Shiite Endowment ‫ديوان الوقف الشيعي‬

University of Imam Khadhum for Isalmic

‫كلية اإلمام الكاظم للعلوم اإلسالمية اجلامعة‬
Sciences
‫قسم هندسة الربجميات‬
Department of Software Engineering

CHAPTER TWO
The Z transform

For more information contact with the lecturer on

www.facebook.com/alnawar.alseelawi
CHAPTER TWO Z Transform

1. Introduction

The use of Digital Computer Compensator (controller) devices has

grown during the past three decades as the price and reliability of digital
computers have improved dramatically. A block diagram of a single-loop
digital control system is shown in Figure 2.1.

The digital computer is this system configuration receives the error

in digital form and performs calculations in order to provide an output in
digital form. The computer may be programmed to provide an output so
that the performance of the process is near or equal to desired
performance.

Many computers are able to receive and manipulate several inputs,

so a digital computer control system can often be a multivariable system.

A digital computer receives and operates on signals in digital

(numerical) form, as contrasted to continuous signals.

Output
Reference Digital- (analog)
Digital (digital) (analog)
input to-analog Actuator Process
Computer
converter

(digital) Analog- (analog) Measurement

to-digital
sensor
converter

Figure 2.1

1.1 Types of Signals

A continuous-time signal is a signal defined over a continuous

range of time. The amplitude may assume a continuous range of values or

1
CHAPTER TWO Z Transform

may assume only a finite number of distinct values. The process of

representing a variable by a set of distinct values is called quantization,
and the resulting distinct values are called quantized values. The
quantized variable changes only by a set of distinct steps.

An analog signal is a signal defined over a continuous range of

time whose amplitude can assume a continuous range of values. Figure
2.2(a) shows a continuous-time analog signal, and figure 2.2(b) shows a
continuous-time quantized signal.

Notice that the analog signal is

special case of continuous-time signal.
In practice, however, we frequently
use the treminology "continuous-
time" in lieu of "analog".

A discrete-time signal is a
signal defined only at discrete instants
of time (that is, one in which the
independent variable t is quantized).
In a discrete-time signal, if the
amplitude can assume a continuous
range of values, then the signal is
called a sampled-data signal. A
sampled-data signal can be generated
by sampling as analog signal at
discrete instants of time. It is an
amplitude-modulated pulse signal.
Figure 2.2(c) shows a sampled-data Figure 2.2
signal.

2
CHAPTER TWO Z Transform

A digital signal is a discrete-time signal with quantized amplitude.

Such a signal can be represented by a sequence of numbers, for example,
in the form of binary numbers.

A digital control system uses digital signals and a digital computer

to control a process. The measurement data are converted from analog
form to digital form by means of analog-to-digital converter shown in
Figure 2.1. after processing the inputs, the digital computer provides an
output in digital form. This output is often converted to analog form by
the digital-to-analog converter shown in Figure 2.1.

1.2 Sampled data system

Computers used in control systems are interconnected to the

actuator and the process by the mean of signal converters. The output of
the computer is processed by a digital-to-analog converter. Assume that
all the numbers that enter or leave the computer do so at the same fixed
period 𝑇, called the sampling period. Thus, for example, the reference
input shown in Figure 2.3 is a sequence of sample values 𝑟(𝑘𝑇). The
variables 𝑟(𝑘𝑇), 𝑚(𝑘𝑇), and 𝑢(𝑘𝑇) are discrete signals in contrast to
𝑚(𝑡) and 𝑦(𝑡), which are continuous functions of time.

r(kT) p(t) y(t)

Digital u(kT) Digital- Actuator
Computer to-analog and
converter process

m(kT) Analog- m(t) Measurement

to-digital sensor
converter

Figure 2.3

3
CHAPTER TWO Z Transform

Sampled data (or a discrete signal) are data obtained for the system
variables only at discrete intervals and are denoted as 𝑥(𝑘𝑇).

A system where part of the system acts on sampled data is called a

sampled-data system. A sampler is basically a switch that closes every 𝑇
secondes for one instant of time. Consider an ideal sampler, as shown in
Figure 2.4. The input is 𝑟(𝑡), and the output is 𝑟 ∗ (𝑓), where 𝑛𝑇 is the
current sample time, and the current value of 𝑟 ∗ (𝑡) is 𝑟(𝑛𝑇). We then
have

𝑟 ∗ (𝑡) = 𝑟(𝑛𝑇)𝛿(𝑡 − 𝑛𝑇)

Where 𝛿 is the impulse function.

Sampler

𝑟(𝑡) 𝑟 ∗ (𝑡)

Continuous Sampled
signal signal

Figure 2.4

A digital-to-analog converter (D/A) serves as a device that converts

the sampled signal 𝑟 ∗ (𝑡) to a continuous signal 𝑝(𝑡). the digital-to-analog
converter can usually be represented by a zero-order hold circuit, as
shown in Figure 2.5.

The zero-order hold (Z.O.H) takes the value 𝑟(𝑘𝑇) and holds it
constant for 𝑘𝑇 < 𝑡 < (𝑘 + 1)𝑇, as shown in Figure 2.6 for 𝑘 = 0. Thus,
we use 𝑟(𝑘𝑇) during the sampling period.

4
CHAPTER TWO Z Transform

Zero-order
Sampler hold

𝑟(𝑡) 𝑟 ∗ (𝑡)
𝑮𝟎 (𝒔) 𝑝(𝑡)

Figure 2.5

𝑝(𝑡)

0 𝑇
Time

Figure 2.6

The transfer function of the zero-order hold is:

1
1 − 𝑒 −𝑇𝑆
𝐺0 (𝑠) =
𝑠

The precision of the digital computer and the associated signal

converters is limited.

Precision is the degree of exactness or disceimination with which a

quantity is stated. The precision of the computer is limited by an ability ti
store its output only in digital logic composed of the a finite number of
binary digits. The converted signal 𝑚(𝑘𝑇) (in Figure 2.3) is then said to
include an amplitude quantization error.

5
CHAPTER TWO Z Transform

When the quantization error and the error due to a computer's finite
word size are small relative to the amplitude of the signal, the system is
sufficiently precise, and the precision limitations can be neglected.

2. The z transform

The z transform method is an operational method that is very

powerful when working with discrete-time systems. In what follows we
shall define the z transform of a time function or a number sequence.

In considering the z transform of a time function x(t), we consider

only the sampled values of 𝑥(𝑡), that is, 𝑥(0), 𝑥(𝑇), 𝑥(2𝑇), …, where 𝑇 is
the sampling period. The z transform of a time function x(t), where t is
nonnegative, or of a sequence of values 𝑥(𝑘𝑇), where k takes zero or
positive integers and T is the sampling period, is defined by the following
equation:

∞ 2
𝑋(𝑧) = 𝒵[𝑥(𝑡)] = 𝒵[𝑥(𝑘𝑇)] = ∑ 𝑥(𝑘𝑇)𝑧 −𝑘
𝑘=0

The z transform defined by the equation above is referred to as the

one-sided z transform.

The symbol 𝒵 denotes "the z transform of". In the one-sided z

transform, we assume 𝑥(𝑡) = 0 for 𝑡 < 0 or 𝑥(𝑘) = 0 for 𝑘 < 0. Note
that z is a complex variable.

2.1 z Transforms of Elementary Functions

In the following we shall present z transforms of several

elementary functions. It is noted that in z transform theory, in sampling
discontinuous function 𝑥(𝑡), we assume that the function is continuous

6
CHAPTER TWO Z Transform

from the right, that is, if discontinuity occurs at t=0, then we assume that
𝑥(0) is equal to 𝑥(0+) rather than to the average at the discontinuity,
[𝑥(0−) + (𝑥(0+)]/2.

Unit-Step Function

Let us find the z transform of the unit-step function

1(𝑡), 𝑡≥0
𝑥(𝑡) = {
0, 𝑡<0

As just noted, in sampling a unit-step function we assume that this

function is continuous from the right; that is, 1(0) = 1. Then,

∞ ∞

𝑋(𝑧) = 𝒵[1(𝑡)] = ∑ 1𝑧 −𝑘 = ∑ 𝑧 −𝑘
𝑘=0 𝑘=0

= 1 + 𝑧 −1 + 𝑧 −2 + 𝑧 −3 + ⋯

1
=
1 − 𝑧 −1
𝑧
=
𝑧−1

7
CHAPTER TWO Z Transform

Unit-Ramp Function

Consider the unit ramp function

𝑡, 𝑡≥0
𝑥(𝑡) = {
0, 𝑡<0

Notice that

𝑥(𝑘𝑇) = 𝑘𝑇, 𝑘 = 0, 1, 2, …

Figure 2.7 depicts the sampled unit-ramp signal. The magnitude of

the sampled values are proportional to the sampling period T.

Figure 2.7

8
CHAPTER TWO Z Transform

The z transform of the unit-ramp function can be written as

∞ ∞ ∞

𝑋(𝑧) = 𝒵[𝑡] = ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 = ∑ 𝑘𝑇𝑧 −𝑘 = 𝑇 ∑ 𝑘𝑧 −𝑘

𝑘=0 𝑘=0 𝑘=0

= 𝑇(𝑧 −1 + 2𝑧 −2 + 3𝑧 −3 + ⋯ )

𝑧 −1
=𝑇
(1 − 𝑧 −1 )2

𝑇𝑧
=
(𝑧 − 1)2

Note that it is a function of the sampling period 𝑇.

