Chapter 10

Z-Transform
 Z-Transform and the Region of Convergence
 Inverse z-Transform
 Properties of z-Transform
 Analysis and Characterization of LTI Systems Using z-Transform
 Unilateral z-Transform
1
Z-Transform and Region of Convergence

The z-transform of a discrete-time signal x[n] is defined as


X ( z)  
n 
x[n]z  n

 z could be any complex number.

 (X(z),x[n]) is called a z-transform pair.
Z
x[n]   X ( z ); Z {x[n]}  X ( z )
 For a discrete-time LTI system with impulse response
x[n], X(z) is the corresponding system function.

2
Example

n
x[n]  a u[n]  X ( z )  ?
<Sol.>
  
X ( z)  
n 
x[n]z n
 
n 
n
a u[n]z n
  (az 1 ) n
n0
1 n
For convergence of X(z), we require that  az   .

n 0

Consequently, the region of convergence is the range of

values of z for which az  1 , or equivalently, z  a .
1


1 z
 X ( z )   ( az )  1 n
1
 , if z  a
n 0 1  az za
3
For example, for a=1, x[n] is the unit step sequence with
z-transform
1
X ( z)  1
, z 1
1 z

Consequently, just as with rational Laplace transforms,

the z-transform can be characterized by its zeros (the
roots of the numerator polynomial) and its poles (the
roots of the denominator polynomial).

1 z
X ( z)  1

1 z z 1
4
 1 z
X ( z)  1
 , if z  a
1  az za

5
Example

n
x[n]  a u[n  1]  X ( z )  ?
<Sol.>
 1
X ( z )    a nu[n  1]z  n    a n z  n
n  n 
 
  a  n z n  1   (a 1 z ) n
n 1 n 0

If a 1 z  1, then
1 1 z
X ( z)  1  1
 1
 ,z a
1  a z 1  az za
6
7
Example
n
1 
x[n]    sin( n)u[ n]  X ( z )  ?
3 4
<Sol.>
1 1 j / 4 n 1 1  j / 4 n
x[n]  ( e ) u[n]  ( e ) u[n]
2j 3 2j 3
  1  1 j / 4 n 1  1  j / 4 
n
  n
 X ( z)     e  u[n]   e  u[n]z
n  2 j  3  2j3  
n n
1   1 j / 4 1  1   1  j / 4 1 
   e z    e
2 j n0  3  2 j n 0  3
z 

1 1 1 1
 
2 j 1  1 e j / 4 z 1 2 j 1  1 e j / 4 z 1
3 3
8
 ROC: (1/ 3)e j / 4 z 1  1, (1/ 3)e j / 4 z 1  1  z  1/ 3

1
z
X ( z)  3 2
1 j / 4 1  j / 4
( z  e )( z  e )
3 3

9
Z-Transform v.s. Discrete-Time Fourier Transform

Consider z  re j
 
 Z {x[n]}  X (re j )  
n 
x[n](re j )  n  x[
n 
n ]r n
e  j n

 F x[n]r  n 

If | z | 1, i.e., z  e j  ROC

X ( z) z  e j
 X (e j )  F x[n]

10
11
 1 z
 Recall X ( z )  1
 , if z  a
1  az za

| a | 1

12
 1 z
X ( z)  1
 , if z  a
1  az za

| a | 1

13
Properties of Region of Convergence

The ROC of the z-transform of x[n] consists of the values

n
of z for which x[n] z is absolutely summable, i.e.,



n 
x[n]z  n  

 Property 1: The ROC of X(z) consists of a ring in the z-

plane centered about the origin.

For z  re j
with 
n 
x[ n] z  n  , we have


 x[n]  re 
j   n
  for all    [0, 2 ]
n 

14
15
 Property 2: The ROC does not contain any poles.

 Property 3: If x[n] is of finite duration, then the ROC is

the entire z-plane, except possibly z = 0 and/or z = ∞.

  [n]    [n]z  1
Z n

n 

ROC: the entire z-plane, including z = 0 and z = ∞.


