Continuous-Time System Analysis

Using The Laplace Transform

Dr. Kumbirayi Nyachionjeka
University of Botswana
 Outline
• Laplace Transform
• Properties of Laplace Transform
• Solution of Differential Equations
• Analysis of Electrical Networks
• Block Diagrams and System Realization
• Frequency Response of an LTIC System
• Filter Design by Placement of Poles and Zeros of H(s)
 𝑥 𝑡 =∑ 𝑋 (𝑠𝑖 )𝑒
𝑠𝑡
𝑥 (𝑡 )
LT
𝑋 (𝑠) ( ) 𝑖 si : complex frequency = i + ji
i : rate of decay
𝑖 i : rate of oscillation
−0 𝑡
𝑥1 =𝑒 cos ⁡( 2 𝜋 0 𝑡 )

−1 𝑡
𝑥 2=0.8 𝑒 cos ⁡( 2 𝜋 0 𝑡 )

−0 𝑡
𝑥 3=1.2 𝑒 cos ⁡( 2 𝜋 5 𝑡 )

− 1𝑡
𝑥 4=1.5 𝑒 cos ⁡(2 𝜋 5 𝑡 )

−1 𝑡
𝑥5 =1.5 𝑒 cos ⁡(2 𝜋 5 𝑡 − 𝜋 / 4)

0.5 𝑡
𝑥 6=0.7 𝑒 cos ⁡(2 𝜋 3 𝑡 )

𝑥 (𝑡 )= 𝑥 1+ 𝑥 2 + 𝑥 3 +…+ 𝑥 6
 Why Laplace Transform?

Laplace transform is a mathematical tool to map signals and

system behavior from the time-domain into the frequency domain.

Time Domain Laplace Domain (frequency)

𝑥 ( 𝑡 )= ∑ 𝑥(𝑛𝑇 )𝛿(𝑡 − 𝑛𝑇) 𝑥 ( 𝑡 )=∑ 𝑋 (𝑠𝑖 )𝑒
𝑠𝑖 𝑡

𝑛=0 𝑖

si : complex frequency = i + ji

i : rate of decay i : rate of oscillation
 Definition of Laplace Transform

Two-sided Laplace transform of x(t)

∞
𝑋 ( 𝑠 )= ∫ 𝑥 (𝑡 ) 𝑒
− 𝑠𝑡
𝑑𝑡
−∞

One-sided (unilateral) Laplace transform of causal signal x(t)

∞
𝑋 ( 𝑠 )=∫ 𝑥 (𝑡 ) 𝑒
− 𝑠𝑡
𝑑𝑡
0

Two-sided (bilateral) inverse Laplace transform of X(s)

𝑐 + 𝑗∞
1
𝑥 ( 𝑡 )= ∫
2 𝜋 𝑗 𝑐 − 𝑗∞
𝑋 (𝑠)𝑒
𝑠𝑡
𝑑𝑠
 Examples of Popular Functions

Find the Laplace transform of the impulse (t).

∞
ℒ [ 𝛿 ( 𝑡 ) ]= ∫ 𝛿 ( 𝑡 ) 𝑒
− 𝑠𝑡
𝑑𝑡 =1 For all s
−∞

Find the Laplace transform of the unit step u(t).

∞
ℒ [ 𝑢 ( 𝑡 ) ] =∫ 𝑢 ( 𝑡 ) 𝑒
− 𝑠𝑡
𝑑𝑡
0

|
∞
1 − 𝑠𝑡 1
¿− 𝑒 = For Re[s]
𝑠 0 𝑠 >0
 Examples of Popular Functions

Find the Laplace transform of .

∞
1
ℒ [ 𝑒 𝑢(𝑡 ) ]=∫ 𝑒 𝑒 For Re[s] > a
𝑎𝑡 𝑎𝑡 − 𝑠𝑡
𝑑𝑡 =
0 𝑠−𝑎 >a

Find the Laplace transform of .

