Tanstorm

7.1 Introduction
The z-transform is discrete counterpart of Laplace transform.

The DTFT can be applied only to stable systems. But z-transform n can be

for unstable systems as well.

calculated
The solution of linear difference equation becomes easy with with ththe
z-transform. The linear difference equation is converted to algebraic eaua help o
the help of z-transform.

Purpose of learning
i) Analysis of DT signals and systems
ii) Digital filter design

ii) Digital filter/systems synthesis.

Input/outputs: For any input sequence, the z-transform is complex. It has real and

imaginary parts.
x(n)- z-transform X(z)
Real and imaginary
parts
Fig. 7.1.1
Content of z-transform study
i) Definition of z-transform.
ii) Region of convergence.
ii) Properties of z-transform.
iv) Methods to calculate inverse z-transform.
v)System analysis using z-transform.
vi) Relationship with other transforms.
vii) Analysis of LTI systems using z-transfroms.
vii) Solution of difference equations.

7.2 z-Transform
7.2.1 Definition of z-Transform July/Aug06, Marks 2
The z-transform of x(n) is denoted by X(z). It is defined as,
 7-3
e m s
z-Transforms-I

X() = x(m)z""
n=-o
. (7.2.1)
panstorn

H e r eZ - 1 5
nlex variable. x(n) and X(z) is called z-transform
pair. It is represented

pair : X{n) s X(2)

ptransform
- t r a

Types of z-Transform: Unilateral and Bilateral

defined above has both sided summation. It

Rilateral z-transform: The z-transform
alled bilateral or both sided z-transform.
sided z-transform : It is defined as,
i Unilateral or one

X() = xn)z" ..(7.2.2)

n=0

to co, i.e. sided.

Here the summation is fromn =0 one

July/Aug.-06, 07, Marks 2

712.3 Region of Convergence (ROC)
From definition, it is
is the region where z-transform converges.
Definition ROC all values
not This series is convergent
Clear that z-transform is an infinite power series.
z-transform.
of z. Hence ROC is useful in mentioning
Significance of ROC:
z-transform c a n be calculated.
of z for which
idea about values
ROC gives an
the system.
determine causality of
) ROC can be used to
of the system.
determine stability
ROC be used to
) can

Standard Signals
z-Transform of
2.4
Examples for Understanding

sequences
Determine
z-transform of following
Example 7.2.1
= {1, 2, 3, 4,
5, 0, 7
i) x() 1,2, 3,4,5,0,71

ii) x2 r)
=

(n) =
(1, 2, 3, 4, 5, 0, 71
Solution i) x
5, x|(5) 0, x1 (6)
= 7
4, X1(4) = =

2, X1(2)
=
3, X1(3) =

(0) = 1, x1(1) =

ie. Xi
TFCHNIGAL PUBLICATIONS"- An up thrust for knowledge
 Signals and Systems 7-4

By definition, X(z) = n ) 2""

n=0
1-Transfoms
X,z) =2x1()z"
1=0

Putting for x(n), = 1+2zl+3z +4z +5zd +0z-5 +7z 6

t-1

Result: i) X1(2) is as calculated above.

il) X1(z) = o for z = 0, i.e. X1 (z) is convergent for all values of z, excent

iii) Hence ROC Entire z-plane except z = 0.

i) z2(n) = {1, 2,3, 4, 5, 0,7)

i.e. x2 (0)= 4, x2(1) =

5, x2 (2) =
0, x2 (3) = 7 and

2-1) =
3, x2(-2) =
2, x2(-3) =1
Xz) =Zz(m)z"" n=-3

Putting for x2(t), = 1-z3 +2.z2 +3z+4z0 +5z-1+0z-2 +7z-3

= z3 +2z2 +3z+4 +4
Result i) Above equation gives X2(z).

ii) X2(z) = oo for z = 0 and z =

ii) Hence ROC Entire z-plane except z = 0 and o.

Example 7.2.2 z-transform of ö (n).

1 for n=0
Solution: We know that ö(r) =
0 for n*0

Xz)= xr)z""

= (n) z "
n=0

TECHNICAL PUBLICATIONS- An up thrust for knowledge

 7-5 z-Transforms-1
nasanoSYSten e m s

1
= 1z =

value for any

z. Hence ROC will be entire z-plane.
T h i s
i sf i x e d

ROC Entire z-plane (7.2.3)

orno 1,

unit step sequence, u(n).

Erample
23 z-transform of
7 . 2 . 32 - t r a

1 for n20
Jor
u(n) = {
Unit
Solution:l
on:
step
sequence,
10 for n<0

Xiz)=xtn)z""

1-o

2al" = 1+(2"l)+2}2 +(213 +(2l)+

Puting ufn),
= 12=
n=0
n=0

1. Then above equation

will be,
+A4 +A° +A* +
=
A <

Here use, 1+A

X(z) <
1-z1
1
but ROC of X(2). Here |l
1. This condition is nothing
Here2 <1 is |z|
equal to >
1 indicates
a circle of radius 1. And |z| >
1 represents
circle in z-plane. |z| Note that X(z) is
=

tepresents a shaded area.

7.2.1 shows the
ROC by
area outside of the circle. Fig. 1 is called unit circle in the
The circle of radius
1' i.e. |z| =

convergent in this area.

to real values
in the z-plane.
values are plotted
with respect
plane. The imaginary
.. (7.2.4)

1-2-1
ROC:2 >1

Imaginaryyz
This shaded area

is z> 1. It is
the circle.
out side of

Real z

This is called
unit circle

7.2.1 ROC of uln)

Fig.
for knowledge
PUBLICATIONS- An up thrust
TECHNICAL
 Signals and Systems

hand sided sequence xn) = a" u(n).

-Trensforns
Example 7.2.4 z-transform of right
Solution: By definition of z-transform, X(2) =x()z
N=-0o

a" u(r)z
11-o

= a z-" since u(n) = 1 for n = 0 to co

n=0

la2ly
n=0

=
1+(azl)+(az-l)2 +(azl)s +(az-)t+..
Here use
1+A +A2 +A3 +. A|< 1. Then above equation will be,

a2
1 T Jaz<1
X(z) =
1-a
Here laz|< 1 is equal to |z| > |a|. This is the RC.
7.2.2 shows this ROC. It Fig.
is the shaded area outside the circle of radius |a|. Note that |z|>|a| indicates the area
outside the circle.
ROC of the right hand sided sequence (i.e. causal sequence) is outside the circle.

Imaginary z

This shaded area

is lz lal. It is ROC

Real z

Circle of radius
lal

Fig. 7.2.2 ROC of a"u(n)

aur)
uo) 1
1-az1
ROC:1zl>lal
ROC 1z|>|al . . (7.2.5)

TECHNICAL PUBLICATIONS"- An up thrust for

knowledge
 S i g n a l sa n a S Y s t e m s 7- 7

z-Transforms-
1.2.5 z-transform of left handsided sequence.
Examp
-a" u(-n-1)
n) =

aa"
for for ns-1|.
Jan./Feb.-07, Jan./Feb.-08, Marks 5
Herex{n)
=

olution: 0 for n20 u-n-1) =1 for n = - 1 to - co

X(z) =

x(m)z-" =

n = -
-a"z"

n = -1, X(z) - -a
Let
l =oo

(al 2
=1
=
-{a z)+(al2)2 +(al 2)3 +(al 2) +
-(a )1+a"l z+(a2) +(al z) +(al z)* +
For the term in bracket use,

1+A+A2+AS+... =
LAA|< 1. ie.,

X(z) -(az) 1-azlaz<1

1
1 - a z - T1 a z | < 1

Here a zl< 1 is equal to |z|<|a|. This ROC is the area that lies inside the circle of
radius | al. It is shown in Fig. 7.2.3.

Imaginary z

Real z
This shaded area
inside the circle of
radius lal is ROC

Fig. 7.2.3 ROC of

-

a"u(-n -1)
ROC
f t sded sequence (i.e. noncausal sequence) is an area inside the circle.
TECHNICAL PLRIGATIQNS"- An up thrust for knowledge
 7-8
Signals and Systems z-Transforms-1
Important comment z-transforms of a" {n) and
-a"u{-7-1) are same, 1
i.e.1
1-az-1
but their ROCs are different.

-a"u(-n-1) 1-az
ROC:1z<|al .. (7.2.6)

Example 7.2.6 z-transform of both sided sequence.

n ) = a" u (n) +b" u(-n-1)

Solution: Here let x (n) = a" u(n) and x2(n) = b" u(-n-1)

(n) = X(n) +*2(r7)

X(z)= lxn)+x2m)]

n=-co n = - co

From equation (7.2.5) and equation (7.2.6) above equation becomes,

1 ROC: |z|>|a|
1-az-l 1-bz-1, |z|<|b|
X(z) = + and

i.e. al < |z| < |b|

Comment on ROC: Jal < Jzl < |b|
i) When |lal>|b|
As shown Fig. 7.2.4, there is no overlap between the shaded areas for |z|>lal ana
z|<|a|. Hence both the terms of X(z) do not converge simultaneously. Therefore RU
is not possible.

ii) When lal<|b|

For this case, as shown in
Fig. 7.2.4,
the shaded area shows the
overlap of |2|2|a
and |z|<|b]. This area is
la|<|z|<|b|. In this area both the terms of X(z) converge
simultaneously. Hence the ring shown by al<|z||<|b| is ROC of X(z).
s i g n a l sa n aS y s t e n s
7-9 z-Transforms-

al>lb Imaginary(z)

1zJa

1zl bl
Real (z)

No overlap, hence
ROC is not possible

lal<b Imaginary (z)

lal <Iz < b]-

Real (z)

z > Jal and |z] < b|

This is possible ROC|

Fig. 7.2.4 ROC of both sided sequence

7.3 Properties of the ROC

July/Aug.-06, Marks 5; Jan./Feb.-06, Marks 6; Jan/Feb.-05, Marks 4
The ROC for a finite duration sequence includes entire z-plane, except
Property 1:
z 0, and/or |z| = o.

Proof Consider the finite duration sequence x{n) = {l 21 2}

X(z) = 1-z2 +2-z +1z° +2z = z2 +2z+1+

Here X(z) = *o for z = 0 and o. This proves first property.

Property 2 ROC does not contain any poles.

TECHNICAL PUBLICATIONS- An up thrust for knowledge

 7-10
Signals and Systems Z-Transforms-1
calculated as,
The z-transform of a" u(n) is
Proof
X(z)
1-az

ROC:12|>|a|

This function has pole at z = a. Note that ROC is |z|>[al, Thi

poles do not lie in ROC. Actually X(z) = oo at poles by definians
of

pole.
in the z-plane centered about origin.
Property 3 ROC is the ring

Proof Consider a"u(n) ,ROC: |z|>|al

1-azl

or -a"u(-n1) . ROC :|zk|a

1-az"

Here observe that |z| is always a circular region (ring) centered around

origin.
Property 4 ROC of causal sequence (right hand sided sequence) is of the form
z> r.

Proof Consider right hand sided sequence a"u(n). It's ROC is |z|> lal, Thus
the ROC of right hand sided sequence is of the form of |z|>r where r
is the radius of the circle.

Property 5: ROC of left sided sequence is of the form |zkr.

Proof Consider left sided sequence -a" u(-n-1). Its ROC is |z|< |a|. Thus the
ROC of left sided sequence is inside the circle of radius 'r'.

Property 6 ROC of two sided sequence is the concentric

ring in z-plane.
Proof We know that ROC of
n) =
a"u(n) +b" u(-n-1) is lal< |zk ||
which is the concentric
ring as shown in Fig. 7.2.5.
Property 7: f
X(n) is finite causal
sequence, then its ROC is entire z-plane excey
2 =0.

Proof: Lonsider the causal sequence, x(n) = {1, 2, 3 Then its z-transsforms
will be
1+2z +3z. Clearly this sequence converges in
excepts z =0. en

TECHNICAL
PUBLICATIONS"- An up thrust for
knowledg
S i g n a l sa n d S y s t e m s
- 11
z-Transforms-
Property 8: The ROC of stable LTI system contains
unit circle in the z-plane.
This property will be
proved in section 7.8.
Proof

Property9 The ROC is a connected region.