Polynomial Function 𝒂𝒌

Let us obtain the z transform of the 𝑥(𝑘) as defined by

𝑎𝑘 , 𝑘 = 0, 1, 2, …
𝑥(𝑘) = {
0, 𝑘<0

Where 𝑎 is a constant. Referring to the definintion os the z transform, we

obtain

∞ ∞

𝑋(𝑧) = 𝒵[𝑎𝑘 ] = ∑ 𝑥(𝑘)𝑧 −𝑘 = ∑ 𝑎𝑘 𝑧 −𝑘

𝑘=0 𝑘=0

= 1 + 𝑎𝑧 −1 + 𝑎2 𝑧 −2 + 𝑎3 𝑧 −3 + ⋯

1
=
1 − 𝑎𝑧 −1
𝑧
=
𝑧−𝑎

9
CHAPTER TWO Z Transform

Exponential Function

Let us find the z transform of

𝑒 −𝑎𝑡 𝑡≥0
𝑥(𝑡) = {
0, 𝑡<0

Since

𝑥(𝑘𝑇) = 𝑒 −𝑎𝑘𝑇 , 𝑘 = 0, 1, 2, …

We have

∞ ∞

𝑋(𝑧) = 𝒵[𝑒 −𝑎𝑡 ] = ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 = ∑ 𝑒 −𝑎𝑘𝑇 𝑧 −𝑘

𝑘=0 𝑘=0

= 1 + 𝑒 −𝑎𝑇 𝑧 −1 + 𝑒 −2𝑎𝑇 𝑧 −2 + 𝑒 −3𝑎𝑇 𝑧 −3 + ⋯

1
=
1 − 𝑒 −𝑎𝑇 𝑧 −1
𝑧
=
𝑧 − 𝑒 −𝑎𝑇

Sinusoidal Function

Consider the sinusoidal function

sin 𝜔𝑡 , 𝑡≥0
𝑥(𝑡) = {
0, 𝑡<0

Noting that

𝑒 𝑗𝜔𝑡 = cos 𝜔𝑡 + 𝑗 sin 𝜔𝑡

𝑒 −𝑗𝜔𝑡 = cos 𝜔𝑡 − 𝑗 sin 𝜔𝑡

We have

10
CHAPTER TWO Z Transform

1 𝑗𝜔𝑡
sin 𝜔𝑡 = (𝑒 − 𝑒 −𝑗𝜔𝑡 )
2𝑗

Since the z transform of the exponential function is

1
𝒵[𝑒 −𝑎𝑡 ] =
1 − 𝑒 −𝑎𝑇 𝑧 −1

We have

1
𝑋(𝑧) = 𝒵[sin 𝜔𝑡] = 𝒵 [ (𝑒 𝑗𝜔𝑡 − 𝑒 −𝑗𝜔𝑡 )]
2𝑗

1 1 1
= ( − )
2𝑗 1 − 𝑒 𝑗𝜔𝑇 𝑧 −1 1 − 𝑒 −𝑗𝜔𝑡 𝑧 −1

1 (𝑒 𝑗𝜔𝑇 − 𝑒 −𝑗𝜔𝑇 )𝑧 −1
=
2𝑗 1 − (𝑒 𝑗𝜔𝑇 + 𝑒 −𝑗𝜔𝑇 )𝑧 −1 + 𝑧 −2

𝑧 −1 sin 𝜔𝑇
=
1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

𝑧 sin 𝜔𝑇
=
𝑧 2 − 2𝑧 cos 𝜔𝑇 + 1

Example 1

Obtain the z transform of the cosine function

cos 𝜔𝑡 , 𝑡≥0
𝑥(𝑡) = {
0, 𝑡<0

If we proceed in a manner similar to the way we treated the z

transform of the sine function, we have

1
𝑋(𝑧) = 𝒵[cos 𝜔𝑡] = 2 𝒵[𝑒 𝑗𝜔𝑡 + 𝑒 −𝑗𝜔𝑡 ]

1 1 1
= ( + )
2 1 − 𝑒 𝑗𝜔𝑇 𝑧 −1 1 − 𝑒 −𝑗𝜔𝑇 𝑧 −1

11
CHAPTER TWO Z Transform

1 2 − (𝑒 −𝑗𝜔𝑇 + 𝑒 𝑗𝜔𝑇 )𝑧 −1
=
2 1 − (𝑒 𝑗𝜔𝑇 + 𝑒 −𝑗𝜔𝑇 )𝑧 −1 + 𝑧 −2

1 − 𝑧 −1 cos 𝜔𝑇
=
1 − 2𝑧 −1 cos 𝜔𝑇 + 𝑧 −2

𝑧 2 − 𝑧 cos 𝜔𝑇
= 2
𝑧 − 2𝑧 cos 𝜔𝑇 + 1

Table 2.1

Time signal s-domain z-domain

𝛿(𝑡) 1 1
1 𝑧
𝑢 (𝑡 )
𝑠 𝑧−1
1 𝑇𝑧
𝑡
𝑠2 (𝑧 − 1)2

2 2 𝑇 2 𝑧(𝑧 + 1)
𝑡
𝑠3 (𝑧 − 1)3
1 𝑧
𝑒 −𝑎𝑡
𝑠+𝑎 𝑧 − 𝑒 −𝑎𝑇

−𝑎𝑡
1 𝑇𝑧𝑒 −𝑎𝑇
𝑡𝑒
(𝑠 + 𝑎)2 (𝑧 − 𝑒 −𝑎𝑇 )2
𝑎 𝑧(1 − 𝑒 −𝑎𝑇 )
1 − 𝑒 −𝑎𝑡
𝑠(𝑠 + 𝑎) (𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )
𝜔 𝑧 sin(𝜔𝑇)
sin(𝜔𝑡)
𝑠2 + 𝜔2 𝑧 2 − 2𝑧 cos(𝜔𝑇) + 1
𝑠 𝑧 2 − 𝑧 cos(𝜔𝑇)
cos(𝜔𝑡)
𝑠2 + 𝜔2 𝑧 2 − 2𝑧 cos(𝜔𝑇) + 1

2.2 Poles and Zeros in the z Plane

In engineering applications of the z transform method, 𝑋(𝑧) may

have the form

12
CHAPTER TWO Z Transform

3
𝑏0 𝑧 𝑚 + 𝑏1 𝑧 𝑚−1 + ⋯ + 𝑏𝑚
𝑋(𝑧) = (𝑚 ≤ 𝑛)
𝑧 𝑛 + 𝑎1 𝑧 𝑛−1 + ⋯ + 𝑎𝑛

or

𝑏0 (𝑧 − 𝑧1 )(𝑧 − 𝑧2 ) … (𝑧 − 𝑧𝑚 )
𝑋(𝑧) =
(𝑧 − 𝑝1 )(𝑧 − 𝑝2 ) … (𝑧 − 𝑝𝑛 )

where the 𝑝𝑖 's (𝑖 = 1, 2, … , 𝑛) are the poles of 𝑋(𝑧) and the 𝑧𝑗 's (𝑗 =
1, 2, … , 𝑚) the zeros of 𝑋(𝑧).

The locations of the poles and zeros of 𝑋(𝑧) determine the

characterstics of 𝑥(𝑘), the dequence of values or numbers. As in the case
of the s plane analysis of linear cintinuous-time control systems, we often
use a graphical display in the z plane of the locations of the poles and
zeros of 𝑋(𝑧).

Note that in control engineering and signal processing 𝑋(𝑧) is

frequently expressed as a ratio of polynomials in 𝑧 −1 , as follows:
4
𝑏0 𝑧 −(𝑛−𝑚) + 𝑏1 𝑧 −(𝑛−𝑚+1) + ⋯ + 𝑏𝑚 𝑧 −𝑛
𝑋(𝑧) =
1 + 𝑎1 𝑧 −1 + 𝑎2 𝑧 −2 + ⋯ + 𝑎𝑛 𝑧 −𝑛

where 𝑧 −1 is interpreted as the unit delay operator. In this chapter, where

the basic properties and theorems of the z transform method are
presented, 𝑋(𝑧) may be expressed in terms of power of z, as given by
Equation (3), or in terms of powers of 𝑧 −1 , as given by Equation (4),
depending on the circumstances.

In finding the poles and zeros of 𝑋(𝑧), it is convenient to express

𝑋(𝑧) as a ratio of polynomials in 𝑧, for exmaple

13
CHAPTER TWO Z Transform

𝑧 2 + 0.5𝑧 𝑧(𝑧 + 0.5)

𝑋(𝑧) = 2 =
𝑧 + 3𝑧 + 2 (𝑧 + 1)(𝑧 + 2)

Clearly, 𝑋(𝑧) has poles at 𝑧 = −1 and 𝑧 = −2 and zeros at 𝑧 = 0

and 𝑧 = −0.5. If 𝑋(𝑧) is written as a ratio of polynomials in 𝑧 −1 ,
however, the preceding 𝑋(𝑧) can be written as

1 + 0.5𝑧 −1 1 + 0.5𝑧 −1
𝑋(𝑧) = =
1 + 3𝑧 −1 + 2𝑧 −2 (1 + 𝑧 −1 )(1 + 2𝑧 −1 )

Although poles at 𝑧 = −1 and 𝑧 = −2 and a zero at 𝑧 = −0.5 are

clearly seen from the expression, a zero at 𝑧 = 0 is not explicitly shown,
and so the begineer may fail to see the existence of a zero at 𝑧 = 0.
Therefore, in dealing with the poles and zeros of 𝑋(𝑧), it is preferable to
express 𝑋(𝑧) as a ration of polynomials in 𝑧, rather than polynomials in
𝑧 −1 . In addition, in obtaining the inverse z transform by use of the
inversion integral method, it is desirable to express 𝑋(𝑧) as a ration of
polynomials in 𝑧, rather than polynomials in 𝑧 −1 , to avoid any possible
errors in determining the number of poles at the origin of function
𝑋(𝑧)𝑧 𝑘−1 .

2.3 Z Transform Using Partial Fraction

When the Laplace transform of a function is known, the

corresponding z transform may be obtained by the partial fraction.
Consider the rational function:
5
𝑄(𝑠) ∑𝑚𝑖=0 𝑏𝑖 𝑠
𝑖
𝐹(𝑠) = =
𝑃(𝑠) ∑𝑛𝑖=0 𝑎𝑖 𝑠 𝑖

where 𝑎𝑛 = 1 and 𝑛 ≥ 𝑚.

