Chapter 5

z-Transform
Click to edit Master subtitle style
Nguyen Khanh Loi, [Link].
nkloi@[Link]
 v The z-transform is a tool for analysis, design and implementation of
discrete-time signals and LTI systems.

v Convolution in time-domain Û multiplication in the z-domain

Digital Signal Processing 2 z-Transform

 Content

1. z-transform
2. Properties of the z-transform
3. Causality and Stability
4. Inverse z-transform

Digital Signal Processing 3 z-Transform

 1. The z-transform

v The z-transform of a discrete-time signal x(n) is defined as the power

series:
$

𝑋(𝑧) = & 𝑥(𝑛)𝑧 #! = ⋯ 𝑥(−2)𝑧 % + 𝑥(−1)𝑧 + 𝑥(0) + 𝑥(1)𝑧 #& + 𝑥(2)𝑧 #% + ⋯

!"#$

v The region of convergence (ROC) of X(z) is the set of all values of

z for which X(z) attains a finite value.
¥
ROC = {z Î C | X ( z ) = å x (
n = -¥
n ) z -n
¹ ¥}

v The z-transform of impulse response h(n) is called the transform

function of the filter:
¥
H ( z) = å h (
n = -¥
n ) z -n

Digital Signal Processing 4 z-Transform

 Example 1

v Determine the z-transform of the following finite-duration signals

a) x1(n)=[1, 2, 5, 7, 0, 1]
b) x2(n)=x1(n-2)
c) x3(n)=x1(n+2)
d) x4(n)=d(n)

e) x5(n)=d(n-k), k>0

f) x6(n)=d(n+k), k>0

Digital Signal Processing 5 z-Transform

 Example 2

v Determine the z-transform of the signal

a) x(n)=(0.5)nu(n)
b) x(n)=-(0.5)nu(-n-1)

v Remark:
When |x| < 1:

Digital Signal Processing 7 z-Transform

 z-transform and ROC

v It is possible for two different signal x(n) to have the same z-

transform. Such signals can be distinguished in the z-domain by their
region of convergence.
v z-transforms:

and their ROCs:

ROC of a causal signal is the ROC of an anticausal signal

exterior of a circle. is the interior of a circle.

Digital Signal Processing 9 z-Transform

 Example 3

v Determine the z-transform of the signal

x(n) = a nu(n) + b nu(-n - 1)

v The ROC of two-sided signal is a ring (annular region).

Digital Signal Processing 10 z-Transform

 2. Properties of the z-transform

v Linearity:
if x1 ( n) ¬¾® z
X 1 ( z ) with ROC1
and x2 (n) ¬ ¾® z
X 2 ( z ) with ROC 2
then
x(n) = x1 (n) + x2 (n) ¬
¾®z
X ( z ) = X 1 ( z ) + X 2 ( z ) with ROC = ROC1  ROC2

v Example: Determine the z-transform and ROC of the signals

x(n)=[3(2)n-4(3)n]u(n)

Digital Signal Processing 11 z-Transform

 2. Properties of the z-transform

v Time shifting:
if x ( n) ¬
¾®z
X ( z)
then x ( n - D) ¬
¾®z
z - D X ( z)
v The ROC of z - D X (z ) is the same as that of X(z) except for z=0 if
D>0 and z=¥ if D<0.
Example: Determine the z-transform of the signal x(n)=2nu(n-1).
v Convolution of two sequence:
if x1 (n) ¬
¾®z
X 1 ( z ) and x2 (n) ¬
¾® z
X 2 ( z)
then x(n) = x1 (n) * x2 (n) ¬
¾®z
X ( z) = X1 ( z) X 2 ( z)
the ROC is, at least, the intersection of that for X1(z) and X2(z).
Example: Compute the convolution of x=[1, 1, 3, 0, 2, 1] and h=[1, -2, 1] ?
Digital Signal Processing 12 z-Transform
 2. Properties of the z-transform

v Time reversal:
if x(n) ¬
¾® z
X ( z ) ROC : r1 £| z |£ r2
1 1
then x ( - n) ¬
¾®z
X ( z -1 ) ROC : £| z |£
r2 r1
Example: Determine the z-transform of the signal x(n)=u(-n).

v Scaling in the z-domain:

if x ( n) ¬
¾®z
X ( z ) ROC : r1 £| z |£ r2
then a n x(n) ¬
¾®z
X (a -1 z ) ROC : | a | r1 £| z |£| a | r2
for any constant a, real or complex

Digital Signal Processing 13 z-Transform

 3. Causality and stability

v A causal signal of the form

x(n) = A1 p1nu(n) + A2 p2nu(n) + 
will have z-transform
A1 A2
X ( z) = -1
+ -1
+  ROC | z |> max | pi |
1 - p1 z 1 - p2 z i

the ROC of causal signals are outside of the circle.

v A anticausal signal of the form

x(n) = - A1 p1nu(-n - 1) - A2 p2nu(-n - 1) + 
A1 A2
X ( z) = -1
+ -1
+  ROC | z |< min | pi |
1 - p1 z 1 - p2 z i

the ROC of causal signals are inside of the circle.

Digital Signal Processing 14 z-Transform
 3. Causality and stability

v Mixed signals have ROCs that are the annular region between two
circles.

v It can be shown that a necessary and sufficient condition for the

stability of a signal x(n) is that its ROC contains the unit circle.