The covergence of the

sequence exists over certain
Proof: area, rather than
discrete points. Hence ROC is a connected
region.
Review Questions

1. Define one sided and two sided z-transform.

2. What is ROC with respect to z-transform ? What are its properties?

Jan./Feb.-05, Marks 4;, Jan/Feb.-06, Marks 8; July/Aug.-11, Marks 5

7.4 Properties of z-Transform

7.4.1 Linearity
If 1") X(), ROC:a1<|z|<b
and 2() X2(2), ROC:a2<|z|<b2 then

a1 X17) +a2 I2 ) a1 X,(2) +a2 X2() 74.1)

The ROC is intersection of ROC's of X1() and X2(:)

Proof: X(z)= n)* 2la1 31() +a2 Xz(n)]z-
n=-o

24131(n) z"+a2xz(n)z-"
n=-o

Since a and a2 are constants

a1(n)z-" +a2 xz(n)z ="
1=-o

a1 X12) +a2 X2 (z)

7.4.2 Time Shifting or Translation Jan/Feb.-04, Marks 3

n) X(2), ROC: z l < r2 then

(7.4.2)
atn-k) 2* X(E) ROC:<|z|<r2
 Signals and Systems 7-12
z-Transfoms-I
Proof: Zx(n -k)J= ExMn-k):"
11-

Let n-k = m. Hence n = k+m and m = - oo to +o. i.e.,

Zx(n-k}}= atm)2 -k+m

xtm)k" =
:* -m

n=-o M=-co

= * X()

7.4.3 Scaling in z-Domain or Multiplication by Exponential

Let n) X(z), ROC n<|zkr2
then
a"x) X ROC: alr1<|z<la|r2
(7.4.3)
Proof Zja tm)) = ar)z"7 = )(a-lz)"
1=-o n-

= X(al z)

ROC: r1< 2 ie. la|ri<|z]<la|2

7.4.4 Time Reversal
Jan/Feb.-04, Marks 3; July/Aug-04, Marks 4
Let n X(z), ROC: 7<|z|kr2
then
f-1)X(:"). ROC (7.4.4)
T1
Proof: Zx(-n)} =

n=-co
{-n)z"

with n =
-m,
=xtm)z" atm)(l)-m
m=oo
=

m=-oo

X(zl)
ROC:1 zl|< r2 ie. <|zl<

TECHNICAL
PUBLICATIONS An up thrust for
knowledq
signais
e n aY s t e m s
7- 13

in z-Domain or
1-Transforms-
1 4 5
Differentiation
D i t f

Multiplication by a
Ramp
Jan. Feb.04, Marke 3
x (n) X(:), ROC:
<|z|< r2 then,

nan) -zX()
dz ROC:z|<r2 . . (7A5)

X() =

Froof 11-

dz 11-o
az
n=-o dz

xn)-(-n). z-"=l = - nxn)z"2

1-00 N-o

=
-1Enxm)]2" =-z-Z{n xn)]
1=-o

d
or Z (n)) = -zX(z), ROC Same as that of x(n)
or
dz

7.4.6 Convolution in Time Domain July/Aug.-04. Marks 4

Let x1(") X1(7), ROC: 41zkb
ROC:a2<|zkb2 then
and 2) X2(?)
(7.46)
x1()*z2(m) X(2) Xz2)
The ROC is intersection of ROCs of X1(z) and Xz(7)

Proof: Xn) *x2(n) =F x )x2 (n-)

k=-oo

-1
Zx(n)* x2(n)} =
n = - o k=-oo

inferchanging orders of summation,

k=- oo n=-oo

An up thrust for knowledge

TECHNICAL PUBLICATIONS
 7- 14
Signals and Systemss
z-Transfoms-
Since x2 (n-k) z-*X2(z),

2 z1){zkx2 ()}
k=-oo

(k)z-* X2() =X1(z)-X2(z)

k=-o

Correlation of Two Sequencees

7.4.7
X1 (7) 4X1(z), ROC:a1<|z|<b1
Let
ROC:a2< zkb2 then
and x2 (n) X2(z),

x(r)2 (n-) X1 (2)X2(z) (7.4.7)

2
n-o

Proof Correlation of two sequences is given as,

2 xn)2 (n-)
T12
=

N=-oo

x1 n)x2 l-(-n)] =
x1()**2(-)
N=-o

(n) x2(n-D= Zlz)**2(-1

n=-o

Zx}-Zx2(-)}
= X1(2) X2(z7)
7.4.13 Initial Value Theorem
Let x(n) X(z), then,
x(0) = lim X(z)
Z . (7.4.13)

Proof: z-transform of a causal sequence is given as,

X(z) =x(n)z-", since x(n) = 0 for n < 0

n=0

= {0)+(1)zl+x{2)2* +x{3)z-3+.

lim X(z) = lim 0)+ lim x(1)zl+ lim {2)z+..

Zo Z-0o

= x{0)+0+0+0+0...

(0) = lim X(z)

Zc

TECHNICAL PUBLICATIONS An up thrust for knowledge

Signals and Systems
.18
2-Transforms-
Examples for Understanding

Determine z-transform of following.

Example 7.4.1
+k)
ô (n -k) ii) r2 (n) =ô (n
i) ri(n) =

na" u{n)
iv) xa(n)
=

ii) x3() = u(-n)

July/Aug.11, Marke 6
na-uln)
v) x5(n)=
Stn-k)
Solution: i) xz(n)
=

1, ROC: entire z-plane

S (n)
property.
n-k) 2*X{:), By time delay
z* Z{6 (n)}
Z6 (n-k)}
=

ROC: entire z-plane except z =0.

= 21 = z ,

i) x2(m)
=
ötn +k)
z* Z8 (7)}
Z8(m+k)}
=

oo

ROC: entire z-plane except

z =

= z*1 = z,

i) xgtn) =
u(-n)
ROC: r1<|z|k2
xn) X(z),
By time reversal property
-n) +X(zl), ROC:<zk
T2
1 1
Here r
1-235ROC:|z]>1
=
ur)

u-n) R O C :1zk1, By time reversal propertyy

iv) xan) = n a" uln)

1
Zla" u(n))= TROC:|z|>|a|
1-az-1

And Zn xn) d differentiation in z-domain property

dz

Z m a"u(n)} = - z d 1
Here x(n) = a" u (n)
dz 1-a z

TECHNICAL PUBLICATIONS"- An up thrust for knowledge

 n a l sa n aS Y S l e m s

7-19
z-Trensforms-

1-az1-10-az) 0+a(-1)z-2
(1-azl) -Z
(1-0z
azl

(1-azl)2' ROC:1z1>|la
na"u(n) 9- az

(1-az-1)2 ROC:1z|>lal
Thus (74.14)

Example
4.2 Determine the z-transform of,
a"
a" for 0SnsN-1
n) =
x()0 otherwise
Solution:
N-1
X2) = a"z By definition of z-transform
n=0

N-1
(az-ly
n=0

N2 aN1-aN2+1
Then above equation will be,
Here usea , N2>N1.
1-a
k=N1
X(2) az")-(azl)N-1+1

1-az

1-(az-lN
1-az

has finite value.

i.e.,
is convergent if al"
ne above z-transform n=0

N-1
laz <o
n=0 terms in
number of such
are
finite
since there
zik oo, 'a' is finite,
Above sum oun will be finite if la
that |al<oo and z *0. If
abov condition laz-l< co implies
mation. The
0.
then ROC will be entire Z-plane except z =

knowledge
thrust for

TECHNICAL
PUBLICATIONS - An up
Signals and Systems 7- 20

Example 7.4.3 Determine z-transforms of

2-Transforms-
i) an) cos (2 n) u(n)
=

i) n) sin (2o n)u{(n)

=

Solution: i) {n) cos(2 n)u{n) =

e120+e-j2gn
-u (n)
2

e20+e-j2on
X() = Z u(n)
2

=z{ un}+ze-/o"u(r)
Here use a" u(n) + 1,
1 - a 2 - 1.

ROC |z|>|a| ie.,

K O C | 2 / > l a l i

1 1 1
X(2) = 1 ROC:12|>le0|
ROC |z|>|eJ0| and
: and
21_ej20z- "21-e-j90z-1
z>le/20|

Hence |e)a0|= ycos 2o +sin- 2o =

1
Here e0 = cos20 +j sin Q0.

Similarly e)0 = cos20 -j sin20. Hence le)40|

=
ycos 20 +sin Qg =1

1
X() = 21-p90z-1"1-i02-1| ROC:>1

F
1-ei0z-1 +1-ej20z-1|
2 1-e0-1)(1-e"a0z-1)|

F
2-z(e0+e"j@0) 12-z1 2 cos20
2 1-2(eio +ej20)+z2 2|1-z2
2cos20 +z

1-z cosSo, ROC:2|>1

1-2z cos 2g +z
i) x(") = sin 2 n u(n)

e1Ston-e-j20n
u (n)
2j
 7-21
a n d Systems
z-Transform
ejon -e-jS40")|
X(z)=z 2j u(n)
Z e o 7 u (n)} - Z le-ion u (n)il1

1 1
1
2j 1-e20z-l 1-e-j20z-1
ROC |z|>|e0| and |z|>|e"/uo| i.e. |z|>1
11-eja02-1-1+ej90z-11
,
ROC: 1z|>1
2j (1-ei0z-1)(1-e-jn0-1,
(ei2o -e-0 )z"l
2j 1-zl(e10 +e-j20) +z-2|
2 j2j sin2021
2j1-z1.2 cos2g +z-2
zsin 2o 2 ROC :|z|>1
1-2zcos2 +2-2

12|>1 .. (74.15)
Thus cos 1 u(n) Z
1-2z-l
c o s 2 + 2 - 2 K O C :

(n) 2
zsin220 ROC l>1 . (74.16)
sin o n u -2
1-2z cos20 +z
ample 7.4.4 Determine the z-transform of
a" a" sin(0 17)"(n).
Jan/Feb.-04. Marks 55
xm) =
cos(Q n)u(n) x(n) ii) =

Solution x(n) a" cos(2 n)u(n)

=

Let 1-z cos20

X1(n) = cos (2 0 n) u (n), hence X1(7)
1 - =21-2z-1
z cos2o+z -2*

From above equations,

x(n) = a" x1(n)

X(z) = Z{a" xi(n)}

Signals and Systems
-
7- 22
z-Transforms-
scaling in z-domain
ROC alr<|zl<|a|r2, By

cos20
Replacing z by in X1(2), = - -2
ROC: |z]> 1 |a| ie. |2|> Ja

1-2cos2 +
i) {n) = a" sin (20 n) u (n)
zsin 20
x(n) = sin ( n)u (n), hence 1-2z cos2o +z
Let

From above equations, x(n) = a" x1(n)

X() = zla" x1(n)}

in z-domain
ROC:a|r1<|zkla|r2, By scaling

sin 2 o
Replacing z by in X1(), =_

-2'
, ROC:|z|> 1|a|ie.|z|>|a|
1-2 cos+

cos2o
a"cos(Q0n)u(n) ROC z | >lal (7.4.17)

1-2cos2o+

a" sin (0 n)u(n) .

sin o
. (7.418)
ROC : 1z]>la|
1-2cos2o+
Example 7.4.5 Determine the z-transform of x(n) "for n2
=
0.
n!
Solution

By definition, X(z) = 2 x(r)z"

n=-oo

TECHNICAL
PUBLICATIONS". An uR thrust for
 z-Transfons-l
7-23
S n a r a n dS Y s t e m s

-n

n=0

n=0

then above equation will be,

= eX,
Here use

n=0

X(z) = 22-1

determine the inverse z-transform for

ditferentiation property,
Examp
7.4.6 Using
a
X(z) =
=/ >
1-22-1)2

Solution:
1
Y(2)=
Let 1-az

= a" u(n)
y(n)
-2
- Az

Y2)
dz (1-az)2

-zY()
dz (1-a2)2

is X(2) 1e.,
of above equation
Ihe right hand side

-z YE) = X(2)

above
side of
dz of lett
hand
z-transform

inverse
property,
the
rentiation

equation is ny (7) 1.e.

x(n) = ny(n)

= na"u()
Sr. Name of the Time domain z-domain Region of
No. representation convergence
property sequence

x (n) X (z)
X (z) ROC 1
x (n) ROC 2
x n) X2 (z)

a2 X22) Intersection of ROC1 and

1 Linearity aX (n) + a2 X2n) a X (z) +
ROC 2

z*X (z) ROC of X (z) except z 0 =

2 Time shifting x (n-k) if k >0 and z = if k<0

lal< l<al
3 Scaling in z-domain ax (n) x

4 Time reversal x (-n) x() 2

Same as ROC of X (2)

Differentiation in nx (n) d )
-dX
5
Z-domain
Intersection of ROC 1 and
X,z) X2 () ROCROC 2
X (n) X2 (n)
*
6 Convolution in time
domain
Intersection of ROC of
7 Correlation x 0)*x2(-) x, a)-*2 (2) X,2) and X2 (z)
www.e

and
xn) X2 n)
1 2 M<z|</2y Ml
8 Multiplication

X2 dv
Same as ROC of X (z2)
9 Conjugation xn) xz)
Includes ROC of X z)
10 Real part of x (n) Re x (n e)+x e
Includes ROC of
X (2)
11 Imaginary part of Im x (n) 27)-x*e]
x (n)
x (0)= im X (2)
12 Initial value theorem Causal x (n) Z o

13 nal value theorem Causal x(n) x(oolim (1-21)x)

Z1

Parseval's relation
14
n=-
) 0)-277fx,0x|a C

Table 7.4.1 Properties of z-transform

 z-Transforrms-1
Signals and Systems 7- 25

(7) ROC
Time domain sequence x Z-transform X (z)
SrNo.
Sr.No.

