EE 322M : Signal processing

Department of Electronics and Electrical Engineerng

Indian Institute of Technology, Guwahati
Winter 2023

Topic 12 : The z transform

Instruction and notes by : Manish

1 The z -transform
We previously learned that the Laplace transform can be very useful for stability analysis of
the continuous signal. We can obtain similar stability analysis capabilities for discrete-time
signals by using a discrete-time counterpart to the Laplace transform. We had defined Laplace
transform as Z ∞
X(s) = x(t)e−st dt, s ∈ C
−∞

In a very similar way, we define the discrete time counterpart, also known as, the z transform
as ∞
X
X(z) = x[n]z −n , z ∈ C
n=−∞

where z = rejΩ . Here, z is any complex number expressed in the polar coordinates form.

1.1 Definition
Let’s begin with a function z = rejΩ . We can observe that the signal x[n] = z n is a complex
exponential signal, and thus, it will be an eigenfunction to an LTI system with impulse response
h[n]. Thus the output:

y[n] = h[n] ∗ z n
X∞
= h[k]z n−k
k=−∞
∞
X
=z ×n
h[k]z −k
k=−∞

so that we may write

∞
X
H(z) = h[k]z −k
k=−∞

which is the transfer function of this discrete-time system h[n]. This leads to the definition,
the z -transform of an arbitrary signal x[n] is
∞
X
X(z) = x[n]z −n , z ∈ C (1)
n=−∞

1
and the corresponding inverse transform is
I
1
x[n] = X(z)z n−1 dz (2)
2πj
In practice, we do not use the inverse z -transform formula much since it is not straight
forward to evaluate above integral.

1.2 Convergence
As is clear from the Eq. 1, the z -transform exists when the infinite sum converges, i.e., absolute
summability of x[n]z −n . Since |x[n]z −n | = |x[n]r−n |, we must have
∞
X
|x[n]r−n | < ∞
n=−∞

We define the region of convergence (ROC) as the range of r for which the above condition
holds true, or where z -transform converges.

We can remember that due to extra convergence, the Laplace transform could be defined
for the signals for which CTFT did not exist. In similar way, the r−n term provide extra con-
vergence here and hence z-transform exists for signals that may not have a DTFT. Remember
that DTFT required absolute summability of x[n]. In here, we can ensure that x[n]r−n is
absolutely summable even though x[n] is not.

1.3 Relationship between Laplace transform and z-Transform

The z transform can be obtained from Laplace transform by making the assignment z = esTs =
e(σ+jω)Ts = eσTs ejωTs . The magnitude of z is r = |z| = eσTs and the phase is ωTs .
Here, we are assuming Laplace transform for discrete signals i.e. Discrete time Laplace
transform (t = nTs ; and x(nTs ) = x[n])
Z ∞
LT: X(s) = x(t)e−st dt
−∞
∞
X
DTLT: X(esTs ) = x[n]e−nsTs
n=−∞

Replacing esTs = z will give us z-trnasform formula.

1.4 The z-plane

The z-plane and a unit circle in the z-plane is shown here in the figure below.

2
 The point z = rejΩ can be seen as a point located at a distance r from the origin and an
angle Ω from the positive real axis. If x[n] is absolutely summable, then DTFT is obtained
from the z-transform by setting r = 1. Or by substituting z = ejΩ in the Eq. 1. Or we say,
DTFT, in this case, is simply the z-transform evaluated along the unit circle in the z-plane.

1.5 Poles and Zeros

Most commonly, the z -transform appears as a ratio of two polynomials in z −1 , such as

b0 + b1 z −1 + ... + bM z −M
X(z) =
a0 + a1 z −1 + ... + aN z −N
The roots of the numerator are called zeros of X(z). The roots of the denominator are
called zeros of X(z).

1.6 Properties of ROC

We define the region of convergence (ROC) as the range of r for which
∞
X
|x[n]r−n | < ∞
n=−∞

The ROC is a disk/circular space in z-plane with center 0 and radius r. The ROC can
be an inward circle, a disc, or an outward circle as shown in the figure below. The complete
z-transform is always written along with its ROC. The ROC makes a z-transform unique.
Two very different sequences can have the same z-transform algebraic expressions but different
ROCs. Some important points to note are:

• The ROC of X(z) can not contain any poles.

• The ROC of an infinite length right-sided signal is of the form |z| > r+

• The ROC of an infinite length left-sided signal is of the form |z| < r−

• The ROC of an infinite length two-sided signal is of the form r+ < |z| < r−

• In the case of finite duration sequences, ROC is the entire plane except possibly at z = 0
and/or z = ∞.

• The ROC depends only on r. The Ω does not affect the convergence.

3
2 Examples
Example 1: Find the z -transform of the impulse signal δ[n].
Solution: ∞
X
X[z] = δ[n]z −n = δ[0]z 0 = 1
n=−∞

The ROC is entire z-plane.

