Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Control Systems II
Discrete-Time Systems

Bilal Komati

Lebanese University
Faculty of Engineering - Branch III - Hadath
Department of Electrical and Electronics Engineering

4th Year - Semester 7 - Fall 2021/2022

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 1 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 2 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 3 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Continuous-Time vs Discrete-Time Signals

1 x(t)
Continuous-time signals or analog
0.5 signals are defined for every value of
time.
t
−1 −0.5 0.5 1 Discrete-time signals are defined only
at certain specific values of time.
−0.5 (time instants need not to be
equidistant, but in practice they are
equally spaced).
−1

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 4 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Continuous-Time vs Discrete-Time Signals

x(t) 1 x(t)
1

0.5 0.5

t t
−1 −0.5 0.5 1 −1 −0.5 0.5 1

−0.5 −0.5

−1 −1

Continuous-Time Signal Discrete-Time Signal

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 5 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Continuous-Valued vs Discrete-Valued Signals

x(n)

Continuous-Valued vs Discrete-Valued Signals

Continuous-valued signal takes on all possible
values on a finite or an infinite range.
Discrete-valued signal takes on values from a
finite set of possible values.

P.S: The values of a continuous-time or

discrete-time signal can be continuous or discrete. n
−1 1 2 3 4 5 6 7 8

Digital Signal = Discrete-Time + Discrete-Valued Signal

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 6 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Analog-to-Digital Conversion (ADC)

A/D converter or ADC

xa (t) x(n) xq (n) 01011 · · ·

Sampler Quantizer Coder

Analog Discrete-time Quantized Digital

signal signal signal signal

ADC main components

Sampling: Continuous-time (CT) signal xa (t) → Discrete-time (DT) signal x(n).
Quantization: DT and CV signal x(n) → DT and DV (digital) signal xq (n).
Coding: xq (n) → b-bits binary vector or sequence.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 7 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Sampling

Sampling (Uniform)
taking samples of the analog signal every
T sec.
x(n) = xa (nT )
where:
−∞ < n < ∞ is an integer or n ∈ Z
T is the sampling time.
1
Fs = is the sampling rate
T How to select T or Fs ?
(samples/sec) or frequency (Hz).

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 8 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Quantization

Quantization
It is the process of converting a discrete-time (DT), continuous-valued (CV) signal into a
digital signal by representing the continuous-valued signal by a finite set of discrete value levels.

x(n) xq (n) = Q [x(n)]

Quantizer Quantization error
The error introduced from this process is called
quantization error or quantization noise:
Discrete-time
Quantized
signal eq (n) = xq (n) − x(n)
signal

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 9 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Quantization
Discrete-time (DT) continuous-valued (CV) signal

Example x(n)
1
Consider the analog signal: x(n) = 0.9n
0.8 xa (t) = 0.9t
t
xa (t) = 0.9 , t≥0
0.6
After Sampling:
( 0.4
T =1 sec 0.9n , n≥0
−−−−−→ xn = 0.2 T
Fs =1 Hz 0, n<0
···
n
1 2 3 4 5 6 7 8

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 10 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Quantization
Discrete-time (DT) continuous-valued (CV) signal Discrete-time (DT) discrete-valued (DV) signal (digital)

x(n) x(n) xa (t) = 0.9t xq (n) Levels of quantization

1 1
x(n) = 0.9n 0.9
0.8 xa (t) = 0.9t 0.8
0.7
∆: Resolution
0.6 Range 0.6
of the 0.5
0.4 quantizer 0.4
0.3
···
0.2 T 0.2
··· 0.1
n n
1 2 3 4 5 6 7 8 1 2 3 4 5 6 7 8 ···

∆: Quantization step or resolution L: number of quantization levels

xmin and xmax : minimum and maximum values of x(n) xmax − xmin
∆=
xmax − xmin : dynamic range of the signal L−1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 11 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Quantization
x(n) xq (n) xq (n) eq (n) = xq (n) − x(n)
n Discrete-time signal (Truncation) (Rounding) (Rounding)
x(t) = 0.9t 0 1 1.0 1.0 0.0
for t ≥ 0 1 0.9 0.9 0.9 0.0
2 0.81 0.8 0.8 −0.01
3 0.729 0.7 0.7 −0.029
T = 1sec 4 0.6561 0.6 0.7 0.0439
5 0.59049 0.5 0.6 0.00951
x(n) = 0.9n 6 0.531441 0.5 0.5 −0.031441
for n ≥ 0 7 0.4782969 0.4 0.5 0.0217031
8 0.43046721 0.4 0.4 −0.03046721
9 0.387420489 0.3 0.4 0.012579511

Quantization error (rounding)

−∆/2 ≤ eq (n) ≤ ∆/2
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 12 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Coding of Quantization Samples

The coding process assigns a unique binary number to each quantization level.
If we have L levels we need at least L different binary numbers.
With a word length ofb bits we can create 2b different binary numbers.

