusoidal signals

• 4. The sinusoid

– The above expression can also be obtained from the DTFT of the
complex exponential through the Euler’s formula.

18
2.5 Important DTFT Pairs (5) Real exponential

• 5. The real exponential

19
2.1 Important DTFT Pairs (2) Step signals

• 1. Find the DTFT of = − [ − ]

• 2. Find the DTFT of =

20
2.6 Important DTFT Pairs (6) Rectangular pulse train

• 6. Rectangular pulse train

21
3. Important theorems of DTFT
• Theorem 1: DTFT is periodic with 2π.
• Theorem 2: The digital frequency 2π corresponds to the
linear sampling frequency of the signal.
• Theorem 3: DTFT only exists for sequences that are
absolutely summable.

22
3.1 Theorem 1 - Periodicity
• The DTFT of a discrete sequence is periodic with
the period 2π, that is

• The periodicity of DTFT can be easily verified

from the definition:

23
3.2 Theorem 2 - Implications of the periodicity

• The discrete frequency 2π corresponds to the sampling

frequency Ωs used to sample the original continuous signal
x(t) to obtain x[n].

24
3.3 Theorem 3 - Existence of DTFT
• The DTFT of a sequence exists if and only if, the sequence
x[n] is absolutely summable, that is

because

– This is the sufficient condition for the existence of DTFT;

– Certain sequences that do not satisfy this requirement also have
DTFTs, if they satisfy “mean square convergence” :

25
3.3 Example
F -1

• The sequence {x(n)} is not absolutely summable

• But it is mean-square convergent

26
4.1 DTFT Properties (1)
• Linearity
– Given x1[n] and X1(ω) form a DTFT pair, and x2[n] and
X2(ω) form another DTFT pair i.e.
DTFT
[ ] ( )
DTFT
[ ] ( )
– We can show that
DTFT
+ [ ] + ( )
– Example:

27
4.2 DTFT Properties (2)
• Time-reversal: A reversal of the time domain variable
causes a reversal of the frequency variable
DTFT
[− ] (− )

28
4.3 DTFT Properties (3)
• Symmetric
∗[ DTFT ∗ ∗ DTFT ∗(
] (− ) [− ] )

= ∗ (− )
– 1. If x[n] is real:

| |=| − | = (− )
=− − = − (− )
– 2. If = +
DTFT
( )
DTFT
( ) 29
4.4 DTFT Properties (4)
• Time Domain Shifting (TD Delay) => FD Phase Shift
DTFT
[ − ] ( )
– Note that the magnitude spectrum is unchanged by time shift.

• Frequency Domain Shifting => TD Phase Shift

DTFT
[ ] ( − )

• Example: Find DTFT of = cos( + ) [ ]

30
4.5 DTFT Properties (5)
• Differencing in Time
DTFT
− [ − 1] 1− ( )

• Differentiation in Frequency
DTFT ( )
[ ]

31
4.6 DTFT Properties (6) Convolution Property

• Convolution in TD = multiplication in FD
DTFT
∗ ℎ[ ] ( ) ( )

– The convolution property illustrates the system’s response

to input, in TD, the output is the result of convolution, and
in FD, the output is the result of multiplication;
– By designing the ( ) carefully, we can pass certain
frequency components, then make ≅ 1; stop
certain frequency components, i.e. ≅ 0. This is the
concept of FILTERING

32
4.6 DTFT Properties (6) Example
• Consider the system shown in Figure 5.18(a) with input x[n] and output
y[n]. The LTI systems with frequency response H1p(ω) are ideal
lowpass filters with cutoff frequency π/4 and unity gain in the passband.
• Find the overall frequency
response of this system.

33
4.7 DTFT Properties (7) Multiplication Property

• Multiplication in TD = convolution integral in FD

DTFT 1
ℎ[ ] ( ) −
2

– h[n] can be considered as either system impulse response

or another signal;
– This property is also called the modulation property, since
it involves the modulation of one signal x[n] with the other
h[n];

34
4.7 DTFT Properties (7) Example

• Find the Fourier transform ( ) of a signal = [ ] [ ]

where: sin(3 /4)
=
sin( /2)
=

35
4.8 DTFT Properties (8)
• Parseval Theorem: The energy of the signal , whether
computed in TD or FD, is the same!
1
| [ ]| = | ( )|
2
Energy density spectrum
of the signal

∗
1 ∗
[ ]= ( ) ( )
2

36
4.8 DTFT Properties (8)
• Example - Determine and sketch the energy density
spectrum of the signal

• Result:

37
4. Full table of
DTFT properties

A.V. Oppenheim,
Signals and Systems
2nd, p391

38
5. Inverse DTFT Uniqueness

• The DTFT is a unique relationship between x[n] and X(ω).

– Two different signals cannot have the same DTFT.
– If we know a DTFT representation, we can start in either the time or
frequency domain and easily write down the corresponding
representation in the other domain.

• The uniqueness property implies we can always go back and

forth between the time-domain and frequency domain
representations.

• Most frequently used pair:

1
, <1
1− 39
5. Inverse DTFT Examples

• Example 1: Consider an LTI system with impulse

response h[n] and input x[n] given by:

• Find the output y[n].

40
5. Inverse DTFT Partial Fraction Expansion

• A.V. Oppenheim, Appendix A.3

• Suppose, then, that we have a rational function in the form:

• 1. Factorise the denominator

• 2. Write in the form of the partial fraction expansion as:

– where coefficient Bik is:

41
5. Inverse DTFT Example

• Find the inverse DTFT of

42
6. DT systems represented by its DTFT
• A general linear constant-coefficient difference equation for
an LTI system with input x[n] and output y[n] is of the form:

• Convert both sides to frequency domain

• The system in frequency domain is represented by

– The coefficients of the numerator are the same as for x[n];

– The coefficients of the denominator are the same as for y[n].
43
6. DT systems represented by its DTFT Example

• Consider a causal LTI system that is characterized by the

difference equation:
3 1
− −1 + −2 =2 [ ]
4 8

• The input to the system is

1
= [ ]
4

• Find the output in TD and FD.

44
Lect-D4 Practices
• Exercise 1. Determine the frequency response of the following
sequences.
– a) n = −1 , < 1;
– b) n = , < 1;
– c) n = − −1 , < 1;
| |
– d) n = , < 1.

• Exercise 2. Find the inverse DTFT of the following frequency

spectrums:
– a) =1+2 +4 2 ;
/
– b) = 1+2 + 4 2 cos( /2) ;
– c) = 1+2 + 4 2 sin .
/
– d) = 1+2 + 4 2 sin( /2) ;

45
Lect-D4 Practices
• Exercise 3. If ( ) is the DTFT of the sequence =
{3, 1, −4, 0, −5, 2, 1; −4 ≤ ≤ 2}. Calculate the values of following
expressions without calculating the DTFT:
– (a) ( );
– (b) ( );
– (c) ∫ ( ) ;

– (d) ∫ ( ) ;
( )
– (e) ∫ .

46
Lect-D4 Practices
• Exercise 4. The impulse response of an LTI system is given by:
1
ℎ = [ − 2]
3

Find its frequency response ( ).

• Exercise 5. The difference equation of a causal stable LTI system is

given as:
6y k + 5y k − 1 + y k − 2 = x k + x[k − 1]
Find its frequency response ( ) and impulse response h[n].

47