Lecture 7 – 10

Z - Transform

1
 Z-Transform
 Transform techniques are an important tool in the
analysis of signals and LTI systems.
 Z-transform plays the same role in the analysis of D.T
signals and LTI systems as the Laplace transform does
in the analysis of C.T signals and LTI systems.
 In z-domain convolution of two time domain signals is
equivalent to multiplication of their corresponding z-
transforms. This property simplifies the analysis of
response of LTI systems to various signals.
 Z-transform provides a means of characterizing an LTI
system and its response to various signals by its pole-
zero location.
 From mathematical point of view, z-transform is an
alternative representation of signals. 2
The Direct Z-Transform
The z-transform of a discrete time signal is defined as the
power series

X ( z)   x[ n
n  
] z n

(1)

Where z is a complex variable. For convenience, the z-

transform of a signal x[n] is denoted by
X(z) = Z{x[n]}
Since the z-transform is an infinite series, it exists only for
those values of z for which this series converges. The
Region of Convergence (ROC) of X(z) is the set of all
values of z for which this series converges.
3
We illustrate the concepts by some simple examples.
Example 1: Determine the z-transform of
the following signals
(a) x[n] = [1, 2, 5, 7, 0, 1]
Solution: X(z) = 1 + 2z-1+ 5z-2 + 7z-3 + z-5,
ROC: entire z plane except z = 0

(b) y[n] = [1, 2, 5, 7, 0, 1]

Solution: Y(z) = z2 + 2z + 5 + 7z-1 + z-3

ROC: entire z-plane except z = 0 and z = .

(c) z[n] = [0, 0, 1, 2, 5, 7, 0, 1]

Solution: z-2 + 2z-3 + 5z-4 + 7z-5 + z-7, ROC: all z except z=0 4
 p[n] = [n]
(d) 


Solution: P(z) = 1, ROC: entire z-plane.

(e) q[n] = A[n – k], k > 0

Solution: Q(z) = Az-k, entire z-plane except
z=0.
(f) r[n] = A[n+k], k > 0
Solution: R(z) = Azk,
ROC: entire z-plane except z = .
g) x[n] = u[n], X(z) =
5
Example 2: Determine the z-transform of
x[n] = (1/2)nu[n]

Solution: 
X( z )   x[n]z
n  
n

 n  n
 1  n  1 1 
   z    z 
n 0  2  n0  2 
2
1 1 
 1  z 1   z   .......
2 2 
1

1
1  z 1
2
ROC: |1/2 z-1| < 1, or equivalently |z| > 1/2
6
Example 3: Determine the z-transform of the
signal x[n] = anu[n]
Solution:
  
n

 
 
X( z )   a n z n   az 1
n0 n0

1
 1  az  az  
1 2
 .......
1

1  az 1
ROC :| z || a |

example4: x[n]= , find z-transform.

7
 Properties of z-transform
 Linearity
If x1[n]  X1(z)
and x2[[n]  X2(z)
then
a1x1[n] + a2x2[n]  a1X1(z) + a2X2(z)

8
 Example: Determine the z-transform of
the signal x[n] = [3(2n) – 4(3n)]u[n]
Solution:
1
z[a u[n]] 
n

1  az 1
1 1
 z[3( 2)  4 3 ]  3
n
n
1
4
1  2z 1  3 z 1
Example 4: Determine the z-transform of
the signal (cosw0n)u[n]
1 1
 cos w 0nu[n]  e jw 0n  e  jw 0n
2 2
1 1 1 1
 z  cos w 0nu[n]  
jw 0 1
2 1e z 2 1  e  jw 0 z 1
1  z 1 cos w 0

1  2z 1 cos w 0  z  2 9
Time Shifting Property:
If x[n]  X(z) then x[n-k]  z-kX(z)

Proof:
since 
z[x[n  k ]]   x[n  k ]z
n  
n

then the change of variable m = n-k

produces

z[x[n  k ]]   x[m
m  
]z ( m  k )


z k
 x[m
m  
]z m
 z k
X( z )

10
 Example: Find the z-transform of a unit step
function. Use time shifting property to find z-
transform of u[n] – u[n-N].