(n) Complete z-plane

1
u (n) z>1
2
1-z-1

a u (n) 1 1z>ja
3
1-az

- a"u(-n- 1) 1 z<la
4
1-az-1

na" u (n) z>lal

5
(1-az-1)
- na" u (-n- 1) az-1 |z< a
(1-az)
cos (9 n) u (n) 1-z cos 2o z>1
1-2z- cos 2g +zz
8 sin (20 n) u (n) z sin 20 z>1
1-2z- cos 2o + zz

9 a cos (20 n) u (n)

1 z >a
cos S20

1- cos 20 +

10 a sin (20 n) u (n) |z >lal

sin S2o
2
1-cos2o
for 0 sn sN 1 1 - a zN >0
11 a -

otherwise 1-az

12 Ak Ak |z >lPk l= k
2AI cos(B, n +
ax) u (n)

A Aklel "k and

1- Pk 2 1-
Pk z-1
P e Bk
Table 7.4.2 Standard z-transfer pairs
 7-26
signals and Systems
z-Transforms-
Examples with Solutions

Example 7.4.7 Determine z-transform of following sequences.

i)(0.2)"{u(n)-u(n-4) Jan. /Feb.-04, Marks

i a,0<|a |«1 July/Aug.-04, July/Aug.-11, Marks q

ii) Ar" cos(n+0) u("), 0<r<l July/Aug.-04, Marks 4
iv) a,0<la l<1 Jan./Feb.-05, Marks 5
/7t 3
v) u(n +3) Jan/Feb.-05, Marks 5; July/Aug.-10, Marks 6
i) 2 ) m 8 ( - 1 Jan/Feb.-05, Marks 5, July/Aug.-10, Marks 6
vit) 2"u(n) +3" u(-n-1) July/Aug.-05, Marks 3
viile 10 sin () July/Aug.05, Marks 3
ix) n uln) July/Aug.-06, Marks 3
x) cos n0 u(n) July/Aug.-06, Marks 3
Solution: i) x(r) =(0.2)" {u(n)-u(n-4)}

X() = (m) z " = (0.2)" u (r) -u(n-4)} z"

n=-o°

Here un)-u(n-4) = 1 for 0n 3

0 for n 2 4 and n -1

..
X(2) =

2 (02)" :""
n=0
=
2 (02")"
n=0

Here use 2

n=0
a"1-aN+1
1-a
Or aN+a-1 then

X(z) 1-(0.21)
1-0.2 z

TECHNICAL
PUBLICATIONS An up thrust for
knowledge
 a n a l sa n d S Y s t e m s

7-27

above equation is
nvergent if
z-Transforms-
The

1=0
0.2 z-1< o.
This summation will be finite
This condition
n states
that ROC is
ala2: entire
z-plane except =0.z

=
a,,0<a<11
il) x(n)

Here
x (1) =
a "u(n) +a u(-n-1)
=a"u(n)+ (-n-1)

1-a z-1 ROCz|>a

and (-n-1)- - ROC:z

X() ROC : a<|z<

1-d21-4z1
1
li) x(n) =
Ar" cos(2pn +o) u(n), 0 < r<

X(2) =2 x(n) z-"

Ar" cos (2g +0) z7

n = 0

j0 +e j20n-jo

A ne120n+
=

2
n =0

n=0

(reo 2"+e («)":

n=0
n=0

-1

2 1- 1-res20z

for kaowledge
Signals and Systems 7-28 z-Transfoms-
These two conditio
reo-1|<1 or re"0z"|1. condition lead
his equation converges for
be further
simplified to,
equation can
to ROC of |z|>r. The above
cos-rcos(20 -0)z ROC : |Z |>r
X(?) A 1-2rcos2z+r2z-2

iv) x(n) = a,0<|a|<1

()+a"u(-n-1)

Cu(n) +a"u(-n-1) =
1 z a n d |z|<a
xe)1 1 1 - a

-a ROC:<|zka
1+a1+2

v)x(n) =r un +3)

u(n +3)

n=-o

u(n+3) z

u(n +3) 7

Here u
(n+3) =1 for -3 ng Hence above
co,
equation becomes,
X()

Let
n= -3
n=

TECHNICAL PUBLICATIONS -

An up thrust for
knowledge
Signals ana Systems

7-29
z-Transforms-
+1

(32)
1 3z-1

We know that, n x(n) -zX(2)

dz 32-1

X()

(32-1):3(4:3)-(3:)*3
(32-1)2
=
.(32-1)-12z° -94 12 (3-1) z*-9:3
(32-1) (32-1)
32(4-92)
(32-1)

x(n) =

Z
o | z / j s(e-n
X(z)
=

Herez M
knowledge

thrust for
PLURLICATIONS- An up
...
Signals and Systems 7-30
z-Transforms-1
z{8 (n)} = 1

z- 1-
X(2) = 1+-

ROC:

(
ROC:zP
(
vii) x(n) = 2"u(n) + 3" u(-n -1)

X() =z{2"u(r)}+z {3"-n-1)}

1 1
z2, zk3
1-2z 1-3z-1
z ROC: 2<|z|<3
1-52-14+62-2

viil) x(n) =e 10 sin u(n)

-1
sin20
a"sin(9 n) u(") -2
1- cos2 t
Here A e10

Hence a"
a = 0.9

And 27tT
 7-31

z-Transforms-
X(2) = 09 Sin

1-2 9co co09 z09

0.9

0.636z-1
1-1.272z+081z 2 z09

un)
n)
=
n
n(m) -X()

dz

1
1-2
Here

1
dz 1-z

( 1 - )2
d
-z
dz dz (2-1)
-z2+1
a-1)*]
- z

(z-1)4
z(z-1)(2+1)
i.e.
(-1)
(z-1)
z(z+1)
(z-1)
 Signals and Systemss 7-32

x)x(n)= cos ne u(n)

z-Transfoms-
1-z cos2 0
cos ( ) u(r) 1-2z cos2g +z*2

Here Qo
u(n)
1-2 cose
cOS
n 1-2z cos0 +z-2
Example 7.4.8 Determine the z-transforms of following sequences using z-trane.
form
properties.
ii) x(n) = n a" u(-n)
i) x(n) u(-n)

ji xfn) (2)"u(-n-1)
= ) An) =
{n+1) u(n)

Solution i) x(n) = u(-n)

1
u(m) ROC:1
1-z

By time reversal property, x(-r) x(2")

ROC: >1or |k
(i) x(n) = n a"u(-n)

Z
a "uln)2- or ROC:
1-az-1 Z-C

au(n) . ROC:

By time reversal property, «(-r) > x(z")

z
a "u(-n) - ROC: orll<|a

1
1-z

Now by differentiation property

na"u(-n) 4|1
TECHNICAL PUBLIGATIQNS"- An tun thrust far knqudadae
 7- 33 z-Transforms-

s i n a l sa n aS y s t e
ems

(1-a)

ROC:1z<a

(1-
m)x n ) - 2 ) " u ( n -

=
(2)u(=n-1) u(-n-1)
1()
Let becomes,
1 Z-transform of above equation

Gince
-a"u(-n-1)

X1(z) =

z
X1(n)= nx(»)
Now

X1()
= -z
)
d X(z)
-Z
dz

or
dX(2) 1
dz
becomes,
above equation
d
We know that,log, x=<. Hence
ROC:1
X() = log +2

iv) x(n)= n{n +1) u{n)

(n) = (n2 +n) u(n)
nu(n)+n u(n)
n " A n

up
for knowledge
thrust
 Signals and Systems
7-34
z-Transforms-
z-1 i.e
z
n u(n) 4 .
1-1) (z-1)

u)- a (1-")
d Z
dz (z-1)

+ )

(z-1)3

z(z+1)
(2-1)3

X() =
Z(+1), 2_
(-1) (-1)?
222
ROC: z|>1
(z-1)3
Example 7.4.9 Find z-transform of the folowing sequences. Also draw ROC in ead

i) x(n) = 4ei(/9"u{r)
) xpn) =
2(0.5) u(n)+(2)"u(-n-1)
ii) xsn) =
(0.8) cos
Solution: (i) x,(n) =4e(4 n)°

By definition of z-transform,

X1(2) =2qn)z"

= 4e(/4) u(n)z "

n=-co

= 4E(ein41)"
n=0

1
=4
1-e17/4z-1 e /4«1
TECHNICAL PUBLICATIONS - An up thrust for knowledge
 S y s t e m s
7-35
z-Transforms-
s y n a i s
a n 2 0

4
1-en/4z-1'
ROC: >|ein/4|

i.e. |z> 1

a x40)=(0.5)
un) +(2)*u(-n-1)
equation can be written as,
The given
x,f) = 20.5)" (0.5) u(n) +(2)"u(-n-1)

= 4(0.5)"u(n) +(2)"u(-n-1)

T.ing Z-transform of above equation with the help of standard relations,

1 1
Xz)=4- ROC z|> 0.5 and |z|<2
1-0.5z 1-2z-1

4 1
ROC 0.5 <|z|< 2
1-0.5zl 1-2z-1

ixn)=(0.8)"cosn u(n)
We know that,
-1
coso
cos() u(n) ROC:2 >|a|
1- cos@ +
Here 0.8 and
a =
@o

X(z) = 1-05cos
0.8
ROC:>0.8
-2'

1- 0.8 cos
Example 74.10 Find the initial and final values of the signal.
X(z)= 7-3)z

Solution z-1)(z-0.4)
Initial value, lim
x(0) = z co X(z)
 Signals and Systems
7- 36
z-Transforms-
lim (z-3)z
(z-1)(z-04)

lim 1-

= 1

lim
Final value, -) 1(1-2")x()
= )_(2-3)z
(z-1)(z-04)
lim
1 =-
Example 7.4.11 Determine z-transform of the following functions.

i)zn)=|
i)xn) = )

i ) xtn) (06)"u(n)»(09)"'u(r)
=

Solution: i) x(n) =|n+2)

Here the z-tarnsformn o f | n) will be,

|u(r)- ROC: | >

1-
erty,
Here u(n +2) is time advanced by two samples. Hence using time shifting ptop

2 and z
X(z) =2 =
z2
ROC:12> *
 a l sa n dS y s t e m s
7-37
z-Transforms-1

un)
a2-1
na"u(n) 2
W ek n o w t h a t ,

(1-az)

ROC:z>

i l )x i n ) ( 0 . 6 ) " u ( n ) - ( 0 9 ) " u ( n )

S
1 ROC z | > 0.6
Here
Here
(0.6)u(n)
-0
1 ROC z | > 0.9
nd (09) (") 1-0.92-1

By convolution theorem,
1 1
X(z) =
1-0.6z 1-0.9zi

1 ROC:z|>0.9
(1-0.621)(1-0921)

Find the z-transform and ROC of the following sequence.