2b ≥ L ⇔ b ≥ log2 L
The minimum number of bits required in the coder is: bmin = ceil (log2 L).
In the previous example, L = 11, then the coder should have at least
b = ceil (log2 11) = ceil (log2 3.4) = 4 bits
Number of bits being produced per second
   
Bits Samples produced Bits
Bit rate = = ×
Second Second Sample
= Sampling rate × Quantization bits per sample

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 13 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Coding of Quantization Samples

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 14 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Digital to Analog (D/A) Conversion

To convert a digital signal into an analog signal we can use a digital-to-analog converter.
The task of a D/A converter is to interpolate between samples.
The simplest D/A converter is the zero-order hold or staircase interpolation.

Linear Interpolation
Zero-Order-Hold (ZOH)
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 15 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Sampling and Quantization of a Sinusoidal Signal

eq (t) = xa (t) − xq (t)

xa (t) is almost linear

between quantization levels.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 16 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 17 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Analog systems with piecewise constant inputs

In most engineering applications, it is necessary to control a physical system or plant so that

it behaves according to given design specifications.

the plant is analog,

the control is piecewise constant,
and the control action is updated periodically.

This arrangement results in an overall system that is conveniently described by a discrete-time

model.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 18 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Analog systems with piecewise constant inputs

Example
Consider the tank control system
H = steady-state fluid height in the tank
h = height perturbation from the nominal value
Q = steady-state flow rate through the tank
qi = inflow perturbation from the nominal value
q0 = outflow perturbation from the nominal value

It is necessary to maintain a constant fluid level H by adjusting the fluid flow rate into the
tank around Q.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 19 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Analog systems with piecewise constant inputs

The linearized model for the outflow valve is analogous to an electrical resistor: h = Rq0
h is the perturbation in tank level from nominal,
q0 is the perturbation in the outflow from the tank from a nominal level Q,
and R is the fluid resistance of the valve.
Assuming an incompressible fluid, the principle of conservation of mass reduces to the
volumetric balance: rate of fluid volume increase = rate of fluid volume in - rate of fluid
volume out:
dC (h + H)
= (qi + Q) − (q0 + Q)
dt
where C is the area of the tank or its fluid capacitance.
dh h qi
The analog mathematical model: + =
dt τ C
where τ = RC is the fluid time constant for the tank.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 20 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Analog systems with piecewise constant inputs

The solution of this differential equation is
Z t
−(t−t0 )/τ 1
h(t) = e h(t0 ) + e (t−λ)/τ qi (λ)dλ
C t0

Let qi be constant over each sampling period T –that is, qi (t) = qi (k) = constant for t in the
interval [k · T , (k + 1)T ]. Then we can solve the analog equation over any sampling period to
obtain
h(k + 1) = e −T /τ h(k) + R[1 − e −T /τ ]qi (k), k = 0, 1, 2, · · ·
where the variables at time kT are denoted by the argument k.
This is the desired discrete-time model describing the system with piecewise constant control.

The discrete-time model is known as a difference equation.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 21 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Difference Equations
w(t)
Digital Controller
Disturbances r(kT ) ê(kT ) Digital u(kT ) D/A u(t) Plant y(t)
Continuous Controller +
− and
Controller G(s)
hold
r(t) e(t) u(t) Plant y(t)
+ Controller
− G(s)
ŷ(kT ) Clock

ŷ(t)
Sampler
Sensor Sensor
A/D
T ŷ(t) 1
1
v(t)
The fundamental differences between the two implementations are that the digital system
operates on samples of the sensed plant output rather than on the continuous signal and that
the dynamics represented by D(s) are implemented by algebraic recursive equations called
difference equations.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 22 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Difference Equations
Consider the a controller to be implemented inside
a computer given by the transfer function: Euler’s method to approximate the
derivative
U(s) s +a δx
D(s) = = K0 ẋ = lim
E (s) s +b δt→0 δt
x(k + 1) − x(k)
First, find the differential equation that corresponds ẋ ∼
=
T
to D(s):
T = tk+1 − tk
(s + b)U(s) = K0 (s + a)E (s) tk = kT
Using the inverse Laplace transform: x(k) is the value of x at tk
x(k + 1) is the value of x at tk+1
u̇ + bu = K0 (ė + ae)