The z-transform of u[n] can be found as

 
z[u[n ]]   u[n]z
n  
n
 z
n0
n

1
 1  z 1  z  2  ....... 
1  z 1

Now the z-transform of u[n]-u[n-N] may be

found as follows:
1 N 1
z[u[n]  u[n  N ]]   z
1  z 1 1  z 1
1  z N

1  z 1

11
Scaling in the z-domain
If x[n]  X(z)
Then anx[n]  X(a-1z)
For any constant a, real or complex.
Proof:
   a x[n]z    
 

 x[n] a z
n 1 n
z a x[n] 
n n
  X a 1 z
n   n  
Example 5: Determine the z-transform of the signal
an(cosw0n)u[n].
Solution: since
1  z 1 cos w 0
z[cos( w 0n )u[n] 
1  2z 1 cos w 0  z  2
1
1  az cos w 0
 z[a  cos w 0n  u[n]] 
n

1  2az 1 cos w 0  a 2 z  2 12
 Time reversal
If x[n]  X(z) then x[-n]  X(z-1)
Proof:
 x[m] z 
  
z[x[n]]   x[n]z
n  
n
  x[m]z
m  
m

m  
1  m
 X( z 1 )

Example 6: Determine the z-transform of

u[-n].
Solution: since z[u[n]] = 1/(1 – z-1)
Therefore,
Z[u[-n]] = 1/(1-z)

13
 Differentiation in the z - Domain
x[n]  X(z) then nx[n] = -z(dX(z)/dz)
Tutorial:Q1: Prove the differentiation property
of z – transform.
Example 7: Determine the z-transform of the
signal x[n] = nanu[n].
Solution:
1
z[a n u[n]] 
1  az 1
1
d 1 az
 z[na n u[n]]   z 
dz 1  az 1

1  az 1  2

14
 Convolution of two sequences
If x1[n]  X1(z) and x2[n]  X2(z) then
x1[n]*x2[n]  X1(z)X2(z)
Proof:
The convolution of x1[n] and x2[n] is defined as

x[n]  x1[n] * x 2 [n]   x [n]x [n  k ]
k  
1 2

The z-transform of x[n] is

 
   n
X ( z)   x[n]z n
    x1[k ]x2  n  k   z
n   n    k   

Upon interchanging the order of the summation

and applying the time shifting property, we obtain

  n 

X( z )   x1  k    x 2  n  k z   X 2  z   x1 [k ]z k  X 2  z X1  z  15
k   n    k  
Example 8: Compute the convolution
of the signals x1[n] = [1, -2, 1] and
1, 0  n  5
x 2 [n]  
0, elsewhere

Solution:
X1(z) = 1 – 2z-1 + z-2
X2(z) = 1 + z-1 + z-2 + z-3 + z-4 + z-5
Now X(z) = X1(z)X2(z) = 1 – z-1 – z-6 + z-7
Hence x[n] = [1, -1, 0, 0, 0, 0, -1, 1]
Note: You should verify this result from the
definition of the convolution sum.
16
Correlation of two sequences
If x1[n]  X1(z) and x2[n]  X2(z)
then rx1x2[k] = X1(z)X2(z-1)

The Initial Value Theorem:

If x[n] is causal then x[0]  lim
z 
X( z )

Final Value Theorem

If x[n]  X(z), then
17
Example 9: Find the final value of

2 z 1
X( z ) 
1  1.8z 1  0.8z  2

Solution:
1
   
1  z  1 X( z )  1  z  1
2 z
1  1.8z 1  0.8z  2
1 1
 1  z 1 
2 z 2 z

1  z 1  0.8z  1  0.8z 1
1 1

The final value theorem yields

2 z 1 2
y[ ]  lim   10
z 1 1  0.8z 1 0.2
18
 Inverse z-transform
In general, the inverse z-transform may be
found by using any of the following
methods:
 Power series method
 Partial fraction method