∞
1
ℒ [ cos 𝜔 0 𝑡 𝑢 (𝑡 ) ] =∫ [ 𝑒 𝑗 𝜔 𝑡 +𝑒 − 𝑗 𝜔 ] 𝑒− 𝑠𝑡 𝑑𝑡
0 0 𝑡

0 2

𝑠 For Re[s] > 0

¿ 2 2
𝑠 +𝜔 0 >0
 Laplace Transform
∞
3
x(t) = 3e u(t)
-5t ∫ 𝑥 (𝑡 )𝑒 − 𝑠𝑡
𝑑𝑡 𝑋 ( 𝑠 )=
𝑠 +5
−∞

3
¿ 𝑋 ( 𝑠 ) ∨¿
√ ( 𝜎 +5 ) + 𝜔
2 2

ω=0
 = -4
z =3

ω
 Omega
Sigma
 6 𝑠+34
𝑋 ( 𝑠 )= 𝐻 ( 𝑠 )= 𝑌 (𝑠 )
𝑠 𝑠 2+10 𝑠+34

Inverse Laplace Transform

By Partial Fraction Expansion & the Table

𝑥 ( 𝑡 )=6 𝑢 (𝑡 ) h ( 𝑡 )=10 𝑒
−5 𝑡 o
cos ⁡( 3 𝑡 +12 7 ) 𝑢(𝑡 ) 𝑦 (𝑡 )
 Inverse Laplace
Transform
The inverse Laplace
transform, finding x(t)
from X(s), involves
integration in the
complex plane that is not
covered in this class.

Partial fractions
expansion, Laplace
properties, and the table
will be used to find the
inverse Laplace
transform.
 Example 1 of Inverse Laplace Transform

Use partial fraction expansion to find the inverse Laplace

transform of
7 𝑠+ 6
𝑋 ( 𝑠 )= 2
𝑠 +5 𝑠 +6

𝑥 ( 𝑡 )=[ − 8 𝑒 ] 𝑢 (𝑡 )
−2𝑡 −3 𝑡
Answer: +15 𝑒
 Example 2 of Inverse Laplace Transform

Use partial fraction expansion to find the inverse Laplace

transform of 2
( ) 2 𝑠 +5
𝐻 𝑠= 2
𝑠 +3 𝑠+ 2

Answer: h ( 𝑡 )=2 𝛿 ( 𝑡 )+ [ 7 𝑒− 𝑡 − 13 𝑒 −2 𝑡 ] 𝑢 (𝑡 )
 Example 3 of Inverse Laplace Transform

Use partial fraction expansion to find the inverse Laplace

transform of
6( 𝑠+ 34)
𝑋 ( 𝑠 )= 2
𝑠 ( 𝑠 +10 𝑠+ 34)

𝑘1 𝑘2 𝑘3
𝑋 ( 𝑠 )= + +
𝑠 𝑠 +5 − 𝑗 3 𝑠+5 + 𝑗 3
𝑘1 𝐴𝑠 + 𝐵
𝑋 ( 𝑠 )= + 2
𝑠 𝑠 +10 𝑠 +34

Answer: 𝑥 ( 𝑡 )=[ 6 +10 𝑒 −5 𝑡 cos ⁡( 3 𝑡 +126.9 o) ] 𝑢( 𝑡 )

 Example 4 of Inverse Laplace Transform

Use partial fraction expansion to find the inverse Laplace

transform of
8 𝑠+10
𝐻 ( 𝑠 )= 2
(𝑠 +1) ( 𝑠+ 2 )

𝑘1 𝑘2 𝑘3
𝐻 ( 𝑠 )= + +
(𝑠 +1) ( 𝑠 +2) ( 𝑠 +2)2

h ( 𝑡 )=[ 2 𝑒 ] 𝑢 (𝑡 )
−2 𝑡 −2 𝑡
Answer: −𝑡
− 2𝑒 +6 𝑡 𝑒
Properties of The Laplace
Transform
 Linearity Property

Laplace Transform
∞

𝑥 (𝑡 ) ∫ 𝑥 (𝑡 )𝑒− 𝑠𝑡 𝑑𝑡 𝑋 (𝑠 )
−∞

𝑥 (𝑡 ) 𝑋 (𝑠 )

𝑥 1 (𝑡) 𝑋 1 ( 𝑠 ) and 𝑥 2 (𝑡) 𝑋 2( 𝑠)