Example 7.4.12
tt=58(n+1)+5u(r)+4"u(-n-1)
can also be written as,
olution: The given equation

xn) =(n+1)+5(2)"1u(r)+4u(-n-1)
Using the z-transform properties
z-plane expect z
= .

ROC Entire
1

ROC:2|>2
1
5(2y u(n) 5_2
ROC: <4
1
4"(-n-1) -421 1-42-1

t h u s t for knowledge
 2-1 ransiorms-

Thus,
X(z) 1
=

-2-1 1-42
1-2z-1 1-4z-1
ROC:2<|z|<4
Example 7.4.13 Determine z-transform, ROC, pole-zero locations of the following functiom.
Da get pole-zero plot. i) 0.2"
cos
( n) uén)forS2 =
2 {u(n)-u(n-4))
Solution:
Jan./Feb-04, Marks 10
i) a" cos(9gn) «(n)
Refer to example 7.4.4 (i).
Im(z) With 20 21,

cos 27
X(2) =
2
1-2
cos 2nt+
Re(z) 1-az-l
1-2az- 4+a22-2
Fig. 7.4.1 Pole-zero plot
Z-a
Pole z =a
Zero z =0
i) .2 {u(n) -u(n -4) }
Refer to example 7.4.7
(i).
Example 7.4.14 Find "e z-transform of the
following including ROC.
( x(n) =E -6 4--1)
(i) x(n) a"=

0<a k1
(ii) x(n) =A r" cos(9on +0)u(n) 0<r<
1.
Solution:
July/Aug. 04,Marks 12
i) xm) =

-| u(1-1)
-a"u(-n-1) -
1-az-1 ROC:12< lal
TECHNICAL PUBLICATIONS"-
An up thrust
for knowledge
 z-Transforms-1

Signals and Systems 7 39

X(z) =
1-
ROC:z< 1

i) xn)
= a" 0<a<1
Refer example
7.4.7 (i).

1.
iil x(n) =
Ar" cos(20n +)u(n), 0 <r<

7.4.7 (iii).
Refer example
State and prove differentiation property of z-transform.
Determine the signal
Example 7.4.15
x(n), given
5 1
X(2)
Jan /Feb.-05, Marks 8
Solution
5z3
X(z)

4 21 3
X(z) 5z 5 + 2
2 zt 2

1 1
X() =
5
2
1-

1 1 2
+3

51- 1 1-
x(n) =
u(n)
 Signals and Systems 7-40
z-Transfoms-1
Example 7.4.16 The signal xtt)=eat is sampled every T sec. beginning at t= 0. Find th
he
z-transform of the sampled signal. Give two diferent alues of (a, T) such that
t
Z-transforms of the sampled signals are the same. July/Aug-05, Marks8
Solution:
X(t) = e-at

x(nT) = e-anT

11 -co n=0 n=0

1
1-e-aT z-1 '
ROC:
"|<1 |z|>le
or

In the above equation X(z) will be same for all value of a and T for which the

product aT is constant. For example a =3, T = a =2, T= etc.

Example 7.4.17 Find the z-transform of the signal,

xin) =
n sin -n) Jan./Feb.-08, Dec.-12, Marks 6
Solution :

x(n) = n sin| -)
Let, 1)sin7 ut-n) =« - -u-r)= 2j u-)- 2 u-1)
2j

Since u(-n) = 1 for ns0, above equation can be written as,

n = - oo

=
1 1
n=0 n=0
41-2z 1-e7zl

TECHNICAL PUBLICATIONS- An up thrust for knowledge

sonals and Systems
7-41 z-Transforms-l

for <1 and 1

Z
1+22
ROC:|2|>|e7and|z|>|e'2|

i.e., |z|>1 since e

-
Now let us u s e n x(n) -z

Here nsin ut-n) -

«
z(1-z2) ROC 2>1
n sin |u(-n) (1+z2)2

Example 7.4.18 Determine z-transform and ROC of the following:

Jan.Feb.-08, Marks 10
i) x(n) =|

Solution: i)
t-in"
sino ROC z|>|a|
Z
We know that, a" sin (2 o")u(n)
.

1-2cos2o+
becomes,
above equation
With a = and 2 =
4
(3z)sin 4 1
sinnun) . ROC:|>
-2(3z)cos+(3)

0.235z-
1-0.471zl+0.11z2
ROC:z>
and indicate the region of convergence
Example 7.4.19 Find the z-transform of the following
) x]=n(n-1) a" ulbo)
) a]=2" sin 2, (1-2) u(n
-

2);
July/Aug.-11, Marks 12, July-13, Marks
10
Signals and Systems 7 42
z-Transforms-
Solution: i) x (n)= 2" sin 2 (n-2) u{n -2)

= 207-2+2) sin [2a(n-2)]u (n -2) By rearranging

27-2.22 sin [Q (n-2)1Ju(n-2)

= 4:2 (-2) sin [Q0(n-2)]u(n-2)
-1
sin20
We know that, a" sin (20 n)u(n) s -
1-2 cos
By time shifting property,

sinsin20
an-2) sin [2o (n-2)]lu (n-2)

1-s)
X(2) =
4.2sin20
1- cos

ii) x(n) =n(n-1) a" u(n)

x(n) = n a" u(n) -na" u(n)

1
a" u (n) ,
1-az-

or a" u(r)
na"
u(n) - By differentiation in z-domain
az
i.e. na" u (n) - (74.19)

(z-a)
Now n[na" u (n)] -2 d az
By differentiation in z-domain

or na" u (n) -
az(z +a) ... (7.4.20)

(z-a)3
TECHNICAL PUBLICATIONS- An up thrust for
knowledge
 7- 43 z-Transforms-
Signals and Systemns

(7.4.19) and (7.4.20) we can write the z-transform aas,

Erom equations
a z(z+a) az
X(z) =

(z-a)3 (z-a)2

2az 2a2z-2 2(az)2

(z-a) (1-az-l)3 (1-az-l)3

Review Questions

1. State and explain four properties of z-transform.

Prove following properties of
z-transform
2.

i) Linearity
ii) Time shifting property
9, July/Aug.-04, Marks 5, Jan./Feb-04, Marks 3
July/Aug.-11, Marks
Jan/Feb.-06, Marks 6
ii) Convolution in time domain property.
July/Aug.-10, Marks 4
in z-domain
io) Differentiation
July/Aug.-10, Marks 4, Dec.-12, Marks 3
v) Time advance.
Dec.-12, Marks 3
oi) Complex conjugation.
Jan/Feb.-04, Marks 6
3. State and prove
initial value theorem of z-transform.

7 . 5 Inverse z-Transform

The inverse z-transform

can be obtained by,
i) Power series expansion
i) Partial fraction expansion
ii) Contour integration.

using Power Series Expansion

Inverse z-Transform
5.1
is given as,
z-transform of the sequence x(n)
By definition

+x(1)zl +x(2) z2
+...

+0-2) z +x(-1) z +x(0)

x(n) can be obtained as,
rrom above expansion of z-transform, the sequence
x(2),.
x(n) =
..,x(-2), x(-1), x(0), x(1), division method.
can be obtained directly or by long
he power series expansion

PUBLICATIONS- An up thrust for knowledge

TECHNICAL
Signals and Systems 7-44
z-Transfoms-
Example for Understanding

Example 7.5.1 Determine inverse z-transform of the following:

i )X(z) = ROC:|l>lal
1-azl

i) ROC:1zkla|
X(z) 1 - - 1
Solution
1
i) X(z) = , ROC: z|>la|
1-az-1

1+azl+a2z2 +a3 z3-Negativepower of'z'

1-az) 1
1-az
azl

az -a2z-2
+

a2 z2

a2z7 -a3z3

a32-3
a3-a4z4

az4

Thus we have, X(z) = =1+azl +a2z-2 +a3 z-3.

Taking inverse z-transform, ax(n) =

{1, a, a2, a3,)
= a" u(n)

1
X(z) = , ROC:|zk|a|
ii) 1-azl

of
Equation rearranged to get positive powers Z
- az+1

TECHNICAL PUBLICATIONS An up thrust for knowledge

 7-45 z-Transforrns-
S i g n a l sa n d S y s t e m s

-a-lz -a i zi - a z ' -az4 -Positive powers Z

-azl+1)1
1-az

az
a z -a-2 z2

a2z2

a22 a-3 z3

a3z3

a-3z3-a z4

a4z4

-a-4 z4.
1
- a z - a z 2 - a 3 z3
Thus we have, X2)
X(z) =

1-az-1
=

a-4 z4 -aS z3 -a2 z2 -a-l

Rearranging above equation,

Taking inverse z-transform, n) =

t. -ad,-aS,-a"2,-al

-a u(-n-1)
such that
|z| JaL then expand X()
When the ROC is
>
omments on Results i)
powers of 'z' are
negative.
such that powers of 'z
are positive.
when ROC is |z| < lal, then expand X(z)
CVaneion
 1
Example 7.5.2 Determine inverse z-transform of X(2) =
.
1-1.52+0.5z-2
For i) ROC: z| > 1, ) ROC: |z|< 0.5 and ii) ROC: 0.5 < |z|< 1
Solution: Step 1: First convert X(z) to positive powers of z. i.e.,

TECHNICAL PUBLICATIONS An up thrust for knowledge

 ynais
a n aS Y s t e m

7-47
z-Transforms-
X(z) z 2-1.5z +0.5
Z
X()
2-1.5z+0.5 (z-1)(z-0.5)
x= A2
-0.5
SIep2
(7.5.6)
Z
A 1
)-1)z-0.5) 1-0.5
=

0.5
A2 =
(z-0.5).
gnd (z-1)(z-0.5). 0.5-1--1
0.5-1

1
will be,
Equation(7.5.3) Z z-1 0.5
2z Z
X(z) =
Step3: z-1 z-0.5

2 1
.(7.5.7)
1-z 1-0.5z-1

Step 4:i) x(n) for ROC of |z| > 1 |

Here the poles are at z = 1 and z = 0.5 from equation (7.5.6).

Now RCOC of |z|> 1 indicates that sequence corresponding to the term 2
1-2-1
n

equation (7.5.7) must be causal.

Observe that |z|>1 includes
'ig. 7.5.l shows the ROCs of |z|>1 and 2 |> 0.5.
2> 0.5. 1 must
in equation (7.5.7)
Hence the sequence corresponding to the term
1-0.5z-

also be causal.
Imaginary (z) This area is |zl>1
It also includes

z0.5

Circle of
z = 0.5

Circle of
z = 1 Real (z)

In this a r e a

> 0.5
z> 1 and |z|
both are true

and |z|> 0.5

of jz|>1
7.5.1 ROCs
Fig. thrict
for bnnwlorin
P U B L I C A T I O N S - An 1D

TFCHNICAL
Sigrals and Systems 7 48
z-Transforms-1
Iherefore from equation (7.5.7) inverse z-transform becomes,

n) = 2(1)" u(n) -1.(0.5)" u (7) = [2-(0.5)"]u(n)

ii) xn) for ROC: |z|< 0.5

i n equation (7.5.) will be noncausal.
The sequence corresponding ,1-0.5z

Fig. 7.5.2 shows the ROCs of |z|< 0.5 and |z|<1. Observe that 0.5 incl
|z|<0.5 includes
z<1.
Imaginary (z)|
Circle of This area is |z|<0.5
z = 0.5 It also includes
zl<1
circle of
z= 1
Real (z)-

In this area
Iz<0.5 8& |z| < 1
both are true

7.5.2 ROCs of |z|< 0.5 and |z]< 1

Fig.

Hence sequence corresponding to i n equation (7.5.7) will be noncausal.