The polynomial P(s) may be written as:

14
CHAPTER TWO Z Transform

𝑃(𝑠) = 𝑠 𝑛 + 𝑎𝑛−1 𝑠 𝑛−1 + ⋯ + 𝑎1 𝑠 + 𝑎0

where 𝑎0 , 𝑎1 , … , 𝑎𝑛−1 are real coefficients.

By the fundamental theorem of algebra, the denominator

polynomial equation ∑𝑛𝑖=0 𝑎𝑖 𝑠 𝑖 has n poles (roots).

The methods of Partial fraction expansion will now be given for

the cases of Simple Poles, and Multiple-order Poles of 𝐹(𝑠).

Simple Poles: if all poles of are simple and real, Equation (5) can be
written as:
6
𝑄(𝑠) 𝑄(𝑠)
𝐹(𝑠) = =
𝑃(𝑠) (𝑠 + 𝑠1 )(𝑠 + 𝑠2 ) … (𝑠 + 𝑠𝑛 )

where 𝑠1 ≠ 𝑠2 ≠ ⋯ ≠ 𝑠𝑛 . Applying the partial-fraction-expansion

method, Equation (6) is written as:

𝐾𝑠1 𝐾𝑠2 𝐾𝑠𝑛 7

𝐹(𝑠) = + + ⋯+
𝑠 + 𝑠1 𝑠 + 𝑠2 𝑠 + 𝑠𝑛

The coefficient 𝐾𝑠𝑖 (𝑖 = 1, 2, … , 𝑛) is determined by multiplying both

sides of Equation (6) by the factor (𝑠 + 𝑠𝑖 ) and then setting s equal to
−𝑠𝑖 . To find the coefficient 𝐾𝑠1 , for instance, we multiply both sides of
Equation (6) by (𝑠 + 𝑠1 ) and let 𝑠 = −𝑠1 . Thus

𝑄(𝑠) 𝑄(−𝑠1 ) 8
𝐾𝑠1 = [(𝑠 + 𝑠1 ) ] =
𝑃(𝑠) 𝑠=−𝑠 (𝑠2 − 𝑠1 )(𝑠3 − 𝑠1 ) … (𝑠𝑛 − 𝑠1 )
1

Multiple-order poles: if 𝑟 of the 𝑛 poles of F(s) are identical or we say

that the pole at 𝑠 = −𝑠𝑖 is of multiplicity r, F(s) is written as:

𝑄(𝑠) 𝑄(𝑠) 9
𝐹(𝑠) = =
𝑃(𝑠) (𝑠 + 𝑠1 )(𝑠 + 𝑠2 ) … (𝑠 + 𝑠𝑛−𝑟 )(𝑠 + 𝑠𝑖 )𝑟

15
CHAPTER TWO Z Transform

Then F(s) can be expanded as:

𝐾𝑠(𝑛−𝑟)
10
𝐾𝑠1 𝐾𝑠2 𝐴1 𝐴2 𝐴𝑟
𝐹(𝑠) = + + ⋯+ + + 2
+ ⋯+
𝑠 + 𝑠1 𝑠 + 𝑠2 𝑠 + 𝑠(𝑛−𝑟) 𝑠 + 𝑠𝑖 (𝑠 + 𝑠2 ) (𝑠 + 𝑠𝑖 )𝑟

Then (n-r) coefficients 𝐾𝑠1 , 𝐾𝑠2 and 𝐾𝑠(𝑛−𝑟) which correspond to simple
poles, may be evaluated by the method described by Equation (8). The
determination of the coefficients that correspond to the multiple-order
poles is described as follows:

𝐴𝑟 = [(𝑠 + 𝑠𝑖 )𝑟 𝐹(𝑠)]𝑠=−𝑠𝑖

𝑑
𝐴𝑟−1 = [(𝑠 + 𝑠𝑖 )𝑟 𝐹(𝑠)]𝑠=−𝑠𝑖
𝑑𝑠

1 𝑑2
𝐴𝑟−1 = [(𝑠 + 𝑠𝑖 )𝑟 𝐹(𝑠)]𝑠=−𝑠𝑖
2! 𝑑𝑠 2

1 𝑑 𝑟−1
𝐴1 = [(𝑠 + 𝑠𝑖 )𝑟 𝐹(𝑠)]𝑠=−𝑠𝑖
(𝑟 − 1)! 𝑑𝑠 𝑟−1

Example 2

Determine the z transform for the function whose Laplace

transform is

1
𝐹(𝑠) =
𝑠(𝑠 + 1)

By using partial-fraction expansion method, we have

𝐾1 𝐾2
𝐹(𝑠) = +
𝑠 𝑠+1

𝐾1 = [𝑠𝐹(𝑠)]𝑠=0 = 1

𝐾2 = [(𝑠 + 1)𝐹(𝑠)]𝑠=−1 = −1

16
CHAPTER TWO Z Transform

1 1
𝐹(𝑠) = −
𝑠 𝑠+1

From Table 2.1,

1 𝑧
𝒵[ ] =
𝑠 𝑧−1

1 𝑧
𝒵[ ]=
𝑠+1 𝑧 − 𝑒 −𝑇

Thus,

𝑧 𝑧 𝑧(1 − 𝑒 −𝑇 )
𝐹(𝑧) = − =
𝑧 − 1 𝑧 − 𝑒 −𝑇 (𝑧 − 1)(𝑧 − 𝑒 −𝑇 )

Example 3

Find the z transform of the function

5𝑠 + 3
𝐹(𝑠) =
(𝑠 + 1)(𝑠 + 2)(𝑠 + 3)

The F(s) is written in the partial-fraction-expansion form

𝐾1 𝐾2 𝐾3
𝐹(𝑠) = + +
(𝑠 + 1) (𝑠 + 2) (𝑠 + 3)

5(−1) + 3
𝐾1 = [(𝑠 + 1)𝐹(𝑠)]𝑠=−1 = = −1
(−1 + 2)(−1 + 3)

5(−2) + 3
𝐾2 = [(𝑠 + 2)𝐹(𝑠)]𝑠=−2 = =7
(−2 + 1)(−2 + 3)

5(−3) + 3
𝐾3 = [(𝑠 + 3)𝐹(𝑠)]𝑠=−3 = = −6
(−3 + 1)(−3 + 2)

Then,

17
CHAPTER TWO Z Transform

−1 7 6
𝐹(𝑠) = + −
(𝑠 + 1) (𝑠 + 2) (𝑠 + 3)

𝑧 7𝑧 6𝑧
𝐹(𝑧) = − + −
𝑧 − 𝑒 −𝑇 𝑧 − 𝑒 −2𝑇 𝑧 − 𝑒 −3𝑇

Example 4

Find the z transorm of the function

𝑠 2 + 2𝑠 + 2
𝐹(𝑠) = 2
𝑠 + 3𝑠 + 2

By dividing the numinator of F(s) over denominator, we get

𝑠 𝑠
𝐹(𝑠) = 1 − = 1 −
𝑠 2 + 3𝑠 + 2 (𝑠 + 1)(𝑠 + 2)

By using partial-fraction-expansion method, we have

𝐾1 𝐾2
𝐹(𝑠) = 1 + +
(𝑠 + 1) (𝑠 + 2)

𝐾1 = [(𝑠 + 1)𝐹(𝑠)]𝑠=−1 = 1

𝐾2 = [(𝑠 + 2)𝐹(𝑠)]𝑠=−2 = −2

1 2
𝐹(𝑠) = 1 + −
(𝑠 + 1) (𝑠 + 2)

𝑧 2𝑧
𝐹(𝑧) = 1 + +
𝑧 − 𝑒 −𝑇 𝑧 − 𝑒 −2𝑇

Example 5

Find the z transform of

1
𝐹(𝑠) =
𝑠(𝑠 + 1)2 (𝑠 + 2)

18
CHAPTER TWO Z Transform

Using the format of Equation (2.6) , we have

𝐾1 𝐾2 𝐴1 𝐴2
𝐹(𝑠) = + + +
𝑠 𝑠 + 2 𝑠 + 1 (𝑠 + 1)2

The coefficients correspond to the poles are

𝐾1 = [𝑠𝐹(𝑠)]𝑠=0 = 0.5

𝐾2 = [(𝑠 + 2)𝐹(𝑠)]𝑠=−2 = −0.5

𝑑
𝐴1 = [(𝑠 + 1)2 𝐹(𝑠)]𝑠=−1 = 0
𝑑𝑠

𝐴2 = [(𝑠 + 1)2 𝐹(𝑠)]𝑠=−1 = −1

0.5 0.5 1
𝐹(𝑠) = − −
𝑠 𝑠 + 2 (𝑠 + 1)2

𝑧 𝑧 𝑇𝑧𝑒 −𝑇
𝐹(𝑧) = − −
2(𝑧 − 1) 2(𝑧 − 𝑒 −2𝑇 ) (𝑧 + 𝑒 −𝑇 )2

2.4 z Transform Using Residue Method

This is a powerful technique for obtaining z transforms. The z

transform of 𝑓 ∗ (𝑡) may be expressed in the form

For unrepeated poles:

𝑧 11
𝐹(𝑧) = 𝒵[𝑓 ∗ (𝑡)] = ∑ residues of 𝐹(𝑠) at poles of 𝐹(𝑠)
𝑧 − 𝑒 𝑠𝑇

For repeated poles:

1 𝑑 𝑛−1 𝑧 12
𝐹(𝑧) = ∑ residues of 𝐹(𝑠) at poles of 𝐹(𝑠)
(𝑛 − 1)! 𝑑𝑠 𝑛−1 𝑧 − 𝑒 𝑠𝑇

19
CHAPTER TWO Z Transform

When the denominator of 𝐹(𝑠) contains a linear factor of form

𝑠 − 𝑟 that 𝐹(𝑠) has a first-order pole at 𝑠 = 𝑟, the corresponding residue
𝑅 is:

𝑧 13
𝑅 = lim(𝑠 − 𝑟) [𝐹(𝑠) ]
𝑠→𝑟 𝑧 − 𝑒 𝑠𝑇

Example 6

Determine the z transform of a unit step function.