Example 2: Find the z -transform of the signal x[n] = αn u[n].

Solution:
∞
X
X[z] = αn u[n]z −n
n=−∞
∞ 
X α n
=
n=0
z
1
= ; |z| > |α|
1 − αz −1
z
= ; |z| > |α|
z−α
The sum converges only if |α/z| < 1 or |z| > |α|. The pole and zero are present at z = α and
z = 0 respectively.

4
 Homework 1: Find z-transform of u[n].
1
Solution: From above example, it’s going to be: 1−z −1
; with ROC |z| > 1

Example 3: Find the z -transform of the signal x[n] = −αn u[−n − 1].
Solution:
∞
X
X[z] = −αn u[−n − 1]z −n
n=−∞
−1  
X α n
=−
n=−∞
z
∞  
X z k
=−
k=1
α
∞  
X z k
=1−
k=0
α
1
=1− ; |z| < |α|
1 − αz −1
z
= ; |z| < |α|
z−α
The sum converges only if |z/α| < 1 or |z| < |α|. The pole and zero are present at z = α and
z = 0 respectively. Note that this has the same expression as previous example but different
ROC.

Example 4: Find the z -transform of the signal x[n] = −(3/4)n u[−n − 1] + (−1/3)n u[n].
Answer:
z(2z − 5/12)
X(z) = ; 1/3 < |z| < 3/4
(z − 3/4)(z + 1/3)

5
 Example 5: Identify the ROC associated with the z-transform for each of the following
signals:

x[n] = −(1/2)n u[−n] + 2(1/4)n u[n];

y[n] = −(1/2)n u[n] + 2(1/4)n u[n];
w[n] = −(1/2)n u[−n] + 2(1/4)n u[−n];

Answer:

Example 6: Determine the z-transform for sin[Ωn]u[n].

Solution:
∞
X
X[z] = sin[Ωn]u[n]z −n
n=−∞
X∞
= sin[Ωn]z −n
n=0
∞
1 X jΩn
= (e − e−jΩn )z −n
2j n=0
∞
1 X jΩ −1 n
= (e z ) − (e−jΩ z −1 )n
2j n=0
zsinΩ
=
z2
− 2zcosΩ + 1
The ROC from the convergence of the series ∞ −n
P
n=0 sin[Ωn]z is |z| > 1.

Example 7: Determine the z-transform for sin[Ωn].

6
 P∞
Solution: The series n=−∞ sin[Ωn]z −n does not converge. Hence, z -transform does not
exist.

Homework 2: Determine the z-transform for cos[Ωn]u[n].

Answer: z2z(z−cos Ω)
−2zcosΩ+1
. The ROC is |z| > 1.

3 Properties of z -Transform
The properties of z-transform are similar to DTFT properties. A list is given below:

Derivative property
In addition, the derivative property is given as:

z d
nx[n] ←
→ −z X(z), with ROC of Rx
dz

Multiplication by αn
Let α be a complex number. Then
z
z
αn x[n] ←
→X , with ROC of |α|Rx
α
or, we also write this property in another way:
z
→ X(e−jzo z),
ejzo n x[n] ← with ROC of Rx

Compare this property with the DTFT frequency shift property. In DTFT, we get a similar
LHS when our RHS is in fact X(Ω−Ωo ). In z domain, the RHS of this property is not X(z−zo ),
but rather X(e−jzo z).

7
Pole zero cancellation
Let’s see an example
Example 8:

Solution: The answer to this at a = b would be

−5z/4
aX(z) + aY (z) = a .
(z − 1/4)(z − 3/2)

The ROC would be 1/4 < |z| < 3/2. The zero at 1/2 cancels the pole at 1/2. Since pole define
the ROC boundary, hence cancellation of pole results in larger ROC than the X(z).

Example 9: Determine z-Transform of

x[n] = an cos(Ωn)u[n]

Solution: From the answer of Homework 2:

z z(z − cos Ω)
cos(Ωn)u[n] ←
→
z2 − 2zcosΩ + 1
From the “Multiplication by αn ” property:

z 1 − a cos Ωz −1
an cos(Ωn)u[n] ←
→
1 − 2acosΩz −1 + a2 z −2
Example 10: Find the z-Transform of following signals

• x[n] = u[n − 2] ∗ (2/3)n u[n]

• x[n] = sin(πn/8 − π/4)u[n − 2]

• x[n] = 2n u[−n − 3]
z −1 sin(π/8)
Answers: (a) 1/(z − 1)(z − 2/3) with ROC |z| > 1. (b) z 2 −2z cos(π/8)+1
with ROC |z| > 1. (c)
−z 3
4(z−2)
with ROC |z| < 2.

8
Reference material
1. Textbook: Signals and Systems by Simon Haykin
2. Lecture notes are inspired from the course materials of JHU 520.214; MIT 6.003; Purdue
ECE-301; UVic ECE-260; and Imperial College E2.5
[Please report any typos in the notes by sending an email to the instructor.]