Therefore from equation (7.5.7), inverse z-transform becomes,

xn) = 2[-1" u(-n-1)I-1-0.5" u(-n-1)1

= -2+0.5" ]u(-n-1)

i) xn) for ROC 0.5 <|z|< 1

This ROC can be written as |z|> 0.5 and |z|< 1.
2 noncausal snt
T h e sequence corresponding to in equation (7.5.7) will be
1-z-1

ROC is zl< 1.
nce
The sequence corresponding to 1 causal sin
in equation (7.5.7) will be
1-0.5z1
ROC is |z|> 0.5.
Taking inverse z-transform of
equation 7.5.7,
x(n) =
21-1" u(-n-1)]-(0.5)" u(1) -2u(-n-1) -(0.5)" u (1)

TECHNICAL PUBLICATIONS - An up thrust for knowledge

q 2 8 / sa n aS Y s t e m s
7- 49
z-Transforms-I

Example 5 . 3
Determineinverse z-transform of

1 , ROC: z]> 1
Xz) =
( 1 + 2 ) ( 1 - z - 1 ) 2

CHan 1
1:: (Converting X(Z) to positive powers of z,
Step
olution:

X(z) =

(2 +1)(z -1)2

X() (z+1)(z-1)2
H e r e there is multiple pole at z = 1. Therefore the partial fraction expansion

will be,
A3 ... (7.5.8)
X) *(-1) (z-1)
X()
A =
(z +1).
-1 (z-1)2-1
z
A3 = (z-1)2 X(
z2-1 z+2-1
d X()]| By equation (7.5.3)
A2 2=1
2+1)2z-z4

(z+1)2 z=1

utting values in equation (7.5.8),

X(E) 1/4 3/4

z-1 (2
1/2
2

Step 3 1/4 3/42 1/2z

X()=+Z-1 (2-1)
X(2) = 3/4( 11/2
1/4+ - 1 -
1+21-2-1 (1-2)
Step 4 ROC is Let us following
relations

|z 1. use

un) - 1
1-p, , ROC: 1z|>|Pk I
1-Pk
for knowledge
T E C H N I c A L
PUBLICATIONS - An up thrust
Signals and Systems 7- 50
z-Transforms-
Hence inverse z-transform of first two terms of X(z) will be,

IZT1/4 -1)" u (1) and IZT. 3/4

1+z
For 3 term of X(z) let us use,

np K u(n) . P2 ROC: z|>|Pk|

(1-P

IZTY2
i.e.
(1-2
Putting all the sequences together,

Example 7.5.4 Determine the sequence x(n) from follorwing function

1+z-1
X(z) =
1-z+0.5z2
Assume that x(n) is causal.
Solution : Step 1: Converting X(z) to its positive powers,

X(z)
2(1+ z(z +1)
Z+0.5 z-z+0.5
X(2) +1 +1
Z
z-Z +0.5

Step 2 X() A2
Z

1 X(2)| Z +1
Where

and z+1

TECHNICAL PUBLICATIONS"- An up thrust for knowledge

 a / sa n aS Y s l e m s Bf1STGrtns-

P o i n t :
Here
notet
A: This is aluways true. Hence value
of only A, is
Key determine A,.
to
sufficient

1
X()
-i

X()=
Step3
inverse z-transform of above
causal
sequences are assumed. Taking
Here
:
Step
equation,

u(n)
x()
Convert all the term to their polar form. It can
:
above equations
Simplification of calculator. i.e.,
of P-R function on
converted with the help u(n)
71.56°) (0.7072- +5°)
be
- 71.56°) (0.707 245°)" u(n) + (1.582
xn) =(1.58 e/45 )"u(n)
e)45" u (n) +1.58el7.56 (0.707
1.58e171.56
= (0.707
(1)
(0.707)" el451 +1.58e]71.56 (0.707)" e/4 u
[1.58 e]71.56°
=

(0.707)"Iei(4571-71.56) +e i(45n -71.56) lu(n)

=
1.58
euler's identity
-71.56) u (n) By
1.58 (0.707)" 2 cos (45
n

3.16(0.707)" cos(45n-71.56°)u (11)

1s,
whose z-transform
Example 7.5.5 Determine the sequence

X(z) = 1+2z+z2 ROC: 1z|>1

1--11,-2
traction evpansion.
for partial
Solution : To d
arrag e X(z) in proper
form
suitable
d e n o m i n a t o r
n A(z) are
same

and
The hig neg
C8aive powers
of
numerator

in partial
fractuns.

Hence such C
equ
quation cannot be expanded
*
Let us rearrange X(z) as foilows :
+2z-+1
X(z) =
1+1
 cdlla SyStems
lastorms-i

Now perform the division,

2

-2-321+2
52-1
X(z) = 2+
521-1 -1+5z
.
2+
1-

-1 +5z
= 2+X1 (z),
where X1(z) =

3 ..(7.5.9)

Now apply step 1 to step 3 for X,(z)

-1+5z-1
Step 1: X(z) =

1- 2
z(-z +5)
By converting to positive powers of z

X1(2) z+5
7.
7

-2 +5

(z-1
Step 2: X,() =
+ A2

(z-1).X Z(z)|
-z +5
where
A =

1
=
8
z= 1
221
and
A z-1
-2 +5|
9

TECHNICAL PUBLICATIONS An up thrust toc kran lad

 7- 53
a/sa n d S y s t e m s Transormo
S o n

X()
-1

9
8
X()= 1
Step3 1- 1

for A(z) in equation (7.5.9),

CAan
Step
4: Putting
8 9
X(:) = 2+-
1- - -1
1

8
Hence sequence corresponding to will be Causal
.Here ROC is |z|>1. 1-2

9
.The causal sequence corresponding
to will be 9 u(1). It has ROC o
1-

1 also. Hence x(n) will be.

Note that given ROC of |z|>1 includes |E]>

x(n) = 26 (/n) + 8(1)" u (n) -9

2
u (1)

20(n) +8-9 (11)

Eampe 7 Deternmine inverse z-iransfonm of the following

i) X(2) = sinz
JanFeb 04, Marke
11N(E (2-4 3)
(-3)(z-2}(7=1
ROC of a) 2<|z|< 3

b)z|>3
c)|z<1
ii) X(-)=7
1-1:" use convolution property. JulyAugs 06 Marke 5

TECHNICAL PUBLICATIONS An up thrust for knowledge

 Signals and Systems 7- 54
z-Transforms-
Solution : i) X(z) sin z
()'221*1
sin (2) =
i
1=0 (2 +1)!

Z- 3 ! 5!7 .

z° +x(-5) z° +x(-4) z"

X(z) =
..x(-7) z +x(-6)
+x(-3) z° +x(-2) z* +(-1) Z+ x(0)+x(1) z"+ x(2) z*2
On comparing above equation with sin(z) we get,L

) -
050,0, 1,o

s
ii) X(:-3)(z-2)}-1) (2-4z+5s)

X) 2-42+5)
(z-3)(z-2(z-1)

1
-32
1 1 1
x() 1-3z1-2z-1 1-z-l
a) To obtain x(n) for 2<|z|<3
X(z) has poles at =1,
2 and 3. Hence sequences due to
z will be
poles z = 1 and
causal. Sequence corresponding to pole z 3 will be noncausal. =

x(n) =
-3"u(-n-1)-2"u(n) +u(n)
b) To obtain x(n) for |z|>3
Since all the poles are inside the ROC, all the sequence will be causal.

n) =
3"u(n) 2" u(") +u(n)
-

- 3-2+1]u(r)

TECHNIOA DIIDI inTiC

 55

To
obtain x(n) for z1. z-Transforms-I
the poles are outside of
ROC, the
7) sequence will be noncausal.
3" u(-n i.e.,
=
-1)--2"u(-n-1)+u(-n-1)
-3+2" +1 u(-n-1)
ii) X(:)

X) =
X() X2(2)

X)

X2(2)= = z(")=- )

By convolution property,

():(7) X,(2) X2()

(n)=A1(n) x2()
(n) k ) 2(-k)

un-k)

uk) u(n-k)

0sksn. Hence,
tlere (k) u(n-k)=1 for

PUBLICATIONS
An up thrust for knowedge
CHNICAL.
IE
Signals and Systems 7- 56
z-Transforms-
Example 7.5.7 Find the inverse z-transform of the following

i) X(z) =log 1 - a z - 1 >|a|

1
i) X(z) =log :kla|
1-az
ii) X(z) = ez

Solution: i) 1
X(z) =log (1 -az-1)"

-log(1-az")
log(1-p) is expanded with the help of power series. It is given as,

log(l-P) IPl1
n=1

X(z) = foraz<1 or z|>|a|

n
1=1

n=1
n
-11
2r(n) :
n=1

Here x(n) = J for n21

n

Otherwise

a"
.e x(n) =
n
u (n-1)

1
ii) X(z) =log z<a
1-az
-log(1-al:)
a 1 or |=I<la|
n=1

n
n=1
g n s i sB n aS y s t é l

will be
mthen above equation

()
L en
t =

Z-
-m m
X() m=-1

-177
2x(m) z
m=-1

for m=-1 to- co

a(m) otherwise
Here

x(n)= "u(-n-1)
1

ii) X(z)-e

log X(2)
sides with respect to z,
Differentiating both

1dX(:)
E) d
dX() xE)
dz
Multiplying both sides by -z, we get,

-NE) = X )

aking inverse z-transtorm,

1
a(n) =a (n1)
OT
() = x(n-1)

initial value is
given as,

lim lim
a(0)= X(E)
=1
1
r(0) =
1 i.e r(0) =

An up thrust for knowledge

TECHNICAL PUBLICATIONS
 Signals and Systems 7- 58 z-Transforms-
and for n =1,
()=()=1 =
n=2,

n = 3,

n = 4,

Thus, x(n) u(n)

X() using partia fraction expansion
Find the inverse z-transform of
Example 7.5.8
approach:
+1 July/Aug.-11, Mariks 6
X(z) -ROC:|Z|>1
3z2-4z+1
Z+1
Solution. : X (2)
z z (3z-4z+1)
z+1 (7.5.10)

.(7.5.11)
+ 1 ,-1
2
X() = 1--
1-z

O
=andz =1. The ROC
From equation (7.5.10) observe that the poles lie at z=0, z

Hence time domain sequence will be causal. Taking inve

z>1 is exterior to poles.
z-transform of equation (7.5.11),

x(n) 8()-2ut)+u(m
Example 7.5.9 Find the inverse z transform of the following :
z+2
1/812>1/2 i) X()=-a2.
i) X2-3/4z+1/8
z-a
20
culy/Aug 11, Marks 8, July-13, Marks
TECHNICAL PUBLICATIONS-An up thrust for knowledge
Signalsa n d
Systems 7- 59 1-Transforms-I

i)Thegiven X(z) can be expressed as

z(z* +z)
X() 1

z+z . (7.5.12)
X) 1
of

Hence let us perform an actual division as

are is not a proper function.

Here Z

follows
3
Z+4

3+z

3
1632

3
16
written as,
can be
Ihus equation (7.5.12) 3
23 3
3
+
16
X() *3 16 32
3-*s Z

57/2 17/16
Zt 1
4
17/16
5/2
X(2) =1 -
Signals and Systems 7- 60
z-Transforms-
Taking inverse z-transform,

xo) =
ö(n +2)+ ön+1)+ u-16l u(n)

1-az
i) X(z) =

1
az-1 2-11
a
z-l
a-z 1-az-l 1-az 1-az-

Taking inverse z-transform of above equation,

X() = a"-l u(n-1)-a" u(n)

a
Example 7.5.11 Using long division, determine the inverse z-transform of
X(E) = - 1

1-
when the region of convergence is |z |>1.
July/Aug.-06, Marks 4
 Signals and Systems
7-62 z-Transforms-1
Solution

+4

34

XI) 1 63-5t...
3 7 15 31 63
An) 2'4'816'32.s..

