Chapter 2.

Fourier Representation
of Signals and Systems
2.0 Preliminaries

Frequency-Domain Description of Signals

• Sinusoidal wave
𝑔 𝑡 𝑎 ⋅ cos 2𝜋𝑓𝑡 𝜃
• Frequency, amplitude (or magnitude), phase shift

• Frequency is one of the most important resources in

communication systems.

2
2.0 Preliminaries

Frequency-Domain Description of Signals

• Example: Sonic wave

3
2.1 The Fourier Transform

• Definitions
– Fourier transform of the signal g (t ) : analysis equation

G( f )  

g (t ) exp( j 2ft )dt (2.1)

– Inverse Fourier transform : synthesis equation


g (t )  

G ( f ) exp( j 2ft )df (2.2)

– Only for energy signals

• Notations
–   2f [rad / s ]
– G ( f )  F[ g (t )] (2.3)
g (t )  F1[G ( f )] (2.4)
g (t )  G ( f ) (2.5)
4
2.1 The Fourier Transform

• The Fourier transform G(f) is a complex function of

frequency f
G ( f )  G ( f ) exp[ j ( f )] (2.6)
G ( f ) : continuous amplitude spectrum of g(t)
 ( f ) : continuous phase spectrum of g(t)

• The spectrum of a real-valued signal

– G ( f )  G * ( f ) : complex conjugate
– G ( f )  G ( f ) : even function
–  ( f )   ( f ) : odd function

5
2.1 The Fourier Transform

• Example 2.1: Rectangular pulse

1 1
1, 𝑡
rect 𝑡 2 2
1 1
0, 𝑡 or 𝑡
2 2

t
A rect   ATsinc(fT)
T 

6
2.1 The Fourier Transform

– Sinc function
sin( )
sinc( ) 


7
2.1 The Fourier Transform

• Example 2.2: Exponential pulse

𝑔 𝑡 exp 𝑎𝑡 , 𝑡 0
𝑔 𝑡 exp 𝑎𝑡 , 𝑡 0

8
2.2 Properties of the Fourier Transform

1. Linearity c1 g1 (t )  c2 g 2 (t )  c1G1 ( f )  c2G2 ( f ) (2.14)

(Superposition)
1 f
2. Dilation g (at )  G   (2.20)
a a

3. Conjugation Rule g * (t )  G * ( f ) (2.22)

4. Duality G (t )  g ( f ) (2.24)

5. Time Shifting g (t  t0 )  G ( f ) exp( j 2ft0 ) (2.26)

6. Frequency Shifting exp( j 2f c t ) g (t )  G ( f  f c ) (2.27)


7. Area Under g(t)
8. Area Under G(f)

g (t )dt  G (0) (2.31)

g (0)   G ( f )df (2.32)


9
2.2 Properties of the Fourier Transform

9. Differentiation in the Time Domain d

g (t )  j 2fG ( f ) (2.33)
dt

1

t
10. Integration in the Time Domain g ( )d  G (t ) (2.41)
 j 2f

11. Modulation Theorem

g1 (t ) g 2 (t ) 



G1 ( )G2 ( f   )d (2.49)
12. Convolution Theorem
13. Correlation Theorem


g1 ( ) g 2 (t   )d G1 ( f )G2 ( f ) (2.51)


g1 (t ) g 2* (t   )dt G1 ( f )G2* ( f ) (2.53)

 
14. Rayleigh’s Energy Theorem
 g (t ) dt   G ( f ) df
2 2
(2.55)
 

10
2.2 Properties of the Fourier Transform

• Example 2.3: Double exponential pulse (symmetric and

odd-symmetric)

2a
exp( a | t |) 
a 2  (2f ) 2

 j 4f
exp(a | t |) sgn(t ) 
a 2  (2f ) 2

11
2.2 Properties of the Fourier Transform

• Example 2.4: Sinc pulse

A  f 
Asinc( 2Wt)  rect 
2W  2W 

12
2.2 Properties of the Fourier Transform

• Example 2.5: Radio frequency pulse

t
rect  cos(2f c t )
T 
 sinc[T ( f  f c )]  sinc[T ( f  f c )]
T
2

13
2.2 Properties of the Fourier Transform

• Example 2.9: Sinc pulse (continued)

g (t )  Asinc(2Wt )

g (t ) dt   A  sinc 2 (2Wt )dt 


 
E
2 2
   
2
 A  
2 f 
 
 2W 
  2W df
rect
2 2
 A  W  A 

 2W


W  2W 
df  (2.57)

14
2.3 The Inverse Relationship Between Time
and Frequency

• If the time-domain description of a signal is changed,

the frequency-domain description of the signal is
changed in an inverse manner, and vice versa.