1
For 𝐹(𝑠) = , there is but one pole at 𝑠 = 0. The corresponding
𝑠

residue is:

1 𝑧 𝑧
𝑅 = lim 𝑠 [ ] =
𝑠→0 𝑠 𝑧 − 𝑒 𝑠𝑇 𝑧−1

Example 7

Determine the z transform of 𝑒 −𝑎𝑡 using residue method.

1
For this function, 𝐹(𝑠) = , which has but one pole at 𝑠 = −𝑎.
𝑠+𝑎

Thus,

1 𝑧 𝑧
𝑅 = lim (𝑠 + 𝑎) [ ] =
𝑠→−𝑎 (𝑠 + 𝑎) 𝑧 − 𝑒 𝑠𝑇 𝑧 − 𝑒 −𝑎𝑇

Example 8

Determine the z transform for the function whose Laplace

transform is:

1
𝐹(𝑠) =
𝑠(𝑠 + 1)

20
CHAPTER TWO Z Transform

The poles of 𝐹(𝑠) occur at 𝑠 = 0 and 𝑠 = −1. The residue due to

the pole 𝑠 = 0 is:

1 𝑧 𝑧
𝑅1 = lim 𝑠 [ ] =
𝑠→0 𝑠(𝑠 + 1) 𝑧 − 𝑒 𝑠𝑇 𝑧−1

The residue due to the pole at 𝑠 = −1 is:

1 𝑧 𝑧
𝑅2 = lim (𝑠 + 1) [ ] = −
𝑠→−1 𝑠(𝑠 + 𝑎) 𝑧 − 𝑒 𝑠𝑇 𝑧 − 𝑒 −𝑇

Adding these two residues results in

𝑧 𝑧 𝑧(1 − 𝑒 −𝑇 )
𝑅 = 𝑅1 + 𝑅2 = − =
𝑧 − 1 𝑧 − 𝑒 −𝑇 (𝑧 − 1)(𝑧 − 𝑒 −𝑇 )

2.5 Properties and theorems of the z transform

The use of the z transform ,ethod in the analysis of discrete-time

control systems may be facilitated if theorems of the z transform are
referred to. In this section we present important properties and useful
theorems of the z transform. We assume that the time fuction 𝑥(𝑡) is z-
transformable and that 𝑥(𝑡) is zero for 𝑡 < 0.

Multiplication by a Constant

If 𝑋(𝑧) is the z transform of 𝑥(𝑡), then

𝒵[𝑎𝑥(𝑡)] = 𝑎𝒵[𝑥(𝑡)] = 𝑎𝑋(𝑧)

Where 𝑎 is a constant.

To prove this, note that be difinition

∞ ∞

𝒵[𝑎𝑥(𝑡)] = ∑ 𝑎𝑥(𝑘𝑇)𝑧 −𝑘 = 𝑎 ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 = 𝑎𝑋(𝑧)

𝑘=0 𝑘=0

21
CHAPTER TWO Z Transform

Linearity of the z transform

The z transform possesses an important property: linearity. This

means that, if 𝑓(𝑘) and 𝑔(𝑘) are z-transformable and α and β are scalars,
then 𝑥(𝑘) formed by linear combination

𝑥(𝑘) = 𝛼𝑓(𝑘) + 𝛽𝑔(𝑘)

Has the z transform

𝑋(𝑧) = 𝛼𝐹(𝑧) + 𝛽𝐺(𝑧)

Where 𝐹(𝑧) and 𝐺(𝑧) are the z transforms of 𝑓(𝑘) and 𝑔(𝑘),
respectively.

The linearity property can be proved by

𝑋(𝑧) = 𝒵[𝑥(𝑘)] = 𝒵[𝛼𝑓(𝑘) + 𝛽𝑔(𝑘)]

= ∑ [𝛼𝑓(𝑘) + 𝛽𝑔(𝑘)]𝑧 −𝑘
𝑘=0

∞ ∞

= 𝛼 ∑ 𝑓(𝑘)𝑧 −𝑘 + 𝛽 ∑ 𝑔(𝑘)𝑧 −𝑘
𝑘=0 𝑘=0

= 𝛼𝒵[𝑓(𝑘)] + 𝛽𝒵[𝑔(𝑘)]

= 𝛼𝐹(𝑧) + 𝛽𝐺(𝑧)

Multiplication by 𝒂𝒌

If 𝑋(𝑧) is the z transform of 𝑥(𝑡), then the z transform of 𝑎𝑘 𝑥(𝑘)

can be given by

𝒵[𝑎𝑘 𝑥(𝑘)] = 𝑋(𝑎−1 𝑧)

This can be proved as follows:

22
 13
CHAPTER TWO Z Transform

𝒵[𝑎𝑘 𝑥(𝑘)] = ∑∞ 𝑘
𝑘=0 𝑎 𝑥(𝑘)𝑧
−𝑘
= ∑∞ −1 −𝑘
𝑘=0 𝑥(𝑘)(𝑎 𝑧)

= 𝑋(𝑎−1 𝑧)

Shifting Theorem

The shifting theorem presented here is also referred to as the real

translation theorem. If 𝑥(𝑡) = 0 for 𝑡 < 0 and 𝑥(𝑡) has the z transform
𝑋(𝑧), then
14
−𝑛
𝒵[𝑥(𝑡 − 𝑛𝑇)] = 𝑧 𝑋(𝑧)

𝑛−1 15
𝒵[𝑥(𝑡 + 𝑛𝑇)] = 𝑧 𝑛 [𝑋(𝑧) − ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 ]
𝑘=0

Where 𝑛 is zero or a positive integer.

To prove the Equation (14), note that

𝒵[𝑥(𝑡 − 𝑛𝑇)] = ∑ 𝑥(𝑘𝑇 − 𝑛𝑇)𝑧 −𝑘

𝑘=0

= 𝑧 −𝑛 ∑ 𝑥(𝑘𝑇 − 𝑛𝑇)𝑧 −(𝑘−𝑛)

𝑘=0

By defining 𝑚 = 𝑘 − 𝑛, this Equation (14) can be written as follows:

𝒵[𝑥(𝑡 − 𝑛𝑇)] = 𝑧 −𝑛 ∑ 𝑥(𝑚𝑇)𝑧 −𝑚

𝑚=−𝑛

Since 𝑥(𝑚𝑇) = 0 form 𝑚 < 0, we may change the lower limit of

the summation from 𝑚 = −𝑛 to 𝑚 = 0. Hence,

𝒵[𝑥(𝑡 − 𝑛𝑇)] = 𝑧 −𝑛 ∑ 𝑥(𝑚𝑇)𝑧 −𝑚 = 𝑧 −𝑛 𝑋(𝑧)

𝑚=0

23
CHAPTER TWO Z Transform

Thus, multiplication of a z transform by 𝑧 −𝑛 has the effect of

delaying the time function 𝑥(𝑡) by time 𝑛𝑇. (That is, move the function
to the right by time 𝑛𝑇).

To prove the Equation (15), we note that

𝒵[𝑥(𝑡 + 𝑛𝑇)] = ∑ 𝑥(𝑘𝑇 + 𝑛𝑇)𝑧 −𝑘

𝑘=0

= 𝑧 𝑛 ∑ 𝑥(𝑘𝑇 + 𝑛𝑇)𝑧 −(𝑘+𝑛)

𝑘=0

∞ 𝑛−1 𝑛−1

= 𝑧 𝑛 [∑ 𝑥(𝑘𝑇 + 𝑛𝑇)𝑧 −(𝑘+𝑛) + ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 − ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 ]

𝑘=0 𝑘=0 𝑘=0

∞ 𝑛−1

= 𝑧 𝑛 [∑ 𝑥(𝑘𝑇)𝑧 −𝑘 − ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 ]
𝑘=0 𝑘=0

𝑛−1

= 𝑧 𝑛 [𝑋(𝑧) − ∑ 𝑥(𝑘𝑇)𝑧 −𝑘 ]
𝑘=0

For the number sequence 𝑥(𝑘), the equation can be written as follows:

𝑛−1

𝒵[𝑥(𝑘 + 𝑛)] = 𝑧 𝑛 [𝑋(𝑧) − ∑ 𝑥(𝑘)𝑧 −𝑘 ]

𝑘=0

From this last equation, we obtain

𝒵[𝑥(𝑘 + 1)] = 𝑧𝑋(𝑧) − 𝑧𝑥(0)

𝒵[𝑥(𝑘 + 2)] = 𝑧𝒵[𝑥(𝑘 + 1)] − 𝑧𝑥(1)

= 𝑧 2 𝑋(𝑧) − 𝑧 2 𝑥(0) − 𝑧𝑥(1)

Similarly,

24
CHAPTER TWO Z Transform

𝒵 [𝑥 (𝑘 + 𝑛)] = 𝑧 𝑛 𝑋(𝑧) − 𝑧 𝑛 𝑥 (0) − 𝑧 𝑛−1 𝑥 (1) − 𝑧 𝑛−2 𝑥 (2) − ⋯ − 𝑧𝑥 (𝑛 − 1)

Where 𝑛 is a positive integer.

Remember that multiplication of the z transform 𝑋(𝑧) by 𝑧 has the

effect of advancing the signal 𝑥(𝑘𝑇) by one step (1 sampling period) and
that multiplication of the z transform 𝑋(𝑧) by 𝑧 −1 has the effect of
delaying the signal 𝑥(𝑘𝑇) by one step (1 sampling period).

Example 9

Fine the z transform of unit-step function that are delayed by 1

sampling period and 4 sampling periods, respectively, as shown in Figure
2.8 (a) and (b).