2-for
2
n20
Example T.5.12 Determine the inverse z-transform of the following X(z) by partial fraction
expansion method,
(z +2) 3
Xz) 2-7z+3 if the ROCs are (i)|=l>3 (i) |ak; and (ii)
2
Dec-12, Marks8
Solution

X(z) = (z+2)
3
x() (z +2) (z +2)

-2/3,1/3-1
Z-3 1
2

TECHNICAL
PUBLICATIONS An up thrust for knowledge
 7 63
S g n a l sa n aS y s t e m s

-Transforms-
1/3
X(E) 1-3-1
1-

ROC |z|> 3
for
x(n)
s lie at z =3 and = Since ROC of |z|>3 is exterior to both the poles, all
The poles

will be causal. i.e.,

sequences
the

x(n) =(7)+(3" u(n)-5| u(n)

for ROC Iz1<

(i)x(n)
is inside of both the Hence sequences corresponding to both
The ROC of |z< poles.

at z
=3 and z =will be noncausal. i.e.,
the poles

x =8(r)-3)"u(-n-1)+|u(-n-1)

Here note that, ROC foro (n) is entire z-plane.

i) x(n)for ROC z<3

The ROC of < |ak 3 is the annular region between poles at z=s and : =a Hence

Stquence corresponding to pole z = w i l l be causal and that corresponding to pole = 3

will be
noncausal. i.e.,

X() -5n-(3"u(-n-1)-2
Exam Findthe time domain signals corresponding to the following z-transtorms
i) X(z) =
1--2z/>1

)
X(2) an/Rdb&07Mark B
=
cos(22): |z<

TECHNICAL PURLICATIONS An up thrust for knowledae

 7- 64
Signals and Systemns
z-Transforms-
Solution

i) X = 1-2z/>1

2-1 (: -1)(z +1)

X(E)
-1E+1-1:1

XG)211
Since |>1, the sequence is causal for both the terms.

u(n) for even n

x(n) =lu() +(-1)"u(n)] 5[1+(-1)"]u(n) =
=

0 for
for odd n
ii) X() = cos(2z), | z< o

cosr= y(-1)" 2n =1- **°°***°°

n= 0 (21)! 2!4! 6! 8!

cos(22) =1-2)(22)_(22) (22)8 °°*°**°

2! 4 6! 8!

8
8!
And X(z) =
...x(-8)z + x(-7)z + x(-6)z°+r(-5)z +(-4)z4 +(-3)
+x(-2)z+x(-1)z+x(0) +x(1)z+.
One
comparing above two
equations we
get,
) 28 ,
0 024, 0 , 2 0 ,

Example 7.5.14 Find the Inverse

z-transform of
i)X(2)
ROC<lal<
i)
2:2_11ROC
2 -z+1 ROC: 1 z{k2

TECHNICAL
PUBLICATIONS An up thrust for
knowledge
 ms
7- 65

Solution:
z-Transforms-
z +z2
X(z)
ROC:<lak
X(E) 3+Z
(z +z) 4=

(z +Z)
z-A
X(2) = 4
(3 +2)
1
3 1
1 Z

2 1
1- 1
Here, 2 n) 4, 1 for ROC:z> 1

and u(n) « for ROC:|z>

1-

Doth the sequences are causal for ROC of |z|> By applying time shifting

POperty and taking inverse z-transtorm,

1+ 1
V+ 3 n+T
x(n) =4 u+1)
u(n+3) u(n +3) +| ufn +1)-

X
ROC:1<|zl<2
-222- +1
Z

z2-Z-2
X(z) 1 1/3
(2-2)(2+1)Z-2 z+1

X(2)=1 1
1
31-2z-1 1+2J
Signals and Systeni
7- 66 z-Transsm
Here sequence corresponding to pole at z = 2 will be non causal since ROC is1- 2
and sequence corresponding to pole at z =-1 will be causal since ROC is zl>1 t

(n) I-2"u(-n-1) --1)"u(7)]

Eampe 7 Deternmine inverse z-iransfonm of the following

i) X(2) = sinz
JanFeb 04, Marke
11N(E (2-4 3)
(-3)(z-2}(7=1
ROC of a) 2<|z|< 3

b)z|>3
c)|z<1
ii) X(-)=7
1-1:" use convolution property. JulyAugs 06 Marke 5

TECHNICAL PUBLICATIONS An up thrust for knowledge

 Signals and Systems 7- 54
z-Transforms-
Solution : i) X(z) sin z
()'221*1
sin (2) =
i
1=0 (2 +1)!

Z- 3 ! 5!7 .

z° +x(-5) z° +x(-4) z"

X(z) =
..x(-7) z +x(-6)
+x(-3) z° +x(-2) z* +(-1) Z+ x(0)+x(1) z"+ x(2) z*2
On comparing above equation with sin(z) we get,L

) -
050,0, 1,o

s
ii) X(:-3)(z-2)}-1) (2-4z+5s)

X) 2-42+5)
(z-3)(z-2(z-1)

1
-32
1 1 1
x() 1-3z1-2z-1 1-z-l
a) To obtain x(n) for 2<|z|<3
X(z) has poles at =1,
2 and 3. Hence sequences due to
z will be
poles z = 1 and
causal. Sequence corresponding to pole z 3 will be noncausal. =

x(n) =
-3"u(-n-1)-2"u(n) +u(n)
b) To obtain x(n) for |z|>3
Since all the poles are inside the ROC, all the sequence will be causal.

n) =
3"u(n) 2" u(") +u(n)
-

- 3-2+1]u(r)

TECHNIOA DIIDI inTiC

 55

To
obtain x(n) for z1. z-Transforms-I
the poles are outside of
ROC, the
7) sequence will be noncausal.
3" u(-n i.e.,
=
-1)--2"u(-n-1)+u(-n-1)
-3+2" +1 u(-n-1)
ii) X(:)

X) =
X() X2(2)

X)

X2(2)= = z(")=- )

By convolution property,

():(7) X,(2) X2()

(n)=A1(n) x2()
(n) k ) 2(-k)

un-k)

uk) u(n-k)

0sksn. Hence,
tlere (k) u(n-k)=1 for

PUBLICATIONS
An up thrust for knowedge
CHNICAL.
IE
Signals and Systems 7- 56
z-Transforms-
Example 7.5.7 Find the inverse z-transform of the following

i) X(z) =log 1 - a z - 1 >|a|

1
i) X(z) =log :kla|
1-az
ii) X(z) = ez

Solution: i) 1
X(z) =log (1 -az-1)"

-log(1-az")
log(1-p) is expanded with the help of power series. It is given as,

log(l-P) IPl1
n=1

X(z) = foraz<1 or z|>|a|

n
1=1

n=1
n
-11
2r(n) :
n=1

Here x(n) = J for n21

n

Otherwise

a"
.e x(n) =
n
u (n-1)

1
ii) X(z) =log z<a
1-az
-log(1-al:)
a 1 or |=I<la|
n=1

n
n=1
g n s i sB n aS y s t é l

will be
mthen above equation

()
L en
t =

Z-
-m m
X() m=-1

-177
2x(m) z
m=-1

for m=-1 to- co

a(m) otherwise
Here

x(n)= "u(-n-1)
1

ii) X(z)-e

log X(2)
sides with respect to z,
Differentiating both

1dX(:)
E) d
dX() xE)
dz
Multiplying both sides by -z, we get,

-NE) = X )

aking inverse z-transtorm,

1
a(n) =a (n1)
OT
() = x(n-1)

initial value is
given as,

lim lim
a(0)= X(E)
=1
1
r(0) =
1 i.e r(0) =

An up thrust for knowledge

TECHNICAL PUBLICATIONS
 Signals and Systems 7- 58 z-Transforms-
and for n =1,
()=()=1 =
n=2,

n = 3,

n = 4,

Thus, x(n) u(n)

X() using partia fraction expansion
Find the inverse z-transform of
Example 7.5.8
approach:
+1 July/Aug.-11, Mariks 6
X(z) -ROC:|Z|>1
3z2-4z+1
Z+1
Solution. : X (2)
z z (3z-4z+1)
z+1 (7.5.10)

.(7.5.11)
+ 1 ,-1
2
X() = 1--
1-z

O
=andz =1. The ROC
From equation (7.5.10) observe that the poles lie at z=0, z

Hence time domain sequence will be causal. Taking inve

z>1 is exterior to poles.
z-transform of equation (7.5.11),

x(n) 8()-2ut)+u(m
Example 7.5.9 Find the inverse z transform of the following :
z+2
1/812>1/2 i) X()=-a2.
i) X2-3/4z+1/8
z-a
20
culy/Aug 11, Marks 8, July-13, Marks
TECHNICAL PUBLICATIONS-An up thrust for knowledge
Signalsa n d
Systems 7- 59 1-Transforms-I

i)Thegiven X(z) can be expressed as

z(z* +z)
X() 1

z+z . (7.5.12)
X) 1
of

Hence let us perform an actual division as

are is not a proper function.

Here Z

follows
3
Z+4

3+z

3
1632

3
16
written as,
can be
Ihus equation (7.5.12) 3
23 3
3
+
16
X() *3 16 32
3-*s Z

57/2 17/16
Zt 1
4
17/16
5/2
X(2) =1 -
Signals and Systems 7- 60
z-Transforms-
Taking inverse z-transform,

xo) =
ö(n +2)+ ön+1)+ u-16l u(n)

1-az
i) X(z) =

1
az-1 2-11
a
z-l
a-z 1-az-l 1-az 1-az-

Taking inverse z-transform of above equation,

X() = a"-l u(n-1)-a" u(n)

a
Example 7.5.11 Using long division, determine the inverse z-transform of
X(E) = - 1

1-
when the region of convergence is |z |>1.
July/Aug.-06, Marks 4
 Signals and Systems
7-62 z-Transforms-1
Solution

+4

34

XI) 1 63-5t...
3 7 15 31 63
An) 2'4'816'32.s..

2-for
2
n20
Example T.5.12 Determine the inverse z-transform of the following X(z) by partial fraction
expansion method,
(z +2) 3
Xz) 2-7z+3 if the ROCs are (i)|=l>3 (i) |ak; and (ii)
2
Dec-12, Marks8
Solution

X(z) = (z+2)
3
x() (z +2) (z +2)

-2/3,1/3-1
Z-3 1
2

TECHNICAL
PUBLICATIONS An up thrust for knowledge
 7 63
S g n a l sa n aS y s t e m s

-Transforms-
1/3
X(E) 1-3-1
1-

ROC |z|> 3
for
x(n)
s lie at z =3 and = Since ROC of |z|>3 is exterior to both the poles, all
The poles

will be causal. i.e.,

sequences
the

x(n) =(7)+(3" u(n)-5| u(n)

for ROC Iz1<

(i)x(n)
is inside of both the Hence sequences corresponding to both
The ROC of |z< poles.

at z
=3 and z =will be noncausal. i.e.,
the poles

x =8(r)-3)"u(-n-1)+|u(-n-1)

Here note that, ROC foro (n) is entire z-plane.

i) x(n)for ROC z<3

The ROC of < |ak 3 is the annular region between poles at z=s and : =a Hence

Stquence corresponding to pole z = w i l l be causal and that corresponding to pole = 3

will be
noncausal. i.e.,

X() -5n-(3"u(-n-1)-2
Exam Findthe time domain signals corresponding to the following z-transtorms
i) X(z) =
1--2z/>1

)
X(2) an/Rdb&07Mark B
=
cos(22): |z<

TECHNICAL PURLICATIONS An up thrust for knowledae

 7- 64
Signals and Systemns
z-Transforms-
Solution

i) X = 1-2z/>1

2-1 (: -1)(z +1)

X(E)
-1E+1-1:1

XG)211
Since |>1, the sequence is causal for both the terms.

u(n) for even n

x(n) =lu() +(-1)"u(n)] 5[1+(-1)"]u(n) =
=

0 for
for odd n
ii) X() = cos(2z), | z< o

cosr= y(-1)" 2n =1- **°°***°°

n= 0 (21)! 2!4! 6! 8!

cos(22) =1-2)(22)_(22) (22)8 °°*°**°

2! 4 6! 8!

