Signals & Systems

Chapter 10: The Z-Transform

Adapted from: Lecture notes from MIT, Binghamton University
Hamid R. Rabiee Arman Sepehr Fall 2010

Lecture 15 (Chapter 10)

Outline
Introduction to the z-Transform Properties of the ROC of the z-Transform z Transform Inverse z-Transform Properties of the z Transform z-Transform System Functions of DT LTI Systems o Causality o Stability Geometric Evaluation of z-Transforms and DT Frequency Responses FirstFirst and Second-Order Systems Second Order System Function Algebra and Block Diagrams Unilateral z-Transforms z Transforms
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

The z-Transform
Motivation: Analogous to Laplace Transform in CT

x[n]
We now do not restrict ourselves just to z = ej

H ( z)

y[n] = H ( z ) { z

Eigen function for DT LTI

H ( z) =

n =

h[n]z n

assuming it converges

The (Bilateral) z-Transform

x[n] X ( z ) =
Z

n =

x[n]z n = Z {x[n]}

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

The ROC and the Relation Between zT and DTFT z

z = re j , r =| z |

X (re j ) =

n =

x[n](re j ) n =

n =

( x[n]r n )e jn

= F { x[ n]r n }

ROC = z = re j at which

| x[n]r
n =

| <

depends only on r = |z|, just like the ROC in s-plane only depends on Re(s) y p () Unit circle (r = 1) in the ROC DTFT X(ej) exists

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #1 p

x[n] = a n u[n] - right - sided

X ( z) =
This form for PFE and inverse ztransform

n =-

a n u[n]z n

n =0

1 (az 1 ) n

1 z = 1 az 1 z a
This form to find pole and zero locations

If az 1 < 1, i.e., | z |>| a |

That is, ROC |z| > |a|, outside a circle

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #2:

x[n] = a n u[ n 1] - left - sided

n

X (z) =

{ a
n =

u [ n 1 ]z
n

anz
n

n =1

=1

n =

(a

n=0

1 a 1 z = 1 = 1 1 1 a z a z 1 z = , za

If a 1 z < 1, i.e., | z |<| a |

6

Same X(z) as in Ex #1, but different ROC.

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Rational z-Transforms
x[n] = linear combination of exponentials for n > 0 and for n < 0

X ( ) is rational (z N ( z) X(z) = D( z )

Polynomials in z

characterized (except for a gain) by its poles and zeros

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

The z-Transform
Z x [ n ] X ( z ) =

ROC = z = re

n =

x [ n] z
n =

= Z { x [ n ]}
n

at which

| x [n ]r

|<

-depends only on r = | | j like the ROC in s-plane d d l |z|, just lik h i l only depends on Re(s) Last time: o Unit circle (r = 1) in the ROC DTFT X (e j ) exists o Rational transforms correspond to signals that are linear combinations of DT exponentials
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Some Intuition on the Relation between zT and LT

x(t ) X ( s ) = x(t )e st dt = L{x(t )}

Z

Let t=nT

= lim

T 0

n =

x ( nT ) ( e sT ) n T 123
x [n

= lim T
T 0

x [n ]( e

sT

n =

The (Bilateral) z-Transform

x[n] X ( z ) =

n =

x[n]z n = z{x[n]}

Can think of z-transform as DT version of Laplace transform with

z = e sT

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

More intuition on zT-LT, s-plane - z-plane relationship

e sT = z
j axis in s - plane ( s = j ) p
z = e jT = 1 a unit circle in z - plan p

LHP in s-plane, Re(s) < 0|z| = | esT| < 1, inside the |z| = 1 circle. Special case, Re(s) = - |z| = 0. RHP in s-plane, Re(s) > 0|z| = | esT| > 1, outside the |z| = 1 circle. Special case, Re(s) = + |z| = . A vertical line in s-plane, Re(s) = constant| esT| = constant, a circle in z-plane.

10

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Properties of the ROCs of z-Transforms

(1) The ROC of X(z) consists of a ring in the z-plane centered about the origin (equivalent to a vertical strip in the s-plane) g ( q p p )

(2) The ROC does not

contain any poles (same as in LT).

11

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

More ROC Properties p

(3) If x[n] is of finite duration, then the ROC is the entire zplane, plane except possibly at z = 0 and/or z = . Why?