Figure 2.8
Using the shifting theorem, we have

−1 −1
1 𝑧 −1 1
𝒵[1(𝑡 − 𝑇)] = 𝑧 𝒵[1(𝑡)] = 𝑧 = =
1 − 𝑧 −1 1 − 𝑧 −1 1 − 𝑧

25
CHAPTER TWO Z Transform

Also,

−4 −4
1 𝑧 −4
𝒵[1(𝑡 − 4𝑇)] = 𝑧 𝒵[1(𝑡)] = 𝑧 =
1 − 𝑧 −1 1 − 𝑧 −1

1 1
= =
𝑧 4 − 𝑧 3 𝑧 3 (𝑧 − 1)

(Note that 𝑧 −1 represents a delay of 1 sampling period 𝑇, regardless of

the value of 𝑇).

Example 10

Obtain the z transform of

𝑎𝑘−1 , 𝑘 = 1, 2, 3, …
𝑓(𝑎) = {
0, 𝑘≤0

Referring to Equation (14), we have

𝒵[𝑥(𝑘 − 1)] = 𝑧 −1 𝑋(𝑧)

The z transform of 𝑎𝑘 is

𝑧
𝒵[𝑎𝑘 ] =
𝑧−𝑎

And so

𝑧 1
𝒵[𝑓(𝑎)] = 𝒵[𝑎𝑘−1 ] = 𝑧 −1 =
𝑧−𝑎 𝑧−𝑎

Where 𝑘 = 1, 2, 3, … .

Example 11

Consider the function 𝑦(𝑘), which is a sum of functions 𝑥(ℎ),

where ℎ = 0, 1, 2, … . , 𝑘, such that

26
CHAPTER TWO Z Transform

𝑦(𝑘) = ∑ 𝑥(ℎ) , 𝑘 = 0, 1, 2, …
ℎ=0

Where 𝑦(𝑘) = 0 for 𝑘 < 0. Obtain the transform of 𝑦(𝑘).

First note that

𝑦(𝑘) = 𝑥(0) + 𝑥(1) + ⋯ + 𝑥(𝑘 − 1) + 𝑥(𝑘)

𝑦(𝑘 − 1) = 𝑥(0) + 𝑥(1) + ⋯ + 𝑥(𝑘 − 1)

Hence,

𝑦(𝑘) − 𝑦(𝑘 − 1) = 𝑥(𝑘), 𝑘 = 0, 1, 2, …

Therefore,

𝒵[𝑦(𝑘) − 𝑦(𝑘 − 1)] = 𝒵[𝑥(𝑘)]

Or

𝑌(𝑧) − 𝑧 −1 𝑌(𝑧) = 𝑋(𝑧)

Which yields

1 𝑧
𝑌(𝑧) = 𝑋(𝑧) = 𝑋(𝑧)
1 − 𝑧 −1 𝑧−1

Where 𝑋(𝑧) = 𝒵[𝑥(𝑘)].

Complex Translation Theorem

If 𝑥(𝑡) has the z transform 𝑋(𝑧), then the z transform of 𝑒 −𝑎𝑡 𝑥(𝑡)
can be given by 𝑋(𝑧𝑒 𝑎𝑇 ). This is known as the complex translation
theorem.

To prove this theorem, note that

27
CHAPTER TWO Z Transform

∞ ∞

𝒵[𝑒 −𝑎𝑡 𝑥(𝑡)] = ∑ 𝑥(𝑘𝑇)𝑒 −𝑎𝑘𝑇 𝑧 −𝑘 = ∑ 𝑥(𝑘𝑇)(𝑧𝑒 𝑎𝑇 )−𝑘 = 𝑋(𝑧𝑒 𝑎𝑇 )

𝑘=0 𝑘=0

Thus, we see that replacing z in 𝑋(𝑧) by 𝑧𝑒 𝑎𝑇 gives the z transform of

𝑒 −𝑎𝑡 𝑥(𝑡).

Example 12

Given the z transforms of sin 𝜔𝑡 and cos 𝜔𝑡, obtain the z

transforms of 𝑒 −𝑎𝑡 sin 𝜔𝑡 and 𝑒 −𝑎𝑡 cos 𝜔𝑡, respectively, by using
complex translation theorem.

𝑧 sin(𝜔𝑇)
𝒵[sin 𝜔𝑡] =
𝑧 2 − 2𝑧 cos(𝜔𝑇) + 1

We substitute 𝑧𝑒 𝑎𝑇 for z to obtain the z transform of 𝑒 −𝑎𝑡 sin 𝜔𝑡, as

follows:

−𝑎𝑡
𝑒 𝑎𝑇 𝑧 sin(𝜔𝑇)
𝒵[𝑒 sin 𝜔𝑡] = 2𝑎𝑇 2
𝑒 𝑧 − 2𝑒 𝑎𝑇 𝑧 cos(𝜔𝑇) + 1

Example 13

Obtain the z transform of 𝑡𝑒 −𝑎𝑡

𝑇𝑧 −1
𝒵[𝑡] = = 𝑋(𝑧)
(1 − 𝑧 −1 )2

Thus,

−𝑎𝑡 ] 𝑎𝑇
𝑇𝑒 −𝑎𝑇 𝑧 −1
𝒵[𝑡𝑒 = 𝑋(𝑧𝑒 )=
(1 − 𝑒 −𝑎𝑇 𝑧 −1 )2

28
CHAPTER TWO Z Transform

Initial Value Theorem

If 𝑥(𝑡) has the z transform 𝑋(𝑧) and if lim𝑧→∞ 𝑋(𝑧) exists, then
the initial value 𝑥(0) of 𝑥(𝑡) or 𝑥(𝑘) is given by

𝑥(0) = lim 𝑋(𝑧)

𝑧→∞

To prove this theorem, note that

𝑋(𝑧) = ∑ 𝑥(𝑘)𝑧 −𝑘 = 𝑥(0) + 𝑥(1)𝑧 −1 + 𝑥(2)𝑧 −2 + ⋯

𝑘=0

Letting 𝑧 → ∞ in this equation, we obtain Equation (2.15). The

behavior of the signal in the neighborhood of 𝑡 = 0 or 𝑘 = 0 can thus
determined by the behavior of 𝑋(𝑧) at 𝑧 = ∞.

The initial value theorem is convenient for checking z transform

calculations for possible errors. Since 𝑥(0) is usually known, a check of
the initial value by lim𝑧→∞ 𝑋(𝑧) can easily spot errors in 𝑋(𝑧), if any
exist.

Example 14

Determine the initial value 𝑥(0) if the z transform of 𝑥(𝑡) is given

by

(1 − 𝑒 −𝑇 )𝑧 −1
𝑋(𝑧) =
(1 − 𝑧 −1 )(1 − 𝑒 −𝑇 𝑧 −1 )

By using initial value theorem, we find

(1 − 𝑒 −𝑇 )𝑧 −1
𝑥(0) = lim =0
𝑧→∞ (1 − 𝑧 −1 )(1 − 𝑒 −𝑇 𝑧 −1 )

29
CHAPTER TWO Z Transform

Referring to Example 2, notice that this 𝑋(𝑧) was the z transform

of

𝑥(𝑡) = 1 − 𝑒 −𝑡

And thus x(0)=0, which agrees with the result obtained earlier.

Final Value Theorem

Suppose that 𝑥(𝑘), where 𝑥(𝑘) = 0 for 𝑘 < 0, has the z transform
𝑋(𝑧) and that all the poles of 𝑋(𝑧) lie inside the unit circle, with the
possible exception of a simple pole at 𝑧 = 1. [This is the condition for the
stability of 𝑋(𝑧), or the condition for 𝑥(𝑘)(𝑘 = 0, 1, 2, … ) to remain
finite]. Then the final value of 𝑥(𝑘), that is, the value of 𝑥(𝑘) as 𝑘
approaches infinity, can be given by

lim 𝑥(𝑘) = lim[(1 − 𝑧 −1 )𝑋(𝑧)]

𝑘→∞ 𝑧→1

To prove that final value theorem, note that

𝒵[𝑥(𝑘)] = 𝑋(𝑧) = ∑ 𝑥(𝑘)𝑧 −𝑘

𝑘=0

∞
−1
𝒵[𝑥(𝑘 − 1)] = 𝑧 𝑋(𝑧) = ∑ 𝑥(𝑘 − 1)𝑧 −𝑘
𝑘=0

Hence,

∞ ∞

∑ 𝑥(𝑘)𝑧 −𝑘 − ∑ 𝑥(𝑘 − 1)𝑧 −𝑘 = 𝑋(𝑧) − 𝑧 −1 𝑋(𝑧)

𝑘=0 𝑘=0

Thaking the limit as 𝑧 approaches unity, we have

30
CHAPTER TWO Z Transform

∞ ∞

lim [∑ 𝑥(𝑘)𝑧 −𝑘 − ∑ 𝑥(𝑘 − 1)𝑧 −𝑘 ] = lim[(1 − 𝑧 −1 )𝑋(𝑧)]

𝑧→1 𝑧→1
𝑘=0 𝑘=0

Because of the assumed stability condition and the condition that

𝑥(𝑘) = 0 for 𝑘 < 0, the left-hand side of this last equation becomes

∑ [𝑥(𝑘) − 𝑥(𝑘 − 1)] = [𝑥(0) − 𝑥(−1)] + [𝑥(1) − 𝑥(0)]

𝑘=0

+[𝑥(2) − 𝑥(1)] + ⋯ = 𝑥(∞) = lim 𝑥(𝑘)

𝑘→∞

Hence,

lim 𝑥(𝑘) = lim[(1 − 𝑧 −1 )𝑋(𝑧)]

𝑘→∞ 𝑧→1

The final value theorem is very useful in determining the behavior

of 𝑥(𝑘) as 𝑘 → ∞ from its z transform 𝑋(𝑧).

Example 15

Determine the final value 𝑥(∞) of

𝑧 𝑧
𝑋(𝑧) = − 𝑎>0
𝑧 − 1 𝑧 − 𝑒 −𝑎𝑇

by using the final value theorem.