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 23 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Difference Equations

u̇ + bu = K0 (ė + ae)
Using Euler method for approximating the derivative:
u(k + 1) − u(k) e(k + 1) − e(k)
 
+ bu(k) = K0 + ae(k)
T T
Rearranging to put the difference equation in the desired form:
e(k + 1) − e(k)
  
u(k + 1) = u(k) + T −bu(k) + K0 + ae(k)
T
For computational efficiency, it is convenient to rearrange:
u(k + 1) = (1 − bT )u(k) + K0 (aT − 1)e(k) + K0 e(k + 1)
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 24 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Real-Time Controller Implementation

Algorithm: Real Time Controller implementation

x = 0 (initialization of past values for first loop through)
Define Constants:
α1 = 1 − bT
α2 = K0 (aT − 1)
Read A/D to obtain y and r
e =r −y
u = x + K0 e
OUTPUT u to D/A and ZOH
now compute x for the next loop through
x = α1 u + α2 e
go back to READ when T seconds have elapsed since last READ

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 25 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Difference Equations
Suppose we call:
the input signals to the k-th sample e0 , e1 , e2 , · · · , ek
the output signals prior to that time u0 , u1 , u2 , · · · , uk−1
The next output can be written as a function of the above values:

uk = f (e0 , · · · , ek , u0 , · · · , uk−1 )

This equation is the general difference equation. It can be linear or nonlinear.

In the case of linear difference equation:

uk = −a1 uk−1 − a2 uk−2 − · · · − an uk−n + b0 ek + b1 ek−1 + · · · + bm ek−m

If the coefficients ai and bi are constants, the difference equation is referred to as a linear time
invariant LTI.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 26 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 27 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The z-Transform
Definition
The z-transform of a discrete-time signal x(k) is defined as the power series
∞
X
X (z) = x(k)z −k
k=−∞
where z is a complex variable.

Definition (One-sided z-transform)

Given the causal sequence {x0 , x1 , x2 , · · · , xk , · · · }. We define the z-transform of the signal as:
k=∞
X X∞
−k
X (z) = Z [x(k)] = xk z = xk z −k
k=−∞ k=0
−1
= x0 + x1 z + x2 z −2 + · · · + xk z −k + · · ·
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 28 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The z-Transform

Definition
Given the impulse train representation of a discrete-time signal,
k=∞
X
x ∗ (t) = x0 δ(t) + x1 δ(t − T ) + x2 δ(t − 2T ) + · · · + xk δ(t − kT ) + · · · = xk δ(t − kT )
k=−∞
Its Laplace transform:
k=∞
X  k
X ∗ (s) = x0 + x1 e −sT + x2 e −2sT + · · · + xk e −ksT + · · · = xk e −sT
k=−∞
Let z be defined by: z = e sT
Then the z-transform expression can be determined.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 29 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

z-Transforms of standard discrete-time signals

Having defined the z-transform, we now obtain the z-transforms of commonly used
discrete-time signals such as the sampled step, exponential, and the discrete-time impulse.
The following identities are used repeatedly to derive several important results:
n
X 1 − an+1
ak = , a ̸= 1
1−a
k=0
∞
X 1
ak = , a<1
1−a
k=0

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 30 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

z-Transforms of standard discrete-time signals

Example
Consider the discrete-time impulse
(
1, k = 0
x(k) = δ(k) =
0, k ̸= 0

Applying the definition gives the z-transform:

X (z) = 1
Alternatively, one may consider the impulse-sampled version of the delta function
u ∗ (t) = δ(t). This has the Laplace transform
X ∗ (s) = 1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 31 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

z-Transforms of standard discrete-time signals

Example
Consider the discrete-time step
(
1, k ≥0
x(k) = 1(k) =
0, k <0

Applying the definition gives the z-transform:

∞ ∞
X
−k
X 1
X (z) = x(k)z = z −k = , |z −1 | < 1
1 − z −1
k=−∞ k=0
z
= , |z| > 1
z −1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 32 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

z-Transforms of standard discrete-time signals

Example
Consider the discrete-time one-sided exponential in time
(
ak , k ≥ 0
x(k) =
0, k < 0

Applying the definition gives the z-transform:

∞ ∞ ∞
X
−k
X
k −k
X
k 1
X (z) = x(k)z = a z = az −1 = , |az −1 | < 1
1 − az −1
k=−∞ k=0 k=0
z
= , |z| > |a|
z −a
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 33 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The z-Transform of a few elementary signals