19
 Power Series Method
Example 2: Determine the z-transform of
1
X( z )  1 2
1  1.5z  0.5z
By dividing the numerator of X(z) by its
denominator, we obtain the power series
1
 1  32 z 1  74 z  2  158 z  3  16
31  4
z  ...
1  32 z 1  12 z  2

 x[n] = [1, 3/2, 7/2, 15/8, 31/16,…. ]

20
 Power Series Method
Example 2:Determine the z-transform of
1
4z
X(z )  1 2
2  2z  z
By dividing the numerator of X(z) by its
denominator, we obtain the power series

 x[n] = [2, 1.5, 0.5, 0.25, …..]

21
 Step1: Eliminate all the (–ve) powers of z.

 Step2: Make coefficient of highest power of z in

denominator equal to 1.

 Step3: Factorize X(z).

 Step4: Divide X(z) by z.

 Step5: Multiply both sides by z.

22
 Partial Fraction Method:
Example 1: Find the signal corresponding to
the z-transform
z 3
X( z ) 
2  3 z 1  z  2
Solution:X(z)  z 3

0.5

0.5
2  3z 1  z  2 z 3  1.5z 2  0.5z z  z  1 z  0.5
X( z ) 0.5 3 1 1 4
 2   2 
z z  z  1 z  0.5  z z z  1 z  0.5
1 z z
X( z )  3    ( 4)
z z 1 z  0.5
1 1
or X( z )  3  z  1
1
4
1 z 1  0.5z 1

 x[n]  3[n]  [n  1]  u[n]  4 0.5 u[n]

n
23
 Partial Fraction Method:
Example 2: Find the signal corresponding to the z-
transform
1
Y( z ) 
1  0.2z 1  0.2z 
1 1 2

Solution:
z3
Y( z ) 
 z  0.2  z  0.2  2
Y( z ) z2 0.25 0.75 0.1
   
z  z  0.2 z  0.2 2 z  0.2 z  0.2  z  0.2 2
0.25z 0.75z 0.1z
Y( z )   
z 1 z  0.2  z  0.2  2
1
1 1 0 .2 z
 0.25  0. 75  0 .1
1  0.2z 1 
1  0.2z 1 0.2 1  0.2z 1  2

 y[n]  0.25  0.2 u[n]  0.75 0.2 u[n]  0.5n 0.2 u24[n]
n n n
 The One-Sided z-Transform
It differs from 2-sided z-transform in the lower limit
of summation which is always zero whether or not
the signal is zero for n<0 (i-e causal).

The one-sided or unilateral z-transform of a signal

x[n] is defined by


X ( z)   x[n]z
n 0
n

It can be used to solve difference equation with non

zero initial conditions.

25

 Characteristics:
 
 It does not contain information about the signal
x[n] for negative values of time.
 It is unique only for causal signals because only
these signals are zero for n<0.
 For non-causal it is not unique.
 Examples:
 x1[n] = {1,2,5,7,0,1} ,

 X2[n] = {1,2,5,7,0,1}

26
  x3[n] = {0, 0,1, 2, 5, 7, 0, 1}

x4[n] = {2,4,5,7,0,1}

Observation:
x2[n] x4[n] but
Hence not unique for non-causal signals. 27
 Pulse Transfer Function
It is defined as the ratio of the z-transform of the
output to the z-transform of the input when all
initial conditions are assumed to be zero.
Mathematically,
M