By applying the final value theorem to the given 𝑋(𝑧), we obtain

𝑥(∞) = lim[(1 − 𝑧 −1 )𝑋(𝑧)]

𝑧→1

𝑧 𝑧
= lim [(1 − 𝑧 −1 ) ( − )]
𝑧→1 𝑧 − 1 𝑧 − 𝑒 −𝑎𝑇

1 − 𝑧 −1
= lim (1 − )=1
𝑧→1 𝑧 − 𝑒 −𝑎𝑇

31
CHAPTER TWO Z Transform

Table 2.2
𝑥(𝑡) or 𝑥(𝑘) 𝒵[𝑥(𝑡)] or 𝒵[𝑥(𝑘)]
1 𝑎𝑥(𝑡) 𝑎𝑋(𝑧)
2 𝑎𝑥1 (𝑡) + 𝑏𝑥2 (𝑡) 𝑎𝑋1 (𝑧) + 𝑏𝑋2 (𝑧)
3 𝑥(𝑡 + 𝑇)or 𝑥(𝑘 + 1) 𝑧𝑋(𝑧) − 𝑧𝑥(0)
4 𝑥(𝑡 + 2𝑇) 𝑧 2 𝑋(𝑧) − 𝑧 2 𝑥(0) − 𝑧𝑥(𝑇)
5 𝑥(𝑘 + 2) 𝑧 2 𝑋(𝑧) − 𝑧 2 𝑥(0) − 𝑧𝑥(1)
6 𝑥(𝑡 + 𝑘𝑇) 𝑧 𝑘 𝑋(𝑧) − 𝑧 𝑘 𝑥(0) − 𝑧 𝑘−1 𝑥(𝑇) − ⋯ − 𝑧𝑥(𝑘𝑇 − 𝑇)
7 𝑥(𝑡 − 𝑘𝑇) 𝑧 −𝑘 𝑋(𝑧)
8 𝑥(𝑛 + 𝑘) 𝑧 𝑘 𝑋(𝑧) − 𝑧 𝑘 𝑥(0) − 𝑧 𝑘−1 𝑥(1) − ⋯ − 𝑧𝑥(𝑘 − 1)
9 𝑥(𝑛 − 𝑘) 𝑧 −𝑘 𝑋(𝑧)
𝑑
10 𝑡𝑥(𝑡) −𝑇𝑧 𝑋(𝑧)
𝑑𝑧
𝑑
11 𝑘𝑥(𝑘) −𝑧 𝑋(𝑧)
𝑑𝑧
12 𝑒 −𝑎𝑡 𝑥(𝑡) 𝑋(𝑧𝑒 𝑎𝑇 )
13 𝑒 −𝑎𝑘 𝑥(𝑘) 𝑋(𝑧𝑒 𝑎 )
𝑧
14 𝑎𝑘 𝑥(𝑘) 𝑋( )
𝑎
𝑑 𝑧
15 𝑘𝑎𝑘 𝑥(𝑘) − 𝑋( )
𝑑𝑧 𝑎
16 𝑥(0) lim𝑧→∞ 𝑋(𝑧) if the limit exists

17 𝑥(∞) lim[(1 − 𝑧 −1 )𝑋(𝑧)]

𝑧→1
𝑛
1
18 ∑ 𝑥(𝑘) 𝑋(𝑧)
1 − 𝑧 −1
𝑘=0
𝑑 𝑚
19 𝑚
𝑘 𝑥(𝑘) (−𝑧 ) 𝑋(𝑧)
𝑑𝑧
𝑛
20 ∑ 𝑥(𝑘𝑇)𝑦(𝑛𝑇 − 𝑘𝑇) 𝑋(𝑧)𝑌(𝑧)
𝑘=0
𝑛
21 ∑ 𝑥(𝑘) 𝑋(1)
𝑘=0

32
CHAPTER TWO Z Transform

3. The Inverse z Transform

The z transformation serves the same role for discrete-time control

systems that the Laplace transformation serves for cointinuous-time
control systems. For the z transform to be usefulm we must be familiar
with methods for finding the inverse z transform.

The notation for the inverse z transform is 𝒵 −1 . The inverse z

transform of 𝑋(𝑧) yields the corresponding time sequence 𝑥(𝑘).

It should be noted that only the time sequence at the sampling

instants is obtained from the inverse z transform. Thus, the inverse z
transform of 𝑋(𝑧) yields a unique 𝑥(𝑘), but does not yiels a unique 𝑥(𝑡).

This means that the inverse z transform yields a time sequence that
specifies the value of 𝑥(𝑡) only at discrete instants of time, 𝑡 =
0, 𝑇, 2𝑇, …, and says nothing abut the values of 𝑥(𝑡) at al other times.
That is, many different time functions 𝑥(𝑡) can have the same 𝑥(𝑘𝑇), see
Figure 2.9.

Figure 2.9
When 𝑋(𝑧), the z transform of 𝑥(𝑘𝑇) or 𝑥(𝑘), is given, the
operation that determines the corresponding 𝑥(𝑘𝑇) or 𝑥(𝑘) is called the
inverse z transformation.

33
CHAPTER TWO Z Transform

An obvious method for finding the inverse z transform is to refer to

a z transform table. However, unless we refer to an extensive z transform
table, we may not be able to find the inverse z transform of a complicated
function of 𝑧. (If we use a less extensive table of z transforms, it is
necessary to express a complex z transform as a sum of simpler z
transforms. Refer to the partial-fraction-expansion method presented in
this section).

Other than referring to z transform tables, three methods for

obtaining the inverse z transform are commonly available:

1. Direct division method

2. Partial-fraction-expansion method
3. Inversion integral method

In obtaining the inverse z transform, we assume, as usual, that the

time sequence 𝑥(𝑘𝑇) or 𝑥(𝑘) is zero for 𝑘 < 0.

3.1 Direct Division Method

In the direct division method we obtain the inverse z transform by

expanding 𝑋(𝑧) into an infinite power series in 𝑧 −1 . This method is
useful when it is difficult to obtain the closed-form expression for the
inverse z transform or it is desired to find only the first several terms of
𝑥(𝑘).

The direct division method stems from the fact that if 𝑋(𝑧) is
expanded into a power series in 𝑧 −1 , that is, if

𝑋(𝑧) = ∑ 𝑥(𝑘𝑇)𝑧 −𝑘
𝑘=0

= 𝑥(0) + 𝑥(𝑇)𝑧 −1 + 𝑥(2𝑇)𝑧 −2 + ⋯ + 𝑥(𝑘𝑇)𝑧 −𝑘 + ⋯

34
CHAPTER TWO Z Transform

or

𝑋(𝑧) = ∑ 𝑥(𝑘𝑇)𝑧 −𝑘
𝑘=0

= 𝑥(0) + 𝑥(1)𝑧 −1 + 𝑥(2)𝑧 −2 + ⋯ + 𝑥(𝑘)𝑧 −𝑘 + ⋯

then 𝑥(𝑘𝑇) or 𝑥(𝑘) is the coefficient of the 𝑧 −𝑘 term. Hence, the values
of 𝑥(𝑘𝑇) or 𝑥(𝑘) for 𝑘 = 0, 1, 2, … can be determined by inspection.

If 𝑋(𝑧) is given in the form of a rational function, the expansion

into an infinite power series in increasing power of 𝑧 −1 can be
accomplished by simply dividing the numerator by the denominator,
where both the numerator and denominator of 𝑋(𝑧) are written in
increasing power of 𝑧 −1 . If the resulting series is convergent, the
coefficients of the 𝑧 −𝑘 term in the series are the values 𝑥(𝑘𝑇) of the time
sequence or the values of 𝑥(𝑘) of the number sequence.

Although the present method gives the values of

𝑥(0), 𝑥(𝑇), 𝑥(2𝑇), … or the values of 𝑥(0), 𝑥(1), 𝑥(2), … in a sequential
manner, it is usually difficult to obtain an expression for general term
from a set of values of 𝑥(𝑘𝑇) or 𝑥(𝑘).

Example 16

Find 𝑥(𝑘) for 𝑘 = 0, 1, 2, 3, 4 when 𝑋(𝑧) is given by

10𝑧 + 5
𝑋(𝑧) =
(𝑧 − 1)(𝑧 − 0.2)

First, rewrite 𝑋(𝑧) as a ratio of polynomials in 𝑧 −1 , as follows:

10𝑧 −1 + 5𝑧 −2
𝑋(𝑧) =
1 − 1.2𝑧 −1 + 0.2𝑧 −2

35
CHAPTER TWO Z Transform

Dividing the numerator by the denominator, we have

10𝑧 −1 + 17𝑧 −2 + 18.4𝑧 −3 + 18.68𝑧 −4 + ⋯

1 − 1.2𝑧 −1 + 0.2𝑧 −2 10𝑧 −1 + 5𝑧 −2

10𝑧 −1 − 12𝑧 −2 + 2𝑧 −3

17𝑧 −2 − 2𝑧 −3

17𝑧 −2 − 20.4𝑧 −3 + 3.4𝑧 −4

18.4𝑧 −3 − 3.4𝑧 −4

18.4𝑧 −3 − 22.08𝑧 −4 + 3.68𝑧 −5

18.68𝑧 −4 − 3.68𝑧 −5

Thus,

𝑋(𝑧) = 10𝑧 −1 + 17𝑧 −2 + 18.4𝑧 −3 + 18.68𝑧 −4 + ⋯

By comparing this infinite series expansion of 𝑋(𝑧) with

𝑋(𝑧) = ∑ 𝑥(𝑘)𝑧 −𝑘
𝑘=0

we obtain

𝑥(0) = 0

𝑥(1) = 10

𝑥(2) = 17

𝑥(3) = 18.4

𝑥(4) = 18.68

36
CHAPTER TWO Z Transform

as seen from this example, the direct division method may be

carried out by hand calculations if only the first several terms of the
sequence are desired. In general, the method does not yield a closed-form
expression for 𝑥(𝑘), except in special cases.