Digital Signal Processing 15 z-Transform

 4. Inverse z-transform

- transform
x(n) ¾z¾ ¾¾® X ( z ), ROC
X ( z ), ROC ¾inverse
¾¾z-¾ ¾ ¾® x(n)
transform

x(n) ¬
¾®z
X ( z ), ROC

v In inverting a z-transform, it is convenient to break it into its partial

fraction (PF) expression form, i.e., into a sum of individual pole
terms whose inverse z transforms are known.
1
v Note that with X ( z ) = -1 we have
1 - az
ì a nu (n) if ROC | z |>| a | (causal signals)
x ( n) = í n
î- a u (-n - 1) if ROC | z |<| a | (anticausal signals)

Digital Signal Processing 16 z-Transform

 Partial fraction expansion
v In general, the z-transform is of the form
N ( z ) b0 + b1 z -1 +  + bN z - N
X ( z) = =
D( z ) 1 + a0 z -1 +  aM z - M
v The poles are defined as the solutions of D(z)=0. There will be M
poles, say at p1, p2,…,pM . Then, we can write
D( z ) = (1 - p1 z -1 )(1 - p2 z -1 )(1 - pM z -1 )
v If N < M and all M poles are single poles.

where

Digital Signal Processing 17 z-Transform

 Example 4d
v Compute all possible inverse z-transform of

Digital Signal Processing 18 z-Transform

 Partial fraction expansion

v If N=M

Where and for i=1,…,M

v If N> M

Digital Signal Processing 21 z-Transform

 Example 5od
v Compute all possible inverse z-transform of

Digital Signal Processing 22 z-Transform

 Example 6
v Determine the causal inverse z-transform of

Digital Signal Processing 25 z-Transform

 Example 7
v Determine the causal inverse z-transform of
5 − 5.4𝑧 !"
𝑎) 𝑋 𝑧 =
1 − 2.3𝑧 !" + 1.2𝑧 !#

8 + 7𝑧 !" − 0.4𝑧 !#
𝑏) 𝑋(𝑧) =
1 + 1.8𝑧 !" − 0.4𝑧 !#

7 + 2𝑧 !$
𝑐) 𝑋(𝑧) =
1 − 2𝑧 !"

Digital Signal Processing 27 z-Transform

 Partial fraction expansion
v Complex-valued poles: since D(z) have real-valued coefficients, the
complex-valued poles of X(z) must come in complex-conjugate pairs

Considering the causal case, we have

Writing A1 and p1 in their polar form, say,

with B1 and R1 > 0, and thus, we have

As a result, the signal in time-domain is

Digital Signal Processing 28 z-Transform

 Example 8
v Determine the causal inverse z-transform of

Digital Signal Processing 29 z-Transform

 Some common z-transform pairs

Digital Signal Processing 32 z-Transform

 Review

v Định nghĩa biến đổi z

v Ý nghĩa miền hội tụ của biến đổi z
v Mối liên hệ giữa miền hội tụ với đặc tính nhân quả và ổn định của
tín hiệu/hệ thống-LTI rời rạc.
v Biến đổi z của một số tín hiệu cơ bản: d(n), anu(n), anu(-n-1)
v Một số tính chất cơ bản (tuyến tính, trễ, tích chập) của biến đổi z
v Phân chia đa thức và biến đổi z ngược

Digital Signal Processing 33 z-Transform

 Homework 1

Digital Signal Processing 34 z-Transform

 Homework 2

Digital Signal Processing 35 z-Transform

 Homework 3

Digital Signal Processing 36 z-Transform

 Homework 4

Digital Signal Processing 37 z-Transform

 Homework 5

Digital Signal Processing 38 z-Transform

 Homework 6

v Tìm biến đổi z và miền hội tụ của các tín hiệu sau:
1) d(n + 2) – d(n – 2)
2) u(n – 2)
3) u(n + 2)
4) u(n + 2) – u(n – 2)
5) u(–n)
6) u(n) + u(–n)
7) u(n) – u(–n)
8) u(1–n)
9) u(|n|)
10) 2nu(–n)
11) 2nu(n–1)
12) 2nu(1–n)

Digital Signal Processing 39 z-Transform

 Homework 7

v Tìm biến đổi z và miền hội tụ của các tín hiệu sau:
1) cos(pn)u(n)
2) cos(pn/2)u(n)
3) sin(pn/2)u(n)
4) cos(pn/3)u(n)
5) sin(pn/3)u(n)
6) cos(pn)u(n-1)
7) cos(pn)u(1-n)
8) cos(pn)u(-n-1)
9) 2ncos(pn/2)u(n)
10) 2nsin(pn/2)u(n)
11) 3ncos(pn/3)u(n)
12) 3nsin(pn/3)u(n)

Digital Signal Processing 40 z-Transform

 Homework 8

v Liệt kê giá trị các mẫu (n=0, 1, 2, 3) của tín hiệu nhân quả có biến
đổi z sau:
1) 2z -1 /(1 – 2z -1)
2) 2z -1 /(1 + 2z -1)
3) 2/(1 – 4z -2)
4) 2/(1 + 4z -2)
5) 2z -1 /(1 – 4z -2)
6) 2z -1 /(1 + 4z -2)
7) 2z -2 /(1 – 4z -2)
8) 2z -2 /(1 + 4z -2)
9) 2z -1 /(1 – z -1 – 2z -2)
10) 2z -2 /(1 – z -1 – 2z -2)
11) 2z -1 /(1 – 3z -1 + 2z -2)
12) 2z -2 /(1 – 3z -1 + 2z -2)
Digital Signal Processing 41 z-Transform
 dX ( z )
nx (n) ¬¾® - z
Z
, ROC = Rx
dz