1 The Unit Pulse

X1 (z) = 1
2 The Unit Step
1 z
X2 (z) = −1
= if |z| > 1
1−z z −1
3 Exponential
1 z
X3 (z) = −1
= if |z| > |r |
1 − rz z −r

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 34 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Properties of the z-Transform

1 Linearity : Z [α1 f1 (kT ) + α2 f2 (kT )] = α1 F1 (z) + α2 F2 (z)
2 Convolution of time-Sequences:
" ∞ #
X
Z f1 (l)f2 (k − l) = F1 (z)F2 (z)
l=−∞
3 Time Shift:
Z [f (k + n)] = z n F (z)
4 Scaling in the z-Plane:
Z [r −k f (k)] = F (rz)
5 Multiplication by time:
dF (z)
Z [kTf (k)] = −zT
dz
6 Final-Value Theorem:
lim f (k) = lim (z − 1)F (z)
k→∞ z→1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 35 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The final value theorem

The final value theorem allows us to calculate the limit of a sequence as k tends to infinity, if
one exists, from the z-transform of the sequence.
The main pitfall of the theorem is that there are important cases where the limit does not exist.
The two main cases are as follows:
1 An unbounded sequence
2 An oscillatory sequence

Theorem (The Final Value Theorem)

If a sequence approaches a constant limit as k tends to infinity, then the limit is given by
 
  z −1
f (∞) = lim f (k) = lim 1 − z −1 F (z) = lim F (z) = lim (z − 1) F (z)
k→∞ z→1 z→1 z z→1

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 36 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The final value theorem

Example
Verify the final value theorem using the z-transform of a decaying exponential sequence and
its limit as k tends to infinity.

The z-transform pair of an exponential sequence is

z Z
e −akT −→
z − e −aT
with a > 0, the limit as k tends to infinity in the time domain is
f (∞) = lim e −akT = 0
k→∞
The final value theorem gives
z −1
  
z
f (∞) = lim =0
z→1 z z − e −aT
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 37 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The final value theorem

Example
Obtain the final value for the sequence whose z-transform is

z 2 (z − a)
F (z) =
(z − 1)(z − b)(z − c)

What can you conclude concerning the constants b and c if it is known that the limit exists?

Applying the final value theorem, we have

z(z − a) 1−a
f (∞) = lim =
z→1 (z − b)(z − c) (1 − b)(1 − c)
The condition for the sequence to converge to a constant limit and for the validity of the final
value theorem are |b| < 1 and |c| < 1.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 38 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 39 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Convolution summation
The response of a discrete-time system to a unit impulse is known as the impulse response
sequence.
The impulse response sequence can be used to represent the response of a linear discrete-time
system to an arbitrary input sequence

{u(k)} = {u(0), u(1), u(2), · · · u(i), · · · }

To derive this relationship, we first represent the input sequence in terms of discrete impulses
as follows:

u(k) = u(0)δ(k) + u(1)δ(k − 1) + u(2)δ(k − 2) + · · · + u(i)δ(k − i) + · · ·

X∞
= u(i)δ(k − i)
i=0

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 40 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Convolution summation
For a linear system, the principle of superposition applies, and the system output due to the
input is the following sum of impulse response sequences:
{y (l)} = {h(l)} u(0) + {h(l − 1)} u(1) + · · · + {h(l − i)} u(i) + · · ·

Hence, the output at time k is given by

∞
X
y (k) = h(k) ∗ u(k) = h(k − i)u(i)
i=0
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 41 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Convolution summation
For a causal system, the response due to an impulse at time i is an impulse response starting
at time i and the delayed response h(k − i) satisfies
h(k − i) = 0, i >k
A causal system is one whose impulse response is a causal time sequence.
y (k) = u(0)h(k) + u(1)h(k − 1) + u(2)h(k − 2) + · · · + u(k)h(0)
k
X
= u(i)h(k − i)
i=0
A simple change of summation variable (j = k − i) transforms to
y (k) = u(k)h(0) + u(k − 1)h(1) + u(k − 2)h(2) + · · · + u(0)h(k)
k
X
= u(k − j)h(j)
j=0
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 42 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Convolution summation
Theorem (Response of an LTI System)
The response of an LTI discrete-time system to an arbitrary input sequence is given by the
convolution summation of the input sequence and the impulse response sequence of the
system.

