Norhashimah Mohd Saad

Sampling theorem
 Great practical importance in signal processing where
the discrete-time signal is obtained from the samples
of continuous-time signal.
 Sampling rate/ sampling frequency gives a good
condition of information which will be no loss in the
processing.
 Nyquist Theorom
 Sampling theorem
 The sampling process can be considered as multiplication
of a continuous-time signal x(t ) with a periodic train x (t ) of
unit impulses with Ts as the sampling period.

x (t )    (t  nT )
n  
s

Ts  1

  4  3  2 1 0 1 2 3 4 5 
 Sampling theorem
 Frequency spectrum of a time domain signal is
representation of the signal in the frequency domain.

 From Fourier Transform time domain signal will be

used to find the frequency spectrum.
 
X(f )   x(t )e
 j 2ft
dt x(t )   X ( f )e j 2ft df


 Sampling theorem 

 Frequency spectrum for periodic train x (t ) 


  (t  nT )
n  
s
as below
 
 j 2ft
X ( f )  x (t ) e dt


X ( f )  
n  
e  j 2fnTs
 The spectrum of the periodic train is maximum when f=0 &
fs
 f=0 applies maximum value because of power 0 of an
exponent equation.
 f= fs applies maximum value because of fsTs=1,then integer
multiple of sampling frequency always true for maximum
value for frequency spectrum.
Spectrum of continuous-time and
discrete-time signal
 CT signal
Fourier
x(t) X(f)
Transform

 DT signal
Sampling Fourier
x(t) x(n) X(?)
CT signal Transform


x(n)  x(t ) x (t )  x(t )   (t  nTs )
n  
Discrete-time Fourier Transform
 Obtaining the analysis function of Fourier Transform



 j 2ft
X(f )  x (t ) e dt

 The new DTFT’s analysis function (frequency
spectrum)will be derived

as


 
 j 2ft
X (?)  x (t )  (t  nTs )e dt
 n  

X (e j 2f
)  x (
n  
n ) e  j 2fnTs

 DTFT’s generation function 1 fs

x ( n) 
fs  2
f
 s
2
X (e j 2f )e j 2fnTs df
Discrete-time Fourier Transform 

 Discrete-time Fourier Transform X ( e j 2f

)  x (
n  
n ) e  j 2fnTs

can be viewed from original Fourier Transform as


X (e j 2f )   X ( f  nf )
n  
s

fs fs
n2 n0 n 1
Discrete-time Fourier Transform
 Aliasing Effect
fs fs

fs fs

fs fs fm
Discrete-time Fourier Transform
 Nyquist Theorom
 Sampling frequency must be equal or larger than twice
of frequency maximum

fs  2 fm

fs fs fm
Discrete-time Fourier Transform
 Ex 1.
 A continuous-time signal is defined as
x(t )  cos 2f1t 0  t  T
 Step 1: x(n)?
 Step2:apply Euler Identity
 Step3: Find the analysis function of DTFT
 Step4:if needed, sketch the spectrum of DTFT
 Discrete-time Fourier Transform
 The analysis function will be

 e

 1 j 2f1nTs
X (e j 2f
)  (cos 2f1nTs )e
n  
 j 2fnTs
  
n    2
e  e  j 2f1nTs


 j 2fnTs

 e

 1 j 2f1nTs
  
n    2
e  e  j 2f1nTs


 j 2fnTs


1  1 
  X ( f  nf s )   X 1 ( f  ( f1  nf s ))   X 2 ( f  ( f1  nf s ))
n   2 n   2 n  
fs fs

 f s  f1  f s  f1  f1 f1 f s  f1 f s  f1
Discrete-time Fourier Transform
 Characteristic of Discrete-time Fourier Transform
 Time domain
 Discrete in time represented by n
 Non periodic

 Frequency domain
 Continuous in frequency represented by f  e j 2f
 Periodic
Discrete-time Fourier Transform
 Ex 2.
 A discrete-time signal is a unit step function defined as
x[n] 1 n0
 0 elsewhere
 u[n]
 Calculate the magnitude and phase for this analysis
function.
X e  j 2f

1
 1
2 sin fTs

 e j 2f     fTs
2
Discrete-time Fourier Transform
 Ex 3.
 A discrete-time signal is a pulse function defined as
x ( n) 1 0  n  N 1
0 elsewhere
 u ( n)  u ( n  N )
 Calculate the magnitude and phase for this analysis
function.
sin fNTs

X e j 2f
 
sin fTs
 e j 2f   fTs  fNTs
Discrete-time Fourier Transform
 Discrete-time Fourier Transform and Convolution
 The relation is time domain convolution is equal to the
multiplication frequency domain and vice versa.

y ( n)   h( ) x ( n   )
n  

j 2f j 2f j 2f

Y (e )  H (e ) X (e )
Thank you