• If a signal is strictly limited in frequency, the time-

domain description of the signal will trail on
indefinitely, even though its amplitude may assume a
progressively smaller value.
– A signal cannot be strictly limited in both time and frequency.

15
2.3 The Inverse Relationship Between Time
and Frequency
• Bandwidth
– A measure of extent of the significant spectral content of the signal for
positive frequencies.

• Commonly used three definitions

– Null-to-null bandwidth
• When the spectrum of a signal is symmetric with a main lobe bounded by
well-defined nulls, we may use the main lobe as the basis for defining the
bandwidth of the signal.
– 3-dB bandwidth
• Low-pass type : The separation between zero frequency and the positive
frequency at which the amplitude spectrum drops to 1 / 2 of its peak value.
• Band-pass type : The separation between the two frequencies at which the
amplitude spectrum of the signal drops to 1 / 2 of the peak value at fc.
– Root mean-square (rms) bandwidth
• The square root of the second moment of a properly normalized form of the
squared amplitude spectrum of the signal about a suitably chosen point.
• It lends itself more readily to mathematical evaluation than the other two
definitions of bandwidth.
• Although it is not as easily measured in the lab.

16
2.4 Dirac Delta Function

• The theory of the Fourier transform is applicable only


to time functions including energy signal (  g (t ) dt   )
2

that satisfy the Dirichlet conditions.

• However, it is desirable to extend the theory in two
ways.
– To combine the theory of Fourier series and Fourier transform
into a unified framework, so that the Fourier series may be
treated as a special case of the Fourier transform
– To expand applicability of the Fourier transform to include
power signals, that is, signals for which the condition holds.
1 T

2
lim
T  2T
T
g (t ) dt  

– ex) g(t)=1.0
17
2.4 Dirac Delta Function

• Dirac delta function (Unit impulse function)

– Having zero amplitude everywhere except at t=0, where it is
infinitely large in such a way that it contains unit area under
its curve.
1.   0   
 (t )  0, t  0 (2.61)
 2.   t   0, t  0


 (t )dt  1 (2.62) 
  t  dt  1
3.  

4.   t     t 

– Sifting property : 
g (t ) (t  t0 )dt  g (t0 ) (2.63)


– Replication property :

g ( ) (t   )d  g (t ) ( 2.64)
g (t )   (t )  g (t )
– F [ (t )]  1
 (t )  1 (2.65)
18
2.4 Dirac Delta Function

19
2.4 Dirac Delta Function

• Applications of the Delta Function

1. DC signal
• By applying the duality property to the Fourier transform pair of
Eq.(2.65)
1   ( f ) (2.67)
• A DC signal is transformed in the frequency domain into a delta
function occurring at zero frequency



exp( j 2ft )dt   ( f )



cos(2ft ) dt   ( f ) (2.68)

2. Complex exponential function

• By applying the frequency-shifting property to Eq. (2.67)
exp( j 2f c t )   ( f  f c ) (2.69)

20
2.4 Dirac Delta Function

3. Sinusoidal functions
1
cos(2f c t )  [exp( j 2f c t )  exp( j 2f c t )] (2.70)
2
1
cos(2f c t )  [ ( f  f c )   ( f  f c )] ( 2.71)
2
1
sin( 2f c t )  [ ( f  f c )   ( f  f c )] (2.72)
2j

21
2.5 Fourier Transform of Periodic Signals

 

 g (t  mT )  f  G(nf ) ( f  nf )
m  
0 0
n  
0 0
(2.88)

Periodicity in the time domain has the effect of changing the

spectrum of a pulse-like signal into a discrete form defined at
integer multiples of the fundamental frequency, and vice versa.