2
X
y (2) = u(i)h(2 − i)
i=0
= u(0)h(2) + u(1)h(1) + u(2)h(0)
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 43 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The convolution theorem

Theorem (The Convolution Theorem)
The z-transform of the convolution of two time sequences is equal to the product of their
z-transforms.
Proof:
∞
X
y (k) = h(k) ∗ u(k) = u(i)h(k − i) Using the causality property
i=0
∞ ∞ X
"∞ # "∞ # ∞ 
Z
X X
−k
−→ Y (z) = y (k)z = u(i)h(k − i) z −k
X X
Y (z) = u(i)z −i  h(j)z −j 
k=0 k=0 i=0 i=0 j=0
∞ X
X ∞
Y (z) = u(i)h(j)z −(i+j) taking j = k − i = H(z)U(z)
i=0 j=−i
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 44 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The convolution theorem

Z
y (k) = h(k) ∗ u(k) Y (z) = H(z)U(z)

The function H(z) is known as the z-transfer function or simply the transfer function.

Applying the convolution theorem to the response of an LTI system allows us to use the
z-transform to find the output of a system without convolution as follows:
1 z-transform the input.
2 Multiply the z-transform of the input and the z-transfer function.
3 Inverse z-transform to obtain the output temporal sequence.

An added advantage of this approach is that the output can often be obtained in closed form.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 45 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The convolution theorem

Example
Given the discrete-time system y (k + 1) − 0.5y (k) = u(k), y (0) = 0
Find the impulse response of the system h(k):
1 From the difference equation
2 Using z-transformation

Solution:

1 Let u(k) = δ(k), then

y (1) = 0.5y (0) + u(0) = 1 (
0.5i−1 , i = 1, 2, 3, · · ·
y (2) = 0.5y (1) + u(1) = 0.5 h(i) = y (i) =
0, i <1
y (3) = 0.5y (2) + u(2) = 0.52
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 46 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The convolution theorem

Example
Given the discrete-time system y (k + 1) − 0.5y (k) = u(k), y (0) = 0
Find the impulse response of the system h(k):
1 From the difference equation
2 Using z-transformation

2 z-transforming the difference equation:

Y (z) 1
H(z) = =
U(z) z − 0.5
Inverse-transforming:
(
0.5i−1 , i = 1, 2, 3, · · ·
h(i) = y (i) =
0, i <1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 47 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 48 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The Discrete Transfer Function

The z-transform has the same role in discrete systems that the Laplace transform has in
analysis of continuous systems.
The z-transform of the sequence uk is
X∞
U(z) = uk z −k
k=−∞
If we consider the trapezoid approximation of the integral which is given by:
T
uk = uk−1 + (ek + ek−1 )
2
−k
Multiplying by z and summing over k:
∞ ∞ ∞ ∞
!
X X T X X
uk z −k = uk−1 z −k + ek z −k + ek−1 z −k
2
k=−∞ k=−∞ k=−∞ k=−∞
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 49 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The Discrete Transfer Function

 
∞ ∞ ∞ ∞
X X T  X X
uk z −k = uk−1 z −k + ek z −k + ek−1 z −k 
2
k=−∞ k=−∞ k=−∞ k=−∞

U(z)
let k − 1 = j
E (z) z −1 E (z)
∞
X ∞
X
uk−1 z −k = uj z −(j+1) = z −1 U(z)
k=−∞ j=−∞
We can obtain:
T 
U(z) = z −1 U(z) + E (z) + z −1 E (z)

2
T 1 + z −1 U(z) T z +1
⇒ U(z) = −1
E (z) ⇒ H(z) = =
2 1−z E (z) 2 z −1
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 50 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The Discrete-Transfer Function

A general form of transfer function can be given as follows:
b0 + b1 z −1 + · · · + bm z −m
H(z) =
1 + a1 z −1 + +a2 z −2 + · · · + an z −n
If n ≥ m:
b0 z n + b1 z n−1 + · · · + bm z n−m
H(z) =
z n + a1 z n−1 + a2 z n−2 + · · · + an
Or,
b(z)
H(z) =
a(z)
The general input-output relation between transforms:
U(z) = H(z)E (z)
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 51 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The Discrete-Transfer Function

Qm
b(z) (z − zi )
H(z) = = K Qni=1
a(z) i=1 (z − pi )

The places in z where b(z) = 0 are zeros of the transfer function.