X ( z) =
Examples:

n = N1

N2

x[n]z n

CT counterpart

12

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

ROC Properties Continued p

(4) If x[n] is a right-sided sequence, and if |z| = ro is in the ROC, then all finite values of z for which |z| > ro are also in the ROC.

n = N1

x[n] r1n

converges faster than g

n = N1
13
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

x[n] r0n

Lecture 15 (Chapter 10)

Side by Side
(5) If x[n] is a left-sided sequence, and if |z| = ro is in the ROC, then all finite values of z for which 0 < |z| < ro are also in the ROC. (6) If x[n] is two-sided, and if |z| = ro is in the ROC, then the ROC consists of a ring in the z-plane including the circle |z|| = ro. g p g | What types of signals do the following ROC correspond to?

right-sided
14

left-sided

two-sided

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #1 p

x[n] = b|n| ,

b>0

x[n] = b n u[n] + b n u[ n 1]

From 1 , z >b 1 1 bz 1 1 , b n u[ n 1] z< b 1 b 1 z 1 b n u[n]

15
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #1 continued p

1 1 1 X ( z) = , b< z < + 1 1 1 b 1 bz 1 b z

Clearly, Clearly ROC does not exist if b > 1 No z-transform for b|n|. z transform
16
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Inverse z-Transforms
X ( z ) = X (re j ) = {x[n]r n }, z = re j ROC
x[n]r n = F 1{ X (re j )} = 1 2

X (re j )e jn d

x[n] =

1 2

j X (re j )r n2n d e3 1 2 zn

for fixed r: z = re j dz = jre j d d =

1 1 z dz j

x[n] =
17

1 2j

X ( z ) z n1dz

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #2 p
5 1 z A B 6 X (z) = = = + 1 1 1 1 1 1 1 1 1 1 z 1 z z 1 z 1 z 1 z 4 3 4 3 4 3 3z 2 3 5 z 6

Partial Fraction Expansion Algebra:

X ( z) = b x[n] 1 1 1 z 1 4 b = x1 [n] 2 1 1 z 1 3 b + x2 [n] +

A = 1, B = 2

Note, particular to z-transforms: 1) When finding poles and zeros, express X(z) as a function of z. 2) When doing inverse z-transform using PFE, express 1 X(z) as a function of z-1.
18
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #2 (Continued) p ( )
ROC III: z >
1 3
n

- right - sided singnal

ROC II: 1 < z < 1

4

3
n

- two - sided singnal

1 x1 [n ] = u [n ] 4 1 x 2 [n ] = 2 u [n ] 3
n

1 x1 [n] = u[n] 4 1 x2 [n] = 2 u[ n 1] 3

n

ROC I:

z<

1 4
n

- left - sided singnal

1 x1 [n ] = u [n ] 4 1 x 2 [n ] = 2 u [ n 1] 3
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

19

Lecture 15 (Chapter 10)

Inversion by Identifying Coefficients in the Power Series

X ( z) =
Example #3:

n =

x[n]z n

x[n] - coefficient of z n

X ( z) x[ 3] x[ 1] x[4] x[n]

3 z 3 -z + 2 z 4

= 3 = -1 1 = 2 = 0 for all other n s ns

A finite-duration DT sequence

20

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #4:
(a)

1 X ( z) = = 1 + az 1 + (az 1 ) 2 + L 1 az 1 convergent for az 1 < 1, i.e., z > a

x[n] = a n u[n]
1 1 = a 1 z 1 az 1 1 a 1 z a 1 z (1 + a 1 z + ( a 1 z ) 2 + L ) a 1 a 2 z 2 a 3 z 3 L convergent for a 1 z < 1, i .e., z < a

(b) X ( z )

= = =

x[n] = a n u[ n 1]
21
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Properties of z-Transforms
(1) Time Shifting

x[n n0 ] z n0 X ( z ),

The rationality of X(z) unchanged, different from LT. ROC unchanged except for the possible addition or deletion of the origin or infinity no> 0 ROC z 0 ( (maybe) b ) no< 0 ROC z (maybe) (2) z-Domain Differentiation nx[n] z Derivation: X ( z ) = x[n]z n
dX ( z ) dz =

nx[n]z
n = n =

dX ( z ) dz d

same ROC

n 1

dX ( z ) z = nx[n]z n dx n =
22
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Convolution Property and System Functions

Y(z) = H(z)X(z) , ROC at least the intersection of the ROCs of H(z) and X(z), can b bigger if there is pole/zero cancellation. e.g. be bi th i l / ll ti

H ( z) =

1 , z >a za X ( z ) = z a, z Y ( z ) = 1 ROC all z

H ( z) =

n =

h[n]z n

The System Function

H(z) H( ) + ROC tells us everything about system ll h b

23

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

CAUSALITY
(1) h[n] right-sidedROC is the exterior of a circle possibly including z = :

H ( z) =

n = N1

h[n]z n

If N1 < 0, th the rerm h[ N1 ]z N1 at z = then th t ROC outside a circle , but does not include .
Causal N1 0
No zm terms with m>0 =>z= ROC

A DT LTI system with system function H(z) is causal the ROC of H(z) is the exterior of a circle including z =

24

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Causality for Systems with Rational System Functions y y y

bM z M + bM 1 z M 1 + L + b1 z + b0 H ( z) = a N z N + a N 1 z N 1 + L + a1 z + a0 No poles at , if M N
A DT LTI system with rational system function H(z) is causal y y () (a) the ROC is the exterior of a circle outside the outermost pole; ( ) () p y and (b) if we write H(z) as a ratio of polynomials

H ( z) =

then

N ( z) D( z )

degreeN ( z ) degreeD ( z )

25

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Stability y
LTI System Stable

n =

h[n] < ROC of H(z) includes

the unit circle |z| = 1

Frequency Response H(ej) (DTFT of h[n]) exists.