  [ n  1] Z
   [ n  1]
n 
z n
 z 1

ROC: the entire z-plane, including z = ∞ but excluding

z=0.
16
Example

Consider the signal

a n , 0  n  N  1, a  0
x[n]  
0, otherwise
Then
N 1
X ( z)   a n z n
n 0
N 1
  (az 1 ) n
n 0

1  (az 1 ) N 1 zN  aN
 1
 N 1
1  az z za

17
18
 Property 4: If x[n] is a right-sided sequence, and if the circle is in the
ROC, then all finite values of z for which will also be in the ROC.
z  r0
z  r0

19
If z  r0  ROC
  
X ( z)  
n  N1
x[n]z  n  
n  N1
x[n]z  n  
n  N1
x[n]r0  n  

For z  r1 with r1  r0
  
r0  n 
r n


n  N1
x[n]z  n  
n  N1
x[n]r1 n  
n  N1
x[n]r1 n  
r0  n n  N1
x[ n ]r0
n 1

r0  n
n

 r0  
  x[n]r0 n
    x[ n ]r0
n
< if N1  0
n  N1  r1  n  N1

20
 Property 5: If x[n] is a left-sided sequence, and if the circle
z the
is in  r0ROC, then all values of z for which
0  z  r0will also be in the ROC.

 Property 6: If x[n] is two sided, and if the circle z  r0

is in the ROC, then the ROC will consist of a ring in the
z-plane that includes the circle z  r0 .

21
22
23
Example

n
x[n]  b , b  0

24
x[n]  b nu[n]  b  nu[n  1]

n Z 1 n Z 1 1
b u[n]  1
, z  b; b u[ n  1]  1 1
,z 
1  bz 1 b z b

1 1 1
 X ( z)  1
 1 1
,b  z 
1  bz 1 b z b
b2  1 z 1
 1
,b  z 
b ( z  b)( z  b ) b

25
26
 Property 7: If the z-transform X(z) of x[n] is rational, then its ROC is bounded by poles or extends to
infinity.

 Property 8: If the z-transform X(z) of x[n] is rational, and

if x[n] is right sided, then the ROC is the region in the z-
plane outside the outermost pole, i.e., outside the circle
of radius equal to the largest magnitude of the poles of
X(z). Furthermore, if x[n] is causal (i.e., if it is right sided
and equal to 0 for n < 0), then the ROC also includes z =
∞.

27
 Property 9: If the z-transform X(z) of x[n] is rational, and
if x[n] is left sided, then the ROC is the region in the z-
plane inside the innermost nonzero pole, i.e., inside the
circle of radius equal to the smallest magnitude of the
poles of X(z) other than any at z = 0 and extending
inward to and possibly including z = 0. In particular, if
x[n] is anticausal (i.e., if it is left sided and equal to 0 for
n > 0),then the ROC also includes z = 0.

28
Example

1
X ( z) 
1
(1  z 1 )(1  2 z 1 )
3

29
30
The Inverse z-Transform

For z  re j  ROC, we have X (re j )  F x[ n]r 1

 x[n]r  n  F 1  X (re j )

 x[n]  r n F 1 X (re j )

1 1
 
j j n
r n
X (re )e d   X (re j )(re j ) n d 
2 2 2 2

1


n 1
 x[n]  X ( z ) z dz
2 j

31
Example
5
3  z 1
6 1
X ( z)  , z   x[n]  ?
1 1 1 1 3
(1  z )(1  z )
4 3
<Sol.>
1 2
X ( z)  
1 1 1 1
1 z 1 z
4 3
n n
1 Z 1 1 1 Z 2 1
  u[ n ]   , z  ; 2   u[ n ]   , z 
4
  1 1 4  3  1 1 3
1 z 1 z
4 3
n n
1 1
 x n     u[n]  2   u[ n]
4 3
32
 U ( z) m Ai
 X ( z)   1
 x[n]  ?
V ( z ) i 1 1  ai z

 m
A  m
1  Ai 
 x[n]  Z 
1 i
1 
 Z  1 
 i 1 1  ai z  i 1 1  ai z 
where

1 

Ai   i i u[n],
A a n
if z  ai  ROC
Z  1 
 n
1  ai z   i i u[n  1], if z  ai  ROC
 A a

33
Example

1
X ( z)  1
,z a
1  az

The inverse z-transform can also be obtained by long division.