8
8!
And X(z) =
...x(-8)z + x(-7)z + x(-6)z°+r(-5)z +(-4)z4 +(-3)
+x(-2)z+x(-1)z+x(0) +x(1)z+.
One
comparing above two
equations we
get,
) 28 ,
0 024, 0 , 2 0 ,

Example 7.5.14 Find the Inverse

z-transform of
i)X(2)
ROC<lal<
i)
2:2_11ROC
2 -z+1 ROC: 1 z{k2

TECHNICAL
PUBLICATIONS An up thrust for
knowledge
 ms
7- 65

Solution:
z-Transforms-
z +z2
X(z)
ROC:<lak
X(E) 3+Z
(z +z) 4=

(z +Z)
z-A
X(2) = 4
(3 +2)
1
3 1
1 Z

2 1
1- 1
Here, 2 n) 4, 1 for ROC:z> 1

and u(n) « for ROC:|z>

1-

Doth the sequences are causal for ROC of |z|> By applying time shifting

POperty and taking inverse z-transtorm,

1+ 1
V+ 3 n+T
x(n) =4 u+1)
u(n+3) u(n +3) +| ufn +1)-

X
ROC:1<|zl<2
-222- +1
Z

z2-Z-2
X(z) 1 1/3
(2-2)(2+1)Z-2 z+1

X(2)=1 1
1
31-2z-1 1+2J
Signals and Systeni
7- 66 z-Transsm
Here sequence corresponding to pole at z = 2 will be non causal since ROC is1- 2
and sequence corresponding to pole at z =-1 will be causal since ROC is zl>1 t

(n) I-2"u(-n-1) --1)"u(7)]

 sin 2.
SP-4.23 Find the inverse 2-transform of X (2) =

Solution
Expansion of X(2) in a Taylor's series about z =
0 is as follows:

dX()| 22 dX(z)
X() =
X(E)|z=0 + z
dz 0
2! dz2 +

5
=z T
5!
 The 2-Transform 441

Because, X(2) = T=-0o

r(n) z""
he cOefficients of 2 in Taylor's series expansion can be associated with s n )

Thus, o(n) =
( 5
. 0,19
SP-4.24 Find the inverse 2-transform of X(2) given below:

z+2 e-3T-2z e-2T ROC zl> e-27

X()2-[e-2 +e-3T] 2 + e-
T

Solution
X(2) z + e 3 -2 e-2T
(z(z-e-3T) (z - e-27T)
K1 K2
2 -e - 3 7 Z - e-2T

Ki = } t e s - 2 e - 2 7

= 2
where, -e-27
I2=e-3T

z+e31- 2e-2T|
= -1
-e-3T
2=e2T

Thus X(:) =2--3T 2 e-2T

are surrounded by the ROC. Hence, r(n) will be

The poles at z =
e3 and z =
e
à causal sequence.

z(n) = 2 (e-")" u(n) -

(e-*")" u(n)
= 2e-37 u(n) - e 2 u(n)

jlmz)
Circle:=e-27
z-plane

Re(z)

ROC
 SP-4.26 Using 2-transforms, find y(n) = z(n) * h{n),

where a(n) = a" u(n)

and h(n) = B" u(n)

Solution
X() =-a
X(2) ROC: lal >
a
H(2)
)= B ROC: z> 1B|
 The Z-Transfonm 4551

Gioen yn) = z(n) h{n)

Y(z)= X{z)H(z)
ROC: z maz (a, 3)
-oz -3)
Y(2)
(z-a)(z-)
K2
E
B+-3)
(z a) (z-5

where

-3
(a-)

Y()
Thus,
To-3t-a(o ) (z -

5)

Tuerefore, un)=
yn)(a )uln)-q-3u n )
a+1 - g

a - 3 ) un). a 43
er =3,
Y(z)= z-o

Y a

A2 = z=
A d
.2= 1

Y'z)=- zHOC:
-o ROC z a
yn)=a" uln) + n u T
 452 Signals and System

SP-4.27 Using 2-transforms, find

y(n) = r{(n) * h{n)

where, r(n) = a"u(n)

h(n) =a"u(-Tn), 0 <a<1.

Solution
Given z(n) = a" u(n)

X() ROC: lz| > lal

a

Since h(n) = z(-n), applying the time-reversal property of 2-transform, we get

H(a)=T-iROC: |2|>ja
H) 1-az ROC: a
We want to find y(n) =
zc(n) * h(n).
Taking 2-transform on both the sides, we get

Y() X(a) H(E) ; ROC: (2|> n (l2|<ai

-1
ROC:a<|al <
a
(z-a)(z-
Y()_ -1 1
Let K1 K2
a (z-)(-)-a)*-

where,
K a (-
1

Iz=a *1-0)
1

a z-a|z=1
(1-a2)
Thus, 1
ROC:a< |<
The pole at z
Y)-?-a)1-a?2-
is outside the ROC and hence will
=

give the negative-time sequen

part of y(n). However, the pole at z a is surrounded by the ROC and hence
=

the causal part of the sequence y(n).

proviue
T
Therefore,
un)=za" uln) + u- -1)
1
1-a
Example 8.2.1 Determine the pole zero plot of the following function
1
X()= 1-az-l ... (8.2.4)

And find out the time domain sequence for ROC of |zl> a.
Solution: Let us first make powers of z
positive in X (z). i.e.,
X (z) = 1

(z-a)
1.e. X (Z)
Z-a
which can also be written in standard form
as,
X(2) =2-02- 21
Z a
z-P1
Thus X (z) has one zero at z =
z1 =0, i.e. origin and X (z) has one
pole at z
=p1 =a
Fig. 8.2.1 shows the
pole zero
plot.
The poles and zeros can be
written as combination of real part and
Thus Z1 and pj obtained in this imaginary part.
example can be written as,
Z1 = 0+j0 and
P1 a+j0
Both 21 and pj have zero
imaginary parts. Actually Z1 is at origin.
Im(z)

Zero at
Pole at
z 0+j0 P1 a+j0
Z (0,0)
P (a,0)
0
Re(z)

Z-plane
Fig. 8.2.1 A pole zero
plot of X (z) given by
A circle "O' indicates zero and a cross 'X' equation 8.2.*
TECHNICAL PUBLICATIONS"- An
indicates pole.
uo thrust for knowledge
S i g n a l s a n a S y s t e m s

8-7 Z-Transforms 2

The time domain Luence of X (z) for ROC of |z|> a is given as,
x (n) = a" u (n), causal sequence

ROC of XX(z) is for | z|> a, the pole at pi =a does not lie in the ROC. Thus ROC
Since
ntain any poles of X (:).
does n o t
T f x(1) is a causal sequence with a<1, then it will be decaying in nature. Fig. 8.2.2
shows the pole zero plot and sketch of such sequence

Im(z)
ROC 1z|> Jal x(n)

x(n) = a"u(n)

Re(z)
0

Unit circle

decaying signal
Fig. 8.2.2 Pole lying inside the unit circle represent
And ROC : [z|>|al represent causal signal

pole will
then it will rise exponentially. The
I f x(n) is causal sequence with a>1,

lie outside the unit circle since a>1.

Determine the z-transforms, ROC, and poles and zeros of

Example 8.2.2
i) x(n) = 2"u(m) + 3" u(-n-1)

July/Aug.-2000, Marks 4: Jan./Feb. 03. Marke 12

ii) x(n) =
eS"u(n-1)
Solution i) xtn) = 2" u(n) +3" u(-n-1)
1
1z Jal
We know that
a'u(n)
1-az-1'
y 2>>la|

-u(-n1-1) 1 z<b
and
1-bz-1

e Z-transform of given sequence becomesS,

1 z> 2 and |z|< 3

X(?) 1-2-1 1-3 z

PUBLICATIONS An up thrust for krow'edgo

TFCHNICAL
Signals and Systems 8 8
z-Transforms-2
or 1 1
X(?) =

1-2 z 1-3 z-1

ROC 2<|z|< 3

-Z

(z-2)-(z-3)
Poles are at : z =2 and z = 3

And Zero is at z =0

ii) x(n) =
eS" u(n-1)
Z-transform is given as,

X(z) =2(r)2"
N-oo

e3 u-1)2
n=-o

Here u(n-1) =
1 for n 2 1. Therefore above equation becomes,
X() = 2e-3 "
1=1|

n=1

=
1 +(e321)2 +(e1)3+..t.
= e

e-1 1
1-3fle,-"|e1

1-e 1' forlel<1

Here
e-<1 is same as
|z|> e. Hence the
z-transform becomes,

_3.T
1-32-1 ROC: |zl>e
3
~e-3

TECHNICAL PUD! IrATIONS", An n

Signals and Systems 8-9 z-Transfon

Thus the pole is at : z =

e°.
There are no zeros.
Example 8.2.3 A DT system has a zero at (-0.2) and two poles at 0.5 et ]T/4

i) Find the transfer function of the system.

ii) Find the impulse response of a system in a closed sinusoidal form.

ii) Find the frequency response of the system at the frequencies: 0, and r

Solution: The zero is at, Z -.2

poles are at, pi = 0.5 el7/4 and P2 =

0.5e ]T/4

=
0.5 cos+j
4
sin

1
1
-0.5
TECHNICAL PUBLICATIONS An up thrust fo.
-

knowledge
 8 11
Sgnas
a dS
s andSystems z-Transforms 2
1

Siilarly
P2 2222
Todetermine transi
transfer function

T h e t r a n s t e r

function
f u n c t i o n
is given as,
er
z +0.2
z-Z1
H()(:-P1)(z-P2)
z+0.2

function of the system.

transter
This is the
an impulse response
To obtain

Letus write
H)
Z

Z +0.2
H()
1

0.8-0.4-j1.8139 -0.4 +j 1.8139

22 2
(-0.4-j1.8139) (-0.4+j1.8139)
H(z) = 0.8+

hn) 0.88(n)+(-04 -j1.8139) t5| )

=

+(-0.4 +j1.8139)| )
0.88(n) +(1.8572-1.787)(0.520.785)" u(n)
+(1.85721.787)(0.52-0.785)"u()
Magnitudes anddngles
anEE in above equation can also be written as,

TECHNICAL PUBI ICATIOS". An un thrust for kriowledae

 8- 12
Signals and Systems
Z-Transforms -2
h(n) = 0.88 (1) +1.857 e]1787 (0.5e1075) u(n)

+1.857e/1787 (0.5e-j0785)u(n)
= 0.88(7) +1.857e)1787 -(05)"ej0785n

+1857 e/178 (0.5"e j0785n

(0785n-1787) 1
=
08(n) +1.857(05 "[e(O785-1787) +e
=0.88()+1857(05) "2 cos(0785n-1787)u(n)

= 0.88(n) +(05)" 3714 cos(0785n-1787)u(n)

This is the sinusoidal form of impulse response.

ii) Frequency response at o = 0, and t

We know that,

H(z) =
z+0.2
1 1
2
V2z 4

Frequency response can be obtained by putting z =e . i.e.,

e+0.2
H)
At 2 = 0,
H 2.2
At
H( =
-0.8-j0.57 i.e. 0.9893 3.758

At 2 = T
H ) -04
Above response indicates
lowpass type filter.
Example 8.2.4 A difference equation of the system is given below
y () = O.5 y (n-1) +x(n)
Determine
i) System function
ii) Pole plot of the system function
zero

iii) Unit sample respomse of the

system

TECHNICAL PUBLICATIONS An up thrust for knowledge

 Signalsa n a S y s t e m s

8 13
z-Transforms - 2

lutioni)
To obtain system fucntion

7-transform
Taking
of the given difference equation,
Y (z) = z {0.5y (n-1) +x(n)}

Anplving the linearity property of z-transform,

Y (2) 0.5 Z {y (n-1)} + Z

=
{« (7}
Applying the time shift property of z-transform,
Y(z) = .5 z i Y(z) +X (2)

X(z)
1-0.5 z") Y(z)
=

1 . (8.2.10)
H(2) X(2) 1-0.5 z
This is the required system function.

ii) Pole zero plot

functions to make powers of z positive i.e.,
Let us rearrange the system
Im(z)
H() 1z
1-0.5 Z-plane

Z Z-Z1
Z 0.5 z-P1 0 0.5 Re(z)
Ihus H (z) has one zero at z1 = 0 .e. at

origin and pole at

H() has one

P1 0.5 of the system

8.2.3 Pole zero plot
Fig. 8.2.3
Fig. 8.2.3 shows the pole zero plot. given in example
ii) Unit
sample response inverse
z-transform of system
The h (n) is obtained by taking
t
sample response

function H (E). i.e.,

h (n) = IZT {H (2)}

8.2.10
1 equation
1ZT from
I1-05
=

The inverse z-transform of above equation as,

h (n) = (0.5)" u (n)

knowledge
thrust for
TECHNICAL
PUBLICATIONS - An up
 8.2.3 Causality and Stability interms of z-Transform
Causality
The condition for LTI
system to be causal is given as,
h (n) == 0 n< 0

Here h (17)is the unit sample

response of the LTI system. When the sequence is
causal, its ROC is the exterior of the circle.
Hence,
LTI system is causal if and only if the ROC of the system function is exterior of a circle of
radius r< o,
This is the condition for
causality of the LTI system in terms of z-transform.