1 2
b0  b1 z  b2 z  .....  bM z M b z k
k
Y ( z)
H ( z)   k 0

a0  a1 z 1  a2 z  2  .....  a N z  N N
X ( z)
a
k 0
k z k

The roots of the denominator of a pulse transfer function

are called Poles and those of the numerator are called
Zeros.
The above pulse transfer function has M zeros and N poles.
We can represent H(z) graphically by a pole-zero plot in the
Complex plane, which shows the location of poles by crosses
28
() and the location of zeros by circles (o).
 Stability:
A discrete time system is said to be stable if, and only
if, all of its poles lie inside a unit circle. If any pole(s)
lie(s) on the unit circle, the system is said to be
marginally stable.
The unit circle of z-plane is equal to L.H.P in time
domain.
Im(z) 
 Unit circle 
  Re(z)
Unstable System


Stable System
29
Marginally stable
 Example: A system is characterized by the
difference equation
y[n] – 0.1y[n-1] – 0.02 y[n-2] = 2x[n] – x[n-1].
Find the system transfer function and unit
impulse response.
Solution: Taking the z-transform of both sides of the
difference equation (ignoring the initial
conditions) we have
Y(z) – 0.1 z-1Y(z) – 0.02z-2Y(z) = 2X(z) – z-1X(z)
Y(z) [ 1 –0.1z-1 – 0.02z-2 ]= X(z) [2 – z-1]
Y( z ) 2  z 1

X( z ) 1  0.1z 1  0.02z  2
This transfer function has two poles (i.e. z = -0.1, 0.2) and
Two zeros at z =0, 0.5. This shows that the system is
BIBO stable. The Pole-zero plot is given on the next slide.
30
Pole-zero plot

Imaginary Part 1

-1
-1.5 -1 -0.5 0 0.5 1 1.5
Real Part
31
To find the unit impulse response, we
compute the inverse z-transform of H(z) by
using partial fraction expansion:
2  z 1 2z 2  z 2z 2  z
H ( z)   2 
1
1  0.1z  0.02 z 2
z  0.1z  0.02  z  0.2 z  0.1
H( z ) 2z  1 2 4
  
z  z  0.2 z  0.1 z  0.2 z  0.1
 2z 4z  1   1 
H( z )    2  1 
 4  1  0.1z 1 
z  0.2 z  0.1  1  0.2z   

and, the unit impulse response is

h[n] = -2(0.2)nu[n] + 4(-0.1)nu[n]

32
 Example: A system is characterized by the
difference equation
y[n] – 0.1y[n-1] – 0.02 y[n-2] = 2x[n] – x[n-1].
- Find the system transfer function.
- Response of system to unit step input.

- Hint X(z) =

- Y(z) = H(z) . X(z)

33
 Response of systems with non-zero initial
conditions
 We can use z-transform to solve the difference
equation that characterizes a causal, linear, time
invariant system. The following expressions are
especially useful to solve the difference
equations:
 z[y[(n-1)T] = z-1Y(z) +y[-T]
 Z[y(n-2)T] = z-2Y(z) + z-1y[-T] + y[-2T]
 Z[y(n-3)T] = z-3Y(z) + z-2y[-T] + z-1y[-2T] +
y[-3T]

34
Tutorial 5 Q1: Determine the step response of
the system
y[n]=ay[n-1] + x[n], -1 < a < 1
when the initial condition is y[-1] = 1.

35
Example: Consider the following difference
equation:
y[nT] –0.1y[(n-1)T] – 0.02y[(n-2)T] = 2x[nT] –
x[(n-1)T]
where the initial conditions are y[-T] = -10 and y[-
2T] = 20. Find the output y[nT] when x[nT] is the
unit step input.
Solution:
Computing the z-transform of the difference
equation gives
Y(z) – 0.1[z-1Y(z) + y[-T]] – 0.02[z-2Y(z) + z-1y[-T]
+ y[-2T]] = 2X(z) – z-1X(z)
Substituting the initial conditions we get
Y(z) – 0.1z-1Y(z) +1 – 0.02z-2Y(z) + 0.2z-1 –0.4 =
(2 – z-1)X(z) 36
  1 2
1  0.1z  0.02z Y(z )  2  z
1