Example 17

Find 𝑥(𝑘) when 𝑋(𝑧) is given by

1 𝑧 −1
𝑋(𝑧) = =
𝑧 + 1 1 + 𝑧 −1

By dividing the numerator by the denominator, we obtain

𝑧 −1
𝑋(𝑧) = = 𝑧 −1 − 𝑧 −2 − 𝑧 −3 − 𝑧 −4 + ⋯
1 + 𝑧 −1

By comparing this infinite series expansion of 𝑋(𝑧) with

𝑋(𝑧) = ∑ 𝑥(𝑘)𝑧 −𝑘
𝑘=0

we obtain

𝑥(0) = 0

𝑥(1) = 1

𝑥(2) = −1

𝑥(3) = 1

𝑥(4) = −1

This is an alternating signal of 1 and −1, which starts from 𝑘 = 1.

37
CHAPTER TWO Z Transform

3.2 Partial-Fraction-Expansion Method

The partial-fraction-expansion method presented here, which is

parallel to the partial-fraction-expansion method used in Laplace
transformation, is widely used in routin problems involving z transforms.

The method requires that ahh terms in the partial fraction

expansion be easily recognizable in the table of z transform pairs.

To find the inverse transform, if 𝑋(𝑧) has one or more zeros at the
𝑋(𝑧)
origin (𝑧 = 0), then or 𝑋(𝑧) is expanded into a sum of simple first- or
𝑧

second-order terms by partial fraction expansion, and a z transform table

is used to find the corresponding time function of each expanded term.

𝑋(𝑧)
It is noted the the only reason that we expand into partial
𝑧

fraction is that each expanded term has a form that may easily be found
from commonly available z transform tables.

Example 19

Before we discuss the partial-fraction-expansion method, we shall

review the shifting theorem. Consider the following 𝑋(𝑧):

𝑧 −1
𝑋(𝑧) =
1 − 𝑎𝑧 −1

By writing 𝑧𝑋(𝑧) as 𝑌(𝑧), we obtain

1
𝑧𝑋(𝑧) = 𝑌(𝑧) =
1 − 𝑎𝑧 −1

Referring to Table 2.1, the inverse z transform of 𝑌(𝑧) can be obtained as

follows:

𝒵 −1 [𝑌(𝑧)] = 𝑦(𝑘) = 𝑎𝑘

38
CHAPTER TWO Z Transform

Hence, the inverse z transform of 𝑋(𝑧) = 𝑧 −1 𝑌(𝑧) is given by

13
−1 [𝑋(𝑧)]
𝒵 = 𝑥(𝑘) = 𝑦(𝑘 − 1)

Since 𝑦(𝑘) is assumed to be zero for all 𝑘 < 0, we have

𝑦(𝑘 − 1) = 𝑎𝑘−1 , 𝑘 = 1, 2, 3, …
𝑥(𝑘) = {
0, 𝑘≤0

Consider 𝑋(𝑧) as given by

𝑏0 𝑧 𝑚 + 𝑏1 𝑧 𝑚−1 + ⋯ + 𝑏𝑚−1 𝑧 + 𝑏𝑚
𝑋(𝑧) = 𝑚≤𝑛
𝑧 𝑛 + 𝑎1 𝑧 𝑛−1 + ⋯ + 𝑎𝑛−1 𝑧 + 𝑎𝑛

To expand 𝑋(𝑧) into partial fractions, we first factor the denominator

polynomial os 𝑋(𝑧) and find the poles of 𝑋(𝑧):
16
𝑏0 𝑧 𝑚 + 𝑏1 𝑧 𝑚−1 + ⋯ + 𝑏𝑚−1 𝑧 + 𝑏𝑚
𝑋(𝑧) =
(𝑧 − 𝑝1 )(𝑧 − 𝑝2 ) … (𝑧 − 𝑝𝑛 )

𝑋(𝑧)
We then expand into partial fractions so that each term is easily
𝑧

recognizable in a table of z transforms. If the shifting theorem is utilized

𝑋(𝑧)
in taking inverse z transforms, however, 𝑋(𝑧), instead of , my be
𝑧

expanded into partial fractions. The inverse z transform of 𝑋(𝑧) is

obtained as the sum of the inverse z transforms of the partial fractions.

A commonly used procedure for the case where all the poles are of
simple order and there is at least one zero at the origin (that is, 𝑏𝑚 = 0) is
𝑋(𝑧)
to divide both sides of 𝑋(𝑧) by 𝑧 and then expand into partial
𝑧
𝑋(𝑧)
fractions. Once is expanded, it will be of the form
𝑧

𝑋(𝑧) 𝑎1 𝑎2 𝑎𝑛 17
= + +⋯+
𝑧 𝑧 − 𝑝1 𝑧 − 𝑝2 𝑧 − 𝑝𝑛

39
CHAPTER TWO Z Transform

The coefficient 𝑎𝑖 can be determined by multiplying both sides of this last

13
equation by 𝑧 − 𝑝𝑖 and setting 𝑧 = 𝑝𝑖 . This will result in zero for all the
terms on the right-hand side except the 𝑎𝑖 term, in which the
multiplicative factor 𝑧 − 𝑝𝑖 has been canceled by the denominator.
Hence, we have 13
18
𝑋(𝑧)
𝑎𝑖 = [(𝑧 − 𝑝𝑖 ) ]
𝑧 𝑧=𝑝𝑖

Note that such determination of 𝑎𝑖 is valid only for simple poles.

𝑋(𝑧)
If involves a multiple pole, for example, a douple pole at
𝑧
𝑋(𝑧)
𝑧 = 𝑝1 and no other poles, then will have the form
𝑧

𝑋(𝑧) 𝑐1 𝑐2 19
= +
𝑧 (𝑧 − 𝑝1 )2 𝑧 − 𝑝1

The coefficients 𝑐1 and 𝑐2 are determined from

𝑋(𝑧)
20
𝑐1 = [(𝑧 − 𝑝1 )2 ]
𝑧 𝑧=𝑝1

𝑑 𝑋(𝑧) 21
2
𝑐2 = { [(𝑧 − 𝑝1 ) ]}
𝑑𝑧 𝑧 𝑧=𝑝1

𝑋(𝑧)
It is noted that if involves a triple pole at 𝑧 = 𝑝1 , then the partial
𝑧
(𝑧+𝑝1 )
fractions must include a term
(𝑧−𝑝1 )3

Example 20

Given the z transform

(1 − 𝑒 −𝑎𝑇 )𝑧
𝑋(𝑧) =
(𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )

40
CHAPTER TWO Z Transform

where 𝑎 is a constant and 𝑇 is the sampling period, determine the

inverse z transform 𝑥(𝑘𝑇) by use of the partial-fraction-expansion
mathod.

𝑋(𝑧)
The partial fraction expansion of is found to be
𝑥

𝑋(𝑧) 1 1
= −
𝑧 𝑧 − 1 𝑧 − 𝑧 −𝑎𝑇

Thus,

1 1
𝑋(𝑧) = −
1 − 𝑧 −1 1 − 𝑒 −𝑎𝑇 𝑧 −1

From Table 2.1 we find

1
𝒵 −1 [ ]=1
1 − 𝑧 −1

1
𝒵 −1 [ ] = 𝑒 −𝑎𝑘𝑇
1 − 𝑒 −𝑎𝑇 𝑧 −1

Hence, the inverse z transform of 𝑋(𝑧) is

𝑥(𝑘𝑇) = 1 − 𝑒 −𝑎𝑘𝑇 , 𝑘 = 0, 1, 2, ….

Example 21

Let us obtain the inverse z transform of

𝑧2 + 𝑧 + 2
𝑋(𝑧) =
(𝑧 − 1)(𝑧 2 − 𝑧 + 1)

by use of the partial-fraction-expansion method.

We may expand 𝑋(𝑧) into partial fraction as follows:

4 −3𝑧 + 2 4𝑧 −1 −3𝑧 −1 + 2𝑧 −2
𝑋(𝑧) = + = +
𝑧 − 1 𝑧 2 − 𝑧 + 1 1 − 𝑧 −1 1 − 𝑧 −1 + 𝑧 −2

41
CHAPTER TWO Z Transform

Noting that the two poles involved in the quadratic term of this last
equation are complex conjugates, we rewrite 𝑋(𝑧) as follows:

4𝑧 −1 𝑧 −1 − 0.5𝑧 −2 0.5𝑧 −2
𝑋(𝑧) = − 3( )+
1 − 𝑧 −1 1 − 𝑧 −1 + 𝑧 −2 1 − 𝑧 −1 + 𝑧 −2

−1
1 −1
1 − 0.5𝑧 −2 −1
0.5𝑧 −2
= 4𝑧 − 3𝑧 +𝑧
1 − 𝑧 −1 1 − 𝑧 −1 + 𝑧 −2 1 − 𝑧 −1 + 𝑧 −2

Since

−𝑎𝑘𝑇
1 − 𝑒 −𝑎𝑇 𝑧 −1 cos 𝜔𝑇
𝒵[𝑒 cos 𝜔𝑘𝑇] =
1 − 2𝑒 −𝑎𝑇 𝑧 −1 cos 𝜔𝑇 + 𝑒 −2𝑎𝑇 𝑧 −2

−𝑎𝑘𝑇
1 − 𝑒 −𝑎𝑇 𝑧 −1 sin 𝜔𝑇
𝒵[𝑒 sin 𝜔𝑘𝑇] =
1 − 2𝑒 −𝑎𝑇 𝑧 −1 sin 𝜔𝑇 + 𝑒 −2𝑎𝑇 𝑧 −2
1 𝜋
By identifying 𝑒 −2𝑎𝑇 = 1 and cos 𝜔𝑇 = in this case, we have 𝜔𝑇 =
2 3

3
and sin 𝜔𝑇 = √ . Hence, we obtain
2

−1
1 − 0.5𝑧 −1 𝑘
𝑘𝜋
𝒵 [ ] = 1 cos
1 − 𝑧 −1 + 𝑧 −2 3

And

3
−1 (√2) 𝑧 −1
1 − 0.5𝑧 1 1 𝑘 𝑘𝜋
𝒵 −1 [ ] = 𝒵 −1
= 1 sin
1 − 𝑧 −1 + 𝑧 −2 √3 1 − 𝑧 −1 + 𝑧 −2 √3 3
[ ]