The places in z where a(z) = 0 are poles of the transfer function.
If z0 is a pole and (z − z0 )p H(z) has neither pole nor zero at z0 , we say H(z) has a pole
of order p at z0 .
If p = 1, the pole is simple.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 52 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

ADC Model
Assume that
ADC outputs are exactly equal in magnitude to their inputs (i.e., quantization errors are
negligible).
The ADC yields a digital output instantaneously.
Sampling is perfectly uniform (i.e., occurs at a fixed rate).
Then the ADC can be modeled as an ideal sampler with sampling
period T

Assumptions are idealizations

Quantization errors are typically small but nonzero.
Variations in sampling rate occur but are negligible.
Physical ADCs have a finite conversion time
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 53 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

DAC model

DAC outputs are exactly equal in magnitude to their inputs.

Assume that: The DAC yields an analog output instantaneously.
DAC outputs are constant over each sampling period.
Then the input-output relationship of the DAC is given by
ZOH
{u(k)} −−−→ u(t) = u(k), kT ≤ t ≤ (k + 1)T , k = 0, 1, 2, · · ·

This equation describes a zero-order hold (ZOH).

First-order hold constructs analog signals in terms
of straight lines
Second-order hold constructs them in terms of
parabolas.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 54 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

The transfer function of the ZOH

To obtain the transfer function of the ZOH, we replace the number or discrete impulse by an
impulse δ(t). The transfer function can then be obtained by Laplace transformation of the
impulse response.

1
L {1(t)} =
s
e −sT
L {1(t − T )} =
s
Thus, the transfer function of the ZOH is

1 − e −sT
GZOH (s) =
s

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 55 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Discrete Models of Sampled-Data Systems

The discrete transfer function is the
z-transform of the samples of the
output when the input samples are
the unit pulse at k = 0
(
1 k=0
δ(k) =
0 k ̸= 0
The transfer function of the DAC and analog subsystem cascade is
  G (s)
G1 (s) = G (s)GZOH (s) = 1 − e −Ts
s
The unit pulse response of the plant is:
  G (s)
Y1 (s) = G1 (s) · 1 = 1 − e −Ts
s
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 56 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Discrete Models of Sampled-Data Systems

The required z-transform of the samples of the inverse of Y1 (s) is
G (z) = Z [y1 (kT )] = Z L −1 (Y1 (s)) := Z [Y1 (s)]
 
 
−Ts G (s)
= Z (1 − e )
s
   
G (s) −Ts G (s)
=Z −Z e
s s
where    
−Ts G (s) −1 G (s)
Z e =z Z
s s
Then,  
G (s)
G (z) = 1 − z −1 Z


s
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 57 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 58 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Frequency Response of Discrete-Time Systems

The frequency response studies the steady-state response of a discrete-time system to a

sampled sinusoidal input.
The response is a sinusoid of the same frequency as the input with frequency-dependent
phase shift and magnitude scaling.
The scale factor and phase shift define a complex function of frequency known as the
frequency response.

The well-known relationship between the transfer function Ha (s) and the frequency response
Ha (jω):
Ha (jω) = Ha (s)|s=jω

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 59 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Spectrum of a Sampled Signal

Using the impulse modulation as the mathematical representation of sampling, the
impulse-sampled representation of a discrete-time waveform is
∞
X
∗
r (t) = r (t)δ(t − kT )
k=−∞

where r (kT ) is the value at time kT , and δ(t − kT ) denotes a Dirac delta at time kT .

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 60 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Spectrum of a Sampled Signal

∞
X
The train of impulses δ(t − kT ), being periodic, can be represented by a Fourier series
k=−∞
∞
X ∞
X
δ(t − kT ) = Cn e j(2πn/T )t
k=−∞ n=−∞

where the Fourier coefficients, Cn , are given by the integral over one period as
∞
1 T /2 X
Z
Cn = δ(t − kT )e −jn(2πt/T ) dt
T −T /2
k=−∞ The shifting property of the impulse
Z T /2
1
δ(t)e −jn(2πt/T ) dt
Z ∞
=
T −T /2 f (t)δ(t − a)dt = f (a)
−∞
1
=
T
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 61 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Spectrum of a Sampled Signal

Thus, the representation for the sum of impulses as a Fourier Series is
∞ ∞ ∞
X 1 X j(2πn/T )t 1 X jnωs t 2π
δ(t − kT ) = e = e with ωs =
T n=−∞ T n=−∞ T
k=−∞

Taking the Laplace transform of the output of the mathematical " sampler,
∞ ∞
Z ∞ #
X L 1 X
r ∗ (t) = r (t)δ(t − kT ) −→ L [r ∗ (t)] = r (t) e jnωs t e −st dt
−∞ T n=−∞
k=−∞
∞ Z ∞ ∞ Z ∞
1 X 1 X
⇒R ∗ (s) = r (t)e jnωs t e −st dt = r (t)e −(s−jnωs )t dt
T n=−∞ −∞ T n=−∞ −∞
∞
∗ 1 X
⇒ R (s) = R(s − jnωs )
T n=−∞
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 62 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Spectrum of a sampled system