Stability
Now consider the condition for
stability. A necessary and sufficient condition for the
system to be BIBO stable is given as,
co

.. (8.2.11)

We know that the

system function is given as,

H(z) = h (n) z
n= - o
requlres nat n e u t cicTe sitouTa De Presenn u ue NOC OI TT(A: Tuu5,
LTI system is BIBO stableif and only if the ROC of the system function includes the unit
circle.
 Z-IranSiorms -2

mple 8.2.6 A discrete LTI system is characterized by the difference equation

) + y(n-2)+ x(n-1). Find the system function H(z) and indicate the ROC
e . (i) Stable (1) Lausal. A'so determine the unit sample if
response of the stable
of system
July-13. Marks 10
: (i) ROC
on
Salu

v(n) v (7-1)+ y(n-2)+ x(n -1)

Jakang transform,

) Y(2)+zY(z)+2 X
Yell--z]='x(2)

Y()
H)XX) --1 -0.5-1.25) (z -0.5+ V1.25)
im(z)
Unit ROC
(21.618)(2 + O.618) circie

1.618, P2-06l8
Poleslie at P1
0
P
Zeros lie at 2 o5 11.5 Re(z)

ROC for stable system

ncudes unut
stable ystem

shown in Fig. 8.2.5.

orcde 1his is

0.618< 2 16l8
Hence ROC stable system
Fig. 8.2.5 ROC for

ROC tor causal system

ROC is outside Im(z)
For the causal s t e m , ROC
This s
tirnite radius.
Soe circle of 21618

sthown in Fig 8.2.6. Hene,

KOC 1u18 Re(z)
of stable 0,5
Unit sample response

syste
as,
he stable system is given

H(2) (z-1618)(2+0.618)
system
for c a u s a l
8.2.6 ROC
Fig.
|z/< Loi8 0.447
0447
-1.ol8 (+0.618)

o18)(+ Uoi8)

kiowieoge
mrusl fr
p
A
 Signals and Systems
8-718 Z-Transforms 2
0.447 -0618
H(z) =
0.447 with < 1.618and |2|>
1-1.618 z- 10.618z-
corresponding to pole pi =1619
Here ROC is hence time domain sequence,
|z< 1.618,
will be noncausal. And |z|>-0.618,
hence time domain sequence corresponding to nelle
with this ROC,
inverse z-transform of H(z)
P2 -0.618 will be causal. Taking

h(7) = 0.447 [-(1.618)"u(-n-1)-(-0.618)"u(n)]|

This is unit sample response of a stable system.

Consider a causal LTI system function.

Example 8.2.7
H(2)=a7 wlhere 'a is real.
1-az-l
Determine the value of 'a' for which the system is stable. Show graphically, for 0< a< 1, the

pole zero plot and ROC.

Solution: To obtain value of 'a' for stability
function as,
Let us rearrange the given system
2al

H(z) Z-a multiplying numerator and denominator by z

Im(z)

H(z) = ROC |z|> Jal Pole of H(z)

1.e z-a
Zero of H(z)
Clearly this system has one
pole at z =p1 a. Since the
=

system is causal, its ROC will Relz)

be exterior of the circle. WVe
know that poles do not lie in
the ROC. Hence ROC of H(z)
will be for |z>|a. For the
causal system to be stable, its
ROC should include unit circle.
Fia. 8.2.7 Pole zero plot of H(z) and its ROC. 1ne po
is at z = P1= a and zero is at z = Z1
i.e.,
And 0<a<1
z>al<1
Pole zero plot and ROC for 0<a<1.
Fig. 8.2.7 shows the pole zero plot for 0<a<1.

Example 8.2.8 Compute the response of the system.

fhe
y (n) =07 y (7-1)-012y (n-2) +x(n-1) +x(n -2) to the input r(n)= nu () Is
system stable ?

TECHNICAL PUBLICATIONS -
An up thrust for knowledge
 S i g n a l sa n a S Y s t e m s
8- 19 z-Transforms 2

S o l u t i o n

1: To obtain
system function H(z).
Step

given
difference equation is, (since no initial conditions are given),
The
u()= 07 y (n -1)-0.12 y (n -2) +x(n-1) +x(n-2)

z-transform of this equation becomes,

Y()= 07z Y(z)-012z< Y(:) +2" x (2)+ z2 X(2)

(1-07+012:)
Y()= (+:2) x()
z+2-2
Y()
X() 1-07z +012 z-2

= H(2) i.e. system furnction, we have,

Since X(z)
+z-2
H( 1-07z+0.12 z
+1
z - 0 7 z +0.12

From H(2)
Stability of the System
follows
factored form as
be expressed in
H(7) can
z+1
H) (z-04) (z -0.3)

and = 0.3 and

Poles P1 0.4 P2

Zeros
are inside the unit
circle |z| =1. Hence
of this system
s both the poles pi andp,
this system is stable.

Step 2: To obtain X(z)

Ihe input to the system is,

x(n) =nu (7)

X()=
1 - ) (:-1)
ITOS An up thrust for knowledge
Signals and Systems
8- 20 z-Transforms- 2

Step 3 To obtain Y(z).

The z-transform of output Y (2) is given as,

Y(z) = H (2) X (2)

Z+1
(z-0.4) (z-0.3) (z-1)
inverse z-transform.
Step 4: To obain y (n) by
The above equation can be written as,

Y(z) z+1
.(8.2.16)
z
(z-0.4) (z -0.3) (z-1)*
A1
Z-04Z-0.3
A2+A4.
z-1 (z-1) As . (8.2.17)

Here A, Az, A4 can be obtained

Z+T
A (z-04) 2
lz=0.4 (z -0.3) (2 -1)*|=04

0.4+1 38.89
(0.4-0.3) (04-1)
Y(2) Z+1
A2 (2-0.3) Z z=0.3
(z-0.4) (z-1) |2=0.3
0.3+1
-26.53
(0.3-04) (0.3-1)
Y(2) +1
A (-1) (z-0.4) (z-03)=1
1+1
= 4.76
1-0.4) (1-0.3)
A3 can be obtained by equation

A3 Z
Jz=1
d Z+1
dz
(z-04) (z -0.3)
TECHNICAL PUBLICATIONS An up thrust for knowledge
-
 21
SgnalsBr7 SYste71s

( 0 4 )(-0.3) 1-(+1) (2z- 07)

( 0 7 :012) -
-12.35

8.2.17 becomes,
cquation
Thus
38.89 26.53 12.35 4.76
(E) - 04 -0.3 z-1 (z-1)

26.53z 12.35z 4.76z

38.89 +
Y() -0.4 -0.3 z-1 (z -1)

12.53 4.76z
38.89 26.53
Y(E) 1-042 1-0.3 z-1 1-2-

(n) 476-n (1)" u (r)

-1253 (1)" u
+

(n)
38.89 (04)" u (n)-26.53(03)"
u
=
n)

1253+476 nu (m)
38.89 (04)" 2653 (0.3)"
-

response of the system

sstem

and unit sample

Determine the system function
Example 8.2.9
described by the difference equation,
=0
y(-1)
(n)-v(n-1) =2 x("),
1s,
difference equation
Solution: The given

Y(n)y (n-1)
= 2 x (n), y(-1) =0
of above equation,
unilateral z-transform
aking

Y){y (2)+y(-1)]-2 X() z-transtorai

same as
becomes
Putt 0, above equation
*l /-1) =

Y(:)- Y ( : ) = 2X (2)
YE)2
Y(2)
H(2) X()

h(n)= 2 u(1)

aoaedye
P U B L I C A T I O N S
An up
thrust r
TECHNICAL
 Signals and Systemss 8- 22
z-Transforms-2
Example B.2.10 Determine the system function for the causal LTI system with difference
nce
eguation
ym-y-1)+uo-2) =t

Using z-transform determine y(n) if

x(n) = u(n)

Solution

Step 1 : To determine system function

Taking z-transform of the given difference equation,

Y)-Y)+Y(a)=
4 X(a)
X(z)

Y(2) 1
H(Z) X(Z) .(8.2.18)
1-
This is the system function of LTI
given systen.
Step 2 : To determine X(z)
The input is,
- u(n)

Taking z-transform, X(z) =

1-
Step 3 : To obtain Y(z)
Y(e) =
H(E)- X(2)
1

TECHNICAL PUBLICATIONS"- An up thrust for

knowledge
 8- 23 z-Transforms 2
S i g r n a l sa n a S y s t e m s

z2 Z

above equation into partial fractions.

let us expand
Now

Y A1 Ai A2 . (8.2.19)

= - j 0.577

Here, Ay- Z

Aj and
0.577, i.e. complex
conjugate of
Therefore Ai =
j
= 1

becomes,
Therefore equation 8.2.19
- j0.577 0577 + 1
Y(2) 1
Z

1
- j0.577 j0.577
-1
Y(z)=

Taking inverse z-transform,

u(n) +j0.577
y(n) =-j0.577
L90(0.52-60)" u(1)
260)"u(n) +0.577
0.5772-90(0.5

u(n)

knowledge

thrust for
An up
URL
IGATIONS
-
Signals and Systemns 8- 24
z-Transforms-2
=
(0.577 e/90 ) (0.5ei60)"u(n) + (0.577e (0.5e"/60)"u(n)
u(n)

0.577(0.5" |e10.c1on +e1M0.e160 \l(n) + un

=0.577(0.5)" [el601-90) +e-i(60n-90) lu(n) + u(n)

We know that ele +eJ0= 2 cos 0. Hence above equation becomes,

y(1)= 0.577(0.5)" 2 cos(60n-90)u(1n) +(0.5)"1u(in)

[1.154 cos(6011-90) +1 |(0.5)"u()

This is the required output.

 Emple 82.13 Given: H)2 +1, ROC:|z |>1
i) Is this
system causal ?

i) Is the
system BIBO stable ?
111) Find the unit
samvle response of the sytem.
0) Find the
difference equation realization of the system ana

v) Is the system linear ? July/Aug.-11, Marks 10

Solution : i) Causality :
Here ROC is |z|>1, means it is exterior of circle of radius |z|=1. Hernce this system i1s
causal.

ii) Stability:
The ROC is |z |>1. This the unit circle
means |z =1 is not included in the ROC. Hence
the system is unstable.

ii) Unit sample response of the system:

H(z) =

2 +1

z,-1

1+z-2

Let us
rearrange above equation as,

An up thrust for knowledge

TECHNICAI. PUBLICATIONS
-
 Signals and Systems 8- 28
z-Transforms 2
zsin 1 T
H(z) = Here
,
sin=1, 2
1-2z cos+z2
2
laking inverse z-transform of above cquation,

h (n) =
sinn lu()
iv) Difference equation:

-1
H() =
1+z-2

-1
Y() since H
Y(z)
X(z) 1+z-2
,
(2)x(2)
Y()1+ z7] = z-l x(2)

Y(2)+2Y(E) = 21X(2)

Taking inverse z-transform of above equation,

y(m) +y(n-2) =
x(n-1)
v)Linearity:
The system is linear.
 ple 8.2.14A difference equation of the system iS given u5,

or)-y(n-1)+ y (n-2) =x(n) +x (n-1)R* (n-2)

eterine the transfer
function of the inverse system. Check whether the inverse system 1s
causal and stable.
Solution: July/Aug-11, Marks 4
Taking the z-transform of the given difference equation
Y()- zly
()+2y ()=X(2)+1 x()-2X ()
Y(1-}-x
H()= Y(E)
X(z) 1-2-12-2
4

This is the transfer function of the

given system. The transfer function of the inverse
system is given by equation 8.2.23 as,

H() H() 1+21-z-2

1

1-z-14z -2

1-z-1+ - 2 z2-z+
4

11t1-2 -;)
Thus poles of inverse
system aree

21 -
and
22 2

Thes poles are inside the unit circle. Hence inverse system is causal and stable.
ese