Thus, we have

(𝑘 − 1)𝜋 1 𝑘−1 (𝑘 − 1)𝜋

𝑥(𝑘) = 4(1𝑘−1 ) − 3(1𝑘−1 ) cos + (1 ) sin
3 √3 3

Rewriting, we have

42
CHAPTER TWO Z Transform

(𝑘 − 1)𝜋 1 (𝑘 − 1)𝜋
𝑘 = 1, 2, 3, …
𝑥(𝑘) = {4 − 3 cos 3
+
√3
sin
3 𝑘≤0
0

The first several values of 𝑥(𝑘) are given by

𝑋(0) = 0

𝑋(1) = 1

𝑋(2) = 3

𝑋(3) = 6

𝑋(4) = 7

𝑋(5) = 5

Note that the inverse z transform of 𝑋(𝑧) can also be obtained as follows:

−1
1 𝑧 −1 −1
𝑧 −1
𝑋(𝑧) = 4𝑧 − 3( ) + 2𝑧
1 − 𝑧 −1 1 − 𝑧 −1 + 𝑧 −2 1 − 𝑧 −1 + 𝑧 −2

Since

−1
𝑧 −1 1, 𝑘 = 1, 2, 3, …
𝒵 [ ] = {
1 − 𝑧 −1 0, 𝑘≤0

And

−1
𝑧 −1 2 𝑘 𝑘𝜋
𝒵 [ ] = 1 sin
1 − 𝑧 −1 + 𝑧 −2 √3 3

We have

𝑘𝜋 4 (𝑘 − 1)𝜋
𝑘 = 1, 2, 3, …
𝑥(𝑘) = {4 − 2√3 sin +
3 √3
sin
3 𝑘≤0
0

43
CHAPTER TWO Z Transform

3.3 Invesion Inegral Mathod

This is a useful technique fro obtaining the inverse z transform.

The inversion integral for the z transform 𝑋(𝑧) is given by

1 22
−1 [𝑋(𝑧)]
𝒵 = 𝑥(𝑘𝑇) = 𝑥(𝑘) = ∮ 𝑋(𝑧)𝑧 𝑘−1 𝑑𝑧
2𝜋𝑗 𝑐

Where 𝑐 is the circle with its center at the origin of the z plane such that
all poles of 𝑋(𝑧)𝑧 𝑘−1 are inside it.

The equation for giving the inverse z transform in terms of residues

can be derived by using theory of complex variables. It can be obtained as
follows:

𝑥(𝑘𝑇) = 𝑥(𝑘) = 𝐾1 + 𝐾2 + ⋯ + 𝐾𝑚

𝑚 23
𝑘−1 𝑘−1 ]
= ∑[residue of 𝑋(𝑧)𝑧 at pole 𝑧 = 𝑧𝑖 of 𝑋(𝑧)𝑧
𝑖=1

Where 𝐾1 , 𝐾2 , … , 𝐾𝑚 denote the residue of 𝑋(𝑧)𝑧 𝑘−1 at poles 𝑧1 , 𝑧2 ,

… , 𝑧𝑚 , respectively. In evaluating residues, note that if the denominator
of 𝑋(𝑧)𝑧 𝑘−1 contains a simple pole 𝑧 = 𝑧𝑖 then the corresponding residue
𝐾 is given by
24
𝐾 = lim [(𝑧 − 𝑧𝑖 )𝑋(𝑧)𝑧 𝑘−1 ]
𝑧→𝑧𝑖

If 𝑋(𝑧)𝑧 𝑘−1 contains a multiple pole 𝑧𝑖 of order 𝑞, then the residue 𝐾 is

given by

1 𝑑 𝑞−1 25
𝐾= lim 𝑞−1 [(𝑧 − 𝑧𝑖 )𝑞 𝑋(𝑧)𝑧 𝑘−1 ]
(𝑞 − 1)! 𝑧→𝑧 𝑖 𝑑𝑧

44
CHAPTER TWO Z Transform

Note that the values of 𝑘 in Equations (23), (24), and (5 are nonegative
integer values.

If 𝑋(𝑧) has a zero of order 𝑟 at the origin, then 𝑋(𝑧)𝑧 𝑘−1 in

Equation (23) will involve a zero of order 𝑟 + 𝑘 − 1 at the origin. If
𝑟 ≤ 0, then there will be a pole at 𝑧 = 0 for one or more nonegative
values of 𝑘. in such a case, separate inversion of Equation (23) is
necessary for each such value of 𝑘.

It should be noted that the inversion integral method, when

evaluated by residues, is a very simple technique for obtaining the inverse
transform, provided that 𝑋(𝑧)𝑧 𝑘−1 has no poles at the origin, 𝑧 = 0. If,
however, 𝑋(𝑧)𝑧 𝑘−1 has a simple pole or a multiple pole ar 𝑧 = 0, then
calculations may become cumbersome and the partial-fraction-expansion
method may prove to be simpler to apply. On the other hand, in certain
problems the partial-fraction-expansion method may become laborious.
Then, the inversion integral method proves to be very convenient.

Example 22

Obtain 𝑥(𝑘𝑇) by using the inversion integral method when 𝑋(𝑧) is

given by

𝑧(1 − 𝑒 −𝑎𝑇 )
𝑋(𝑧) =
(𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )

Note that

𝑘−1
(1 − 𝑒 −𝑎𝑇 )𝑧 𝑘
𝑋(𝑧)𝑧 =
(𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )

For 𝑘 = 0, 1, 2, …., 𝑋(𝑧)𝑧 𝑘−1 has two simple poles, 𝑧 = 𝑧1 = 1 and

𝑧 = 𝑧2 = 𝑒 −𝑎𝑇 . Hence, from Equation (24), we have

45
CHAPTER TWO Z Transform

2
(1 − 𝑒 −𝑎𝑇 )𝑧 𝑘
𝑥(𝑘) = ∑ [residue of at pole z = 𝑧𝑖 ]
(𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )
𝑖=1

= 𝐾1 + 𝐾2

where

𝐾1 = [residue at simple pole 𝑧 = 1]

(1 − 𝑒 −𝑎𝑇 )𝑧 𝑘
= lim [(𝑧 − 1) ]=1
𝑧→1 (𝑧 − 1)(𝑧 − 𝑒 −𝑎𝑇 )

𝐾2 = [residue at simple pole 𝑧 = 𝑒 −𝑎𝑇 ]

−𝑎𝑇
(1 − 𝑒 −𝑎𝑇 )𝑧 𝑘
= lim [(𝑧 − 𝑒 ) −𝑎𝑇
] = −𝑒 −𝑎𝑘𝑇
−𝑎𝑇
𝑧→𝑒 (𝑧 − 1)(𝑧 − 𝑒 )

Hence,

𝑥(𝑘𝑇) = 𝐾1 + 𝐾2 = 1 − 𝑒 −𝑎𝑇 , 𝑘 = 0, 1, 2, ….

Example 23

Obtain the inverse z transform of

𝑧2
𝑋(𝑧) =
(𝑧 − 1)2 (𝑧 − 𝑒 −𝑎𝑇 )

By using the inversion integral method

Notice that

𝑘−1
𝑧 𝑘+1
𝑋(𝑧)𝑧 =
(𝑧 − 1)2 (𝑧 − 𝑒 −𝑎𝑇 )

For 𝑘 = 0, 1, 2, …, 𝑋(𝑧)𝑧 𝑘−1 has a simple pole at 𝑧 = 𝑧1 = 𝑒 −𝑎𝑇 and a

double pole at 𝑧 = 𝑧2 = 1. Hence, from Equation (2.24), we obtain

46
CHAPTER TWO Z Transform

2
𝑧 𝑘+1
𝑥(𝑘) = ∑ [residue of at pole z = 𝑧𝑖 ]
(𝑧 − 1)2 (𝑧 − 𝑒 −𝑎𝑇 )
𝑖=1

= 𝐾1 + 𝐾2

Where

𝐾1 = [residue at simple pole 𝑧 = 𝑒 −𝑎𝑇 ]

−𝑎𝑇
𝑧 𝑘+1 𝑒 −𝑎(𝑘+1)𝑇
= lim [(𝑧 − 𝑒 ) ]=
−𝑎𝑇
𝑧→𝑒 (𝑧 − 1)2 (𝑧 − 𝑒 −𝑎𝑇 ) (1 − 𝑒 −𝑎𝑇 )2

𝐾2 = [residue at simple pole 𝑧 = 1]

1 𝑑 2
𝑧 𝑘+1
= lim [(𝑧 − 1) ]
(2 − 1)! 𝑧→1 𝑑𝑧 (𝑧 − 1)2 (𝑧 − 𝑒 −𝑎𝑇 )

𝑑 𝑧 𝑘+1
= lim ( )
𝑧→1 𝑑𝑧 𝑧 − 𝑒 −𝑎𝑇

(𝑘 + 1)𝑧 𝑘 (𝑧 − 𝑒 −𝑎𝑇 ) − 𝑧 𝑘+1

= lim
𝑧→1 (𝑧 − 𝑒 −𝑎𝑇 )2

𝑘 𝑒 −𝑎𝑇
= −
1 − 𝑒 −𝑎𝑇 (1 − 𝑒 −𝑎𝑇 )2

Hence,

𝑒 −𝑎𝑇 𝑒 −𝑎𝑘𝑇 𝑘 𝑒 −𝑎𝑇

𝑥(𝑘𝑇) = 𝐾1 + 𝐾2 = + −
(1 − 𝑒 −𝑎𝑇 )2 1 − 𝑒 −𝑎𝑇 (1 − 𝑒 −𝑎𝑇 )2

𝑘𝑇 𝑒 −𝑎𝑇 (1 − 𝑒 −𝑎𝑇 )
= − 𝑘 = 0, 1, 2, …
𝑇(1 − 𝑒 −𝑎𝑇 ) (1 − 𝑒 −𝑎𝑇 )2

47