The sketch of a spectrum of continuous time signal and discrete-time signals
|R(f )|

−fB fB f
∗
|R (f )|

−2fs −fs −fB fB fs 2fs f

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 63 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist-Shannon Sampling Theorem

|S ∗ (f )|

f
−2fs −fs fs 2fs

One can recover a signal from its samples if the sampling frequency (fs = 1/T ) is at least
twice the highest frequency in the signal: i.e. fs > 2fB

In control, usually the sampling frequency is chosen: 6fB < fe < 25fB
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 64 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Sampling Theorem
Aliasing in the time domain

0.5

2 4 6 8 10

−0.5

−1

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 65 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Table of Contents

1 Digital Systems

2 Difference Equation

3 The z-Transform

4 Time response

5 Transfer Function

6 Sampling Theorem

7 Stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 66 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Stability

Definition
A system is stable if all solutions remain within a small distance of each other after a
specific discrete value of time
Asymptotic stability refers to the condition that all the solution to the discrete model are
stable and the distance between them goes to zero as k → ∞.
A system is called Bounded-Input-Bounded-Output (BIBO) stable if for every Bounded
Input, we should have a Bounded Output.

Asymptotic Stability for a linear system implies both stability and BIBO stability

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 67 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Asymptotic and Exponential Stability

Definition
A system is said to be asymptotically stable if its response to any initial conditions decays to
zero asymptotically in the steady state,that is, the response due to the initial conditions
satisfies
lim y (k) = 0
k→∞

If the response due to the initial conditions remains bounded but does not decay to zero, the
system is said to be marginally stable.

Exponential stability
A system is said to be exponentially stable if it is asymptotically and in addition its response to
any initial conditions decays to zero exponentially in the steady state.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 68 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Bounded-Input–Bounded-Output Stability
The second definition of stability concerns the forced response of the system for a bounded
input. A bounded input satisfies the condition
|u(k)| < bu , k = 0, 1, 2, · · ·
0 < bu < ∞

Definition
A system is said to be bounded-input–bounded-output (BIBO) stable if its response to any
bounded input remains bounded–that is, for any input satisfying the above condition, the
output satisfies
|y (k)| < by , k = 0, 1, 2, · · ·
0 < by < ∞

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 69 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Stability conditions
Theorem (Asymptotic Stability)
In the absence of pole-zero cancellation, an LTI digital system is asymptotically stable if its
transfer function poles are in the open unit disc and marginally stable if the poles are in the
closed unit disc with no repeated poles on the unit circle.
Im(s) Im(z)

Stable Unstable Stable A system is stable if and only if all its

Re(z)
Region RegionRe(s) -1 Region 1 poles are inside the unit circle
Unstable
⇔ |pj | < 1 for all j
Region
s-plane z-plane
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 70 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

BIBO stability
BIBO stability concerns the response of a system to a bounded input. The response of the
system to any input is given by the convolution summation
k
X
y (k) = h(k − i)u(i), k = 0, 1, 2, · · · h(k) : impulse response
i=0

Theorem Because the z-transform of the impulse

A discrete-time linear system is BIBO stable response is the transfer function:
if and only if its impulse response sequence is
Theorem
absolutely summable–that is,
A discrete-time linear system is BIBO stable
∞
X if and only if the poles of its transfer function
|h(i)| < ∞ lie inside the unit circle.
i=0

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 71 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Internal Stability

We consider that system as having two outputs, Y and U, and two inputs, R and D. Thus,
the transfer functions associated with the system are
 
  C (z)GZAS (z) GZAS (z)  
Y (z)  1 + C (z)G
ZAS (z) 1 + C (z)GZAS (z)  R(z)
=
C (z)(z)

C (z)GZAS (z)  D(z)
U(z)
1 + C (z)GZAS (z) 1 + C (z)GZAS (z)
It is not sufficient to prove that the output of the controlled system Y is bounded for bounded
reference input R because U can be unbounded.
The system output
Discrete-Time must be bounded in the
Systems (ULFG) presence
Control Systems II of a disturbance. Fall 2021/2022 72 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Internal Stability

Because internal stability guarantees the stability of the transfer function from R to Y , among
others, it is obvious that an internally stable system is also externally stable.
However, external stability does not, in general, imply internal stability.
Definition
If all the transfer functions that relate system inputs (R and D) to the possible system outputs
(Y and U) are BIBO stable, then the system is said to be internally stable.