1 1 2 2 3 3
 1
 1  az  a z  a z  ....
1  az
1
For X ( z )  1
,z a
1  az
1 1 2 2 3 3
 1
  a z  a z  a z  ....
1  az
34
Properties of z-Transform

 Linearity
 Time Shifting/Scaling/Reversal/Expansion
 Scaling in the z Domain
 Conjugation
 Convolution Property
 Differentiation in the z Domain
 Initial-Value Theorem

35
Linearity

z z
x1[n]   X 1 ( z ) with ROC R1 , x2 [n]   X 2 ( z ) with ROC R2

Z
 ax1[n]  bx2 [n]   aX 1 ( z )  bX 2 (z ) with ROC
containing R1  R2

36
Time Shifting

Z
x[n]  X ( z ), with ROC  R
Z
 x[n  n0 ]   z  n0 X ( z )

ROC = R, except for the possible addition or deletion of

the origin or infinity.

37
Scaling in the z-Domain

Z
x[n]   X ( z ), with ROC  R

n Z  z 
 z x[n]  X   , with ROC  z0 R
0
 z0 

38
Time Reversal

Z
x[n]   X ( z ), ROC  R

Z 1 1
 x[n]  X   , ROC 
z R

39
Time Expansion

Z
x[n]   X ( z ), ROC  R

 x[n / k ], if k | n Z
 x( k ) [n]    X ( z k ), ROC  R1/ k
0, otherwise


X ( z)  
n 
x[n]z  n
  
 X (zk )  
n 
x[n]( z k )  n  
n 
x[n]z  kn  
m 
x( k ) [m]z  m

40
Conjugation

Z
x[n]   X ( z ), ROC  R
Z
 x*[n]   X * ( z * ), ROC  R

 If x[n] is real, we can conclude that

X ( z )  X * ( z * ).
41
Convolution Property

Z Z
x1[n]   X 1 ( z ), ROC  R1; x2 [n]   X 2 ( z ), ROC  R2
Z
 x1[n]  x2 [n]   X 1 ( z ) X 2 ( z ) with ROC containing R1  R2

42
Example

Consider an LTI system for which

y[n]  h[n]  x[n]
with h[n]   [ n]   [ n  1].
Z
 [n]   [n  1]  1  z 1 with ROC equal to the
entire z-plane except the origin
Z Z
x[n]   X ( z ), ROC  R  y[n]  (1  z 1 ) X ( z )
with ROC equal to R, with the possible deletion of z = 0
and/or addition of z = 1.
43
Differentiation in the Z-Domain

Z
x[n]   X ( z ), ROC  R
Z dX ( z )
 nx[n]   z , ROC  R
dz

44
Example

X ( z )  log(1  az 1 ), z  a  x[n]  ?

1
Z dX ( z ) az
nx[n]   z  1
,z a
dz 1  az
n Z a
a (a ) u[n]  1
,z a
1  az
1
Z az
a (a ) n 1 u[n  1]   1
,z a
1  az
(  a ) n
 x[n]  u[n  1]
n
45
The Initial-Value Theorem

If x[n]=0, n < 0, then

x[0]  lim X ( z ).
z 


 X ( z )   x[n]z  n  x[0]  x[1]z 1  x[2]z 2  
n 0

 lim X ( z )  x[0]
z 

46
Example

n 1 z
 a u[n]  1
 , if z  a
1  az za

5
n n 3  z 1
1 1 6 1
   u[n]  2   u[n]  1 1
,z 
4  3 (1  z 1 )(1  z 1 ) 3
4 3

nb2  1 z 1
 b  1
,b  z 
b ( z  b)( z  b ) b

47
Summary of Properties

48
Basic z-Transform Pairs

49
Analysis and Characterization of LTI Systems Using
z-Transforms

x[n] h[n] y[n]  x[n]* h[n]

Z
 Y ( z)  H ( z) X ( z)
Recall that

z n
h[n] H ( z) z n

H ( z)  
n 
h[n]z  n

50
Causality

 A discrete-time LTI system is causal if and only if the

ROC of its system function is the exterior of a circle,
including infinity.