22
2.5 Fourier Transform of Periodic Signals

• Example 2.11: Ideal sampling function

 

  (t  mT )  f   ( f  nf
m  
0 0
n  
0 ) (2.90)

23
 Review: Convolution

• The output of LTI system for an arbitrary input

Impulse
Input Output
response
x(t) y(t)
h(t)

 
  x  h t     y  t   x  t   h  t    x  h  t    d



ℎ 𝜏 𝑥 𝑡 𝜏 𝑑𝜏

24
2.6 Transmission of Signal Through Linear
Systems : Convolution Revisited

• Frequency Response 𝐻 𝑓
– Consider the impulse response h(t) of a linear time-invariant
system.
– Input and output signal

y (t )   h( ) x(t   )d  h(t ) * x(t )

– By convolution theorem (property 12),
Y ( f )  H ( f ) X ( f ) ( 2.109)
– x(t )  exp( j 2ft )
 
⇒ y (t )   h( ) x(t   )d   h( ) exp[ j 2f (t   )]d

 exp( j 2ft )  h( ) exp( j 2f )d


 H ( f ) exp( j 2ft )

25
2.7 Ideal Low-Pass Filters

• Filter: A frequency-selective system that is used to limit the

spectrum of a signal to some specified band of frequencies.
• Conditions for the frequency response of an ideal low-pass
filter
– The amplitude response of the filter is a constant inside the
passband, that is, -B ≤ f ≤ B where B is the bandwidth of the filter.
– The phase response varies linearly with frequency inside the
passband of the filter.
exp( j 2ft0 ),  B  f  B
H( f )   (2.116)
0, f  B

26
2.7 Ideal Low-Pass Filters

• Evaluating the inverse Fourier

B
h(t )   exp[ j 2f (t  t0 )]df (2.117)
B
transform of the transfer sin[ 2B (t  t0 )]

function of Eq. (2.116)  (t  t0 )
 2 Bsinc[2 B(t-t0 )] (2.118)

• Sinc function is noncausal.

• We can build a causal filter

that approximates an ideal
low-pass filter with delay t0.
sinc[2 B(t-t0 )]  1, for t  0

27
2.7 Ideal Low-Pass Filters

• Gibbs phenomenon

28
2.8 Correlation and Spectral Density
: Energy Signals

• The autocorrelation function of an energy signal x(t) is

defined as 
Rx ( )   x(t ) x* (t   )dt


• A measure of the similarity between the signal x(t )

and its delayed version x(t   ) .
• Rx (0) is equal to the energy of the signal x(t ) .

Rx (0)   x(t ) dt
2


29
2.8 Correlation and Spectral Density
: Energy Signals

30
2.8 Correlation and Spectral Density
: Energy Signals

• Energy spectral density

x  f   X  f 
2

– The energy spectral density is a nonnegative real-valued

quantity for all f, even though the signal x(t) may itself be
complex valued.
• Wiener-Khintchine Relations for Energy Signals
– The autocorrelation function and energy spectral density form
a Fourier-transform pair.

 x ( f )   Rx ( ) exp( j 2 f  )d


Rx ( )    x ( f ) exp( j 2 f  )df


– Next two relations hold.



Rx ( )d   x (0)


 x ( f )df  Rx (0) 31
2.8 Correlation and Spectral Density
: Energy Signals

• Example 2.13: Autocorrelation of sinc pulse

– x(t )  Asinc(2Wt )
A  f 
X(f )  rect 
2W  2W 
2
 A   f 
 x( f )    rect 
 2W   2W 
A2
Rx ( )  sinc2W 
2W

32
2.9 Power Spectral Density

• The average power of a signal is

1

T 2
P  lim x(t ) dt (2.147)
T 
2T T

– Power signal : P  
– Truncated version of the signal x(t)
 t   x(t ),  T  t  T
xT (t )  x(t )rect 
 2T  0, otherwise (2.148)
1 


2
P  lim xT (t ) dt (2.149)
T 
2T 

– By Rayleigh energy theorem,

Power spectral density
1 
T  2T  
2
P  lim X ( f ) df 1 2
S x ( f ) : lim
T
XT ( f )
T  2T
  1 2 
   lim X T ( f ) df   S x ( f )df
  T  2T
  
33