Theorem
The closed-loop system is internally stable if and only if all its closed-loop poles are in the
open unit disc.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 73 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Jury Stability Test

b0 + b1 z −1 + b2 z −2 + · · · + bm z −m
H(z) =
a0 + a1 z −1 + a2 z −2 + · · · + an z −n
z n b0 + b1 z −1 + b2 z −2 + · · · + bm z −m


H(z) =
a0 z n + a1 z n−1 + a2 z n−2 + · · · + an−1 z + an
We draw the table:
× cj = a0 × aj − an−j × an
Row 1- a0 a1 a2 · · · aj · · · an−2 an−1 an
Row 2- an an−1 × an−2 · · · an−j · · · a2 a1 a0
Row 3- c0 c1 c2 · · · cj · · · cn−2 cn−1
Row 4- c n−1 cn−2 cn−3 · · · c n−j−1 · · · c1 c0
Row 5- d0 d1 d2 · · · dj · · · dn−2
.. .. .. ..
. . . . dj = c0 × cj − cn−j−1 × cn−1
Row 2n − 3 x0 x1 x2

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 74 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Jury Stability Test

We need 2n − 3 lines to perform the Jury stability test.
The system is stable if all the below conditions are valid:
a0 > |an |
D(1) > 0 if n is even where:
D(1) < 0 if n is odd
D(z) = a0 z n + a1 z n−1 + · · · + an−1 z + an
For each row:
|c0 | > |cn−1 |



|d0 | > |dn−2 | The main advantage of Jury Criteria is that it is an easy


.. to implement algorithm to study the stability.


 .

|x0 | > |x2 |


Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 75 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist criterion
The Nyquist criterion allows us to answer two questions:
1 Does the system have closed-loop poles outside the unit circle?
2 If the answer to the first question is yes, how many closed-loop poles are outside the unit
circle?
We begin by considering the closed-loop characteristic polynomial

NL (z)
pcl (z) = 1 + KD(z)G (z) = 1 + L(z); L(z) = : loop gain
DL (z)
NL (z) NL (z) + DL (z)
pcl (z) = 1 + =
DL (z) DL (z)

P = Number of of open-loop poles outside the unit circle; known

Z = Number of closed-loop poles outside the unit circle; unknown
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 76 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist criterion

In summary the discrete Nyquist criterion is

1 Determine the number, P, of unstable poles of KDG .
2 Plot KD(z)G (z) for the unit circle, z = e jωT and 0 ≤ ωT ≤ 2π. This is a
counter-clockwise path around the unit circle. Points for the plot can be conveniently
taken from a discrete Bode plot of KDG .
3 Set N equal to the net number of counter-clockwise (same direction) encirclements of the
point −1 on the plot.
4 Compute Z = P − N. The system is stable if and only if Z = 0.

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 77 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist criterion

Theorem (Nyquist Criterion)

Let the number of counterclockwise encirclements of the point (−1, 0) for a loop gain L(z)
when traversing the stability contour be N (i.e., −N for clockwise encirclements), where L(z)
has P open-loop poles inside the contour. Then the system has Z closed-loop poles outside
the unit circle with Z given by
Z = −N + P

Corollary
An open-loop stable system is closed-loop stable if and only if its Nyquist plot does not
encircle the point (−1, 0) (i.e., if N = 0).

Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 78 / 80

Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist criterion
Example
1
Consider a linearized model of a furnace: G (s) =
+ 3s + 1 s2
Determine the closed-loop stability of the system with digital control and a sampling period of
0.01.

4.95z + 4.901
G (z) = 10−5
z2 − 1.97z + 0.9704

No encirclement for the point (−1, 0): N = 0.

No unstable poles for the open loop transfer function:
P = 0.
Z = −N + P = 0 ⇒ the closed-loop system is stable.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 79 / 80
Digital Systems Difference Equation The z-Transform Time response Transfer Function Sampling Theorem Stability

Nyquist criterion

Example
Consider the previous model of the furnace after adding
the discrete-time first-order actuator
0.9516
Ga (z) = K
z − 0.9048
Determine the closed-loop stability of the system with
digital control and a sampling period of 0.01.

The open-loop transfer function is given by:

4.711z + 4.644
H(z) = G (z) · Ga (z) = 10−5
z3 − 2.875z 2 + 2.753z − 0.8781
The system is stable for gain K = 1.
The system is unstable for gain K = 5 since N = 2 and P = 0.
Discrete-Time Systems (ULFG) Control Systems II Fall 2021/2022 80 / 80