 A discrete-time LTI system with rational system function

H(z) is causal if and only if: (a) the ROC is the exterior of
a circle outside the outermost pole; and (b) with H(z)
expressed as a ratio of polynomials in z, the order of the
numerator cannot be greater than the order of the
denominator.

51
Example

1 1
H ( z)   , z 2
1 1 1  2 z 1
1 z
2
5 1 2 5
2 z 2z  z
 H (z)  2  2
1 1 1 2 5
(1  z )(1  2 z ) z  z  1
2 2

 1  n n

 h[n]     2  u[n]
 2  

52
Stability

 An LTI system is stable if and only if the ROC of its

system function H(z) includes the unit circle, z  1 .

 


n 
h[n]    H ( z )  
n 
h[n]z  n with z  1  ROC

 A causal LTI system with rational system function H(z)

is stable if and only if all of the poles of H(z) lie inside
the unit circle, i.e., they must all have magnitude
smaller than 1.

53
Example

n
1 1  1 n

 H ( z)   , z  2  h[n]     2  u[ n]
1 1 1  2 z 1
 2  
1 z
2
n
1 1 1 1 n
 H ( z)   ,  z  2  h[ n ]    u[ n ]  2 u[n  1]
1 1 1  2 z 2
1
2
1 z
2

1 1 1  1  n n 
 H ( z)   , z   h[n]      2  u[ n  1]
1 1
1  z 1 1  2 z 2  2  
2
54
 Consider an LTI system for which the input and output
satisfy a linear constant-coefficient difference equation
of the form
N m

a
k 0
k y[n  k ]   bk x[n  k ].
k 0

N M N M
  ak z  k Y ( z )   bk z  k X ( z )  Y ( z ) ak z  k  X ( z )  bk z  k
k 0 k 0 k 0 k 0

M   N
k 
 H ( z )  Y ( z ) / X ( z )    bk z  /   ak z 
k

 k 0   k 0 
55
Example

Consider an LTI system for which the input x[n] and output
y[n] satisfy the following equation:
1 1
y[n]  y[n  1]  x[n]  x[n  1].
2 3

1 1 1 1
 Y ( z)  z Y ( z)  X ( z)  z X ( z)
2 3
1 1
1 z
Y (z) 3
 H ( z)  
X ( z ) 1  1 z 1
2
56
 For ROC equal to z  1 / 2

n n 1
 1 1  1 1 1 1
H ( z )  1  z   h[n]    u[n]    u[n  1]
 3  1  1 z 1  2 3 2
2

 For ROC equal to z  1 / 2

n n 1
 1 1  1 1 1 1
H ( z )  1  z   h[n]     u[n  1]    u[n]
 3  1  1 z 1 2 3 2 
2
57
Example

Suppose that we are given the following information about

an LTI system with initial rest:
n
1. If the input to the system is 1 x [ n ]  (1 / 6 ) , then the
output is
  1 n   
1
n

y1[n]   a    10    u[n]
  2   3  
n
2. If the input to the system is 2x [ n ]  ( 1) , then the
output is
7
y2 [n]  (1) n
4
 H ( z) ?
58
 1 1
X1 ( z)  ,z 
1 1 6
1 z
6
a 1
(a  10)  (5  ) z
a 10 3 1
Y1 ( z )    ,z 
1 1 1 1 1 1 1 1 2
1 z 1 z (1  z )(1  z )
2 3 2 3