1 z 1
1
 0.2z  0.6  1

 1
Y ( z ) 1  0.1z  0.02 z 2
 
2  z 1
1 z 1
 0 .2 z 1
 0.6

1.4  0.6z 1  0.2z 2 1.4  0.6z 1  0.2z 2

Y( z )  
1
1
1  z 1  0.1z  0.02z 2
 
1  z 1 1  0.2z 1 1  0.1z 1  
1.4z 3  0.6z 2  0.2z

 z  1 z  0.2  z  0.1
Y( z ) 1.136  0.567 0.830
  
z z  1 z  0.2 z  0.1
1 1 1
Y( z )  1.136  0 .567  0. 830
1  z 1 1  0.2z 1 1  0.1z 1

and the output signal y[nT] is

y[nT]  1.136u[nT]  0.567(0.2) u[nT]  0.830( 0.1) u[nT

37
] n n
 Pole-Zero Cancellation
 When a z-transform has a pole that is at the
same location as a zero, the pole is cancelled by
the zero and, consequently, the term containing
that pole in the inverse z-transform vanishes.
 Pole-zero cancellation may occur either in the
system function itself or in the product of the
system function with the z-transform of the
input signal.
 When the zero is located very near the pole but
not exactly at the same location, the term in the
response has a very small amplitude.
 Pole-zero cancellation may cause unstability
and should be avoided in most cases.
38
 Example: Determine the unit impulse response of
the system characterized by the difference
equation
y[n] = 2.5y[n-1] – y[n-2] + x[n] – 5x[n-1] + 6x[n-2]
Solution: Taking the z-transform of both sides of
the above difference equation and ignoring all
initial conditions, we obtain the following pulse
transfer function:
H( z ) 
1  5 z 1  6 z  2

  
1  2 z  1 1  3 z 1

1  3 z 1
 1
2.5z 1
1
1  2.5z  z 2
 1 1

1  2 z 1  2z 
1 1 1
1 2 z 1  12 z 1
and therefore

h[n]  [n]  2.5 

1 n 1
2
u[n  1]

39
 System Frequency response
The frequency response of a BIBO stable
system is defined as
H(w) = H(z)|z = ejw
The frequency response function is usually
expressed in terms of its magnitude |H(w)|
and phase (w), where
H(w) = |H(w)|ej(w)
Usually, the magnitude is plotted on a
logarithmic scale as
|H(w)|dB = 20log10|H(w)|

40
Example: Determine the frequency response
function H(w) and the magnitude |H(w)|dB for the
LTI system characterized by the difference
equation
y[n] = 1.8y[n-1] – 0.81y[n-2] + x[n] + 0.95x[n-1]
Solution: The pulse transfer function is
1  0.95z 1 1  0.95z 1
H( z )  
1
1  1.8z  0.81z 2

1  0.9z 12

Now the frequency response function may be obtained as

1  0.95e  jw
H w  
1  0.9e  jw
 2

The magnitude response is

1.9025  1.9 cos w
| H( w ) |
1.81  1.8 cos w
41
We note that |H(w)| has its maximum at w = 0,
where |H(0)| = 195. It is customary to normalize
|H(w)| by its peak value and plot
20log(|H(w)|/|H(w)|max). A plot of the normalized
magnitude of the frequency response is shown
below:
0
-10
-20
-30
-40
-50
-60
-70
-80
-90
- 0 
42
 Example: Find the frequency response for the
system
y[n] = -0.1y[n-1] + 0.2y[n-2] + x[n] + x[n-1]

Solution: The system function is

1  z 1
H( z ) 
1  0.1z 1  0.2z  2
1 1
1  z 1  z 2  z  z
Now H(z )H(z 1 )  
1  0.1z  0.2z 1  0.1z  0.2z 1.05  0.08( z  z 1 )  0.2(z 2  z  2 )
1 2 2

1 2  2 cos w
| H( w ) | | H( z )H( z ) |z e jw
2

1.45  0.16 cos w  0.8 cos 2 w

2  2 cos w
| H( w ) |
1.45  0.16 cos w  0.8 cos 2 w

43