A causal LTI system with rational system function is stable all poles are inside the unit circle, i.e. have magnitudes < 1

26

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Geometric Evaluation of a Rational z-Transform

Example #1:
X 1 ( z ) = z a - A first - order zero

Example #2:

1 X 2 ( z) = - A first - order pole za 1 X 2 ( z) = , X 2 ( z ) = X 1 ( z ) X 1 ( z)

Example #3:
iR=1 ( z i ) X ( z) = M P j =1 ( z j )

X ( z) = M
R

iR=1 z i P=1 z j j
P i j =1

X ( z ) = M +

( z ) ( z )
j i =1

All same as in s-plane p

27

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Geometric Evaluation of DT Frequency Responses q y p

First-Order System one real pole
H ( z) = 1 z = 1 az 1 z a h[n] = a n u[n] , , z>a

a <1

H ( e j ) =

1 1 , H ( e j ) = 1 = , 2 2 2

H (e j ) = 1 2 = 2

28

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Second-Order System Two poles that are a complex conjugate pair (z1= rej =z2*)
z2 1 H(z) = = , 0 < r <1, 0 1 2 2 (z z1 )(z z2 ) 1 (2r cos )z + r z 1 sin [(n + 1) ] , h[n] = r n H ( e j ) = j u [n] j j j sin (e re )(e re ) Clearly, |H| peaks near =

29

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Demo: DT pole-zero diagrams, frequency response

18
30
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

DT LTI Systems Described by LCCDEs y y

a y[n k ] = b x[n k ]
N M k k k =0 k =0

Use the time-shift property

k =0

ak z k Y ( z ) =

k =0

bk z k X ( z )

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

k =0 N

bk z k ak z k

Rational

k =0

ROC: Depends on Boundary Conditions, left-, right-, or two-sided. ROC is outside the outermost pole For Causal Systems
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

31

Lecture 15 (Chapter 10)

System Function Algebra and Block Diagrams y g g

Feedback System (causal systems)
negative feedback configuration

H ( z) =

Example #1:

H1 ( z) Y ( z) = X ( z) 1 + H1 ( z)H 2 ( z)

H ( z) =

z-1

D Delay

y[n] =

1 1 1 z 1 1 4

1 y[n 1] + x[n] 4

32

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Example #2:

1 2 z 1 1 H ( z) = = 1 1 1 2 z 1 1 1 z 1 1 4 z 4

) Cascade of

two systems y

33

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Unilateral z-Transform

( z) =
Note: (1) If x[n] = 0 for n < 0, then

n =0

x[n]z n

( z) = X ( z)

UZT of x[n] = BZT of x[n]u[n]ROC always outside a circle (2) and includes z = (3) For causal LTI systems,

H ( z) = H ( z)

34

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Properties of Unilateral z-Transform

Convolution property (for x1[n<0] = x2[n<0] = 0)
UZ x1 [ n ] x2 [ n ] 1 ( z ) 2 ( z )

But there are important differences. For example, time-shift

y[n] = x[n 1] Y( z ) = x 1 + z 1 ( z )
Initial condition
y [n ]z
n

Derivation:
Y(z) =

n=0

x [n 1 ]z

= x [ 1 ] +

= x[ 1] + z 1

x[m]z m m= 10 4243 z

n=0

x [n 1 ]z

n =1

35

Sharif University of Technology, Department of Computer Engineering, Signals & Systems

Lecture 15 (Chapter 10)

Use of UZTs in Solving Difference Equations with Initial Conditions

y[n] + 2 y[n 1] = x[n]

y[1] = , x[n] = u[n]

1 z 1

UZT of Difference Equation

UZ{ y [ n 1]} 64 744 4 8 Y ( z ) + 2 + z 1Y ( z ) = 1 z 1 2 Y(z) = + 1 42 41 + (1 + 2 z 1 )( 1 z 1 ) 1 2z 3 1 4 442 4 4 4 3 ZIR ZSR

Output purely due to the initial conditions, Output purely due to the input. input
36
Sharif University of Technology, Department of Computer Engineering, Signals & Systems

ZIR ZSR

Lecture 15 (Chapter 10)

Example (continued)
System is initially at rest: ZSR
Y( z ) = H ( z ) X ( z ) = 1

=0

14 z4 1 23 + 2 1 1 z4 2 3 4 1 1 H ( z) X ( z) 1 H ( z) = H ( z) = 1 + 2z 1

Get response to initial conditions ZIR

=0

Y ( z) =

1 + 2z 1

y[n] = 2 (2) n u[n]

37
Sharif University of Technology, Department of Computer Engineering, Signals & Systems