 a 1   1 1 
 (a  10)  (5  ) z  1  z 
Y1 ( z )  3  6 
 H ( z)  
X1 ( z) 1 1 1 1
(1  z )(1  z )
2 3
59
Furthermore, we know that the response to x2 n  (1)
n

n
must equal (1) multiplied by the system function H(z)
evaluated at z = -1. Thus, we see that
 a  7
 ( a  10)  5 
7 3   6 
  H (1)   3 4
 a  9
4 ( )( )
2 3
1 1 1 13 1 1 2 2 13 1
(1  2 z )(1  z ) 1 z  z z  z
 H ( z)  6 or 6 3 or 6 3
1 1 1 1 5 1 1 2 2 5 1
(1  z )(1  z ) 1 z  z z  z
2 3 6 6 6 6
60
From the convolution property, we know that the ROC of
must include at least the intersections of the ROCsXof1 ( z )
Yand
1 ( z) . Examining the three possible ROCSHfor (z )

z  1/ 3, 1/ 3  z  1/ 2, z  1/ 2

we find that the only choice that is consistent with the

ROCs of X 1 ( z ) and Y1 ( z ) is z  1 / 2 .

5 1 13 1
 y[n]  y[n  1]  y[ n  2]  x[n]  x[ n  1]  x[n  2]
6 6 6 3

61
System Functions for Interconnections of LTI Systems

H ( z )  H1 ( z ) H 2 ( z )

H ( z )  H1 ( z )  H 2 ( z )

62
System Functions for Interconnections of LTI Systems

Y ( z) H1 ( z )
 H ( z) 
X ( z) 1  H1 ( z ) H 2 ( z )

63
Example

Consider the causal LTI system with system function

1
H ( z) 
1 1
1 z
4

1
 y[n]  y[n  1]  x[n]
4

64
Example

Consider the system function

7 1 1  2
1 z  z
H ( z)  4 2 .
1 1 1  2
1 z  z
4 8

 
 1   7 1 1 2 
 H ( z)     1 z  z 
1 1
 1  z 1  z 2   4 2 
 4 8 

65
66
 Cascade-form:
 1 1   
 1  4 z   1  2 z 1 
 H ( z)  
1 1   1 1 
 1 z  1 z 
 2  4 
 Parallel-form:

5/3 14 / 3
 H ( z)  4  
1 1 1 1
1 z 1 z
2 4

67
Unilateral z-Transform

The unilateral z-transform of a sequence x[n] is defined as


X ( z )   x[n]z  n .
n0

 x[ n ] UZ
 X ( z )  UZ x[n]
 Original bilateral z-transform:

X ( z)  
n 
x[n]z  n

68
Example

x[n]  a n 1u[n  1]  x[n]  x[n]u[n]

Bilateral z-transform:

z
X ( z)   x[n]z
n 
n

1  az 1
,z a

Unilateral z-transform:
 
a
X ( z )   x[n]z n
 a n 1  n
z  1
,z a
n 0 n0 1  az

69
Properties of the Unilateral z-Transform

70
Example

Consider the causal LTI system described by the

difference equation

y[n]  3 y[n  1]  x[n]

together with the condition of initial rest. The system
function for this system is
1
H ( z) 
1  3 z 1

71
x[n] = αu[n], where α is a given constant

 (3 / 4) (1/ 4)

Y ( z)  H ( z) X ( z)  1 1
 1

(1  3 z )(1  z ) 1  3 z 1  z 1

1  3  
 y[n]       (3) n  u[n]
4  4  

72
Example

Consider the causal LTI system described by the

difference equation

y[n]  3 y[n  1]  x[n]

together with y[1]   .

x[n] =αu[n] => y[n]?

73
 1 
Y ( z )  3  3 z Y ( z ) 
1  z 1
3 
 Y ( z)   1

1  3z (1  3 z 1 )(1  z 1 )

For example, if α = 8 and β = 1,

3 2
 Y ( z)  1

1  3z 1  z 1

 y[n]  3(3) n  2  u[n], for n  0

74