If

then
𝑎 𝑥 1( 𝑡 ) 𝑎 𝑋 1 (𝑠 )

𝑎 𝑥 1 ( 𝑡 )+ 𝑏 𝑥 2 (𝑡 ) 𝑎 𝑋 1 ( 𝑠 ) +𝑏 𝑋 2 ( 𝑠 )
 Time Shifting Property

𝑥 (𝑡 ) 𝑋 (𝑠 )

𝑥 ( 𝑡 − 𝑡0 ) 𝑋 (𝑠 )𝑒
−𝑡 0 𝑠

Time delaying x(t) by to is equivalent to multiplying X(s) by

(phase shifting).

For causal signal

𝑥 ( 𝑡 ) 𝑢(𝑡 ) 𝑋 (𝑠 )
−𝑡 0 𝑠
) 𝑋 (𝑠 )𝑒
 Example: Time Shifting Property

Find the Laplace transform of the signal x(t)

Answer:
1 − 𝑠 1 − 2𝑠 1 − 4 𝑠
𝑋 ( 𝑠 )= 𝑒 − 2𝑒 𝑒
𝑠
2
𝑠 𝑠
 Frequency Shifting Property

𝑥 (𝑡 ) 𝑋 (𝑠 )

𝑥 ( 𝑡 ) 𝑒𝑠 𝑜 𝑡
𝑋 (𝑠 − 𝑠𝑜 )

Multiplying the signal x(t) by is equivalent to frequency shifting

by so in the Laplace domain.
𝑥 (𝑡 )
𝜎0
𝑠𝑜
𝑡 𝑗𝜔 𝜔0 𝑗𝜔

𝜎 𝜎
Example: Frequency Shifting Property

Find the Laplace transform of

Answer:
𝑠+ 𝑎
𝑋 ( 𝑠 )= 2 2
( 𝑠+ 𝑎) +𝑏
 Time-Differentiation Property

𝑥 (𝑡 ) 𝑋 (𝑠 )
𝑑𝑥 −
𝑑𝑡 𝑠𝑋 ( 𝑠 ) − 𝑥 (0 )

2
𝑑 𝑥
𝑑𝑡
2 𝑠2 𝑋 ( 𝑠 ) − 𝑠𝑥 ( 0− ) − 𝑥(
˙ 0− )

3
𝑑 𝑥
𝑑𝑡
3 𝑠3 𝑋 ( 𝑠 ) − 𝑠 2 𝑥 ( 0− ) − 𝑠 𝑥
˙ ( 0− ) − 𝑥
¨ (0− )

𝑛 𝑛
𝑑 𝑥
𝑑𝑡
𝑛 𝑠 𝑋 ( 𝑠 )− ∑ 𝑠
𝑛 𝑛−𝑘 𝑘 −1
𝑥˙ −
(0 )
𝑘=1
𝑑
Frequency-differentiation property 𝑡𝑥 ( 𝑡 ) − 𝑋 (𝑠 )
𝑑𝑠
Example: Time Differentiation Property

Find the Laplace transform of the

signal x(t)

Answer:
1
𝑋 ( 𝑠 )= 2
( 1 −3 𝑒
−2𝑠
+2 𝑒
−3 𝑠
)
𝑠
Time-Integration Property

𝑥 (𝑡 ) 𝑋 (𝑠 )

∫ 𝑥 (𝜏 ) 𝑑𝜏 𝑋 (𝑠)
𝑠
0−
 Scaling Property

𝑥 (𝑡 ) 𝑋 (𝑠 )

𝑥 ( 𝑎𝑡 )
1
𝑎
𝑋 ( )
𝑠
𝑎
for a>0

Expansion in time domain result in compression in the s

(frequency) domain, and the inverse is true.
𝑥 (𝑡 ) 𝑥 ( 0.5 𝑡 )

-2 2
𝑡
-1 1 𝑡
𝑗𝜔 𝑗𝜔

𝜎 𝜎
 Convolution Property

Convolution in the time domain is equivalent to multiplication in

the Laplace (frequency) domain.

𝑥 (𝑡 ) 𝑋 (𝑠 ) h (𝑡 ) 𝐻 (𝑠 )

𝑥 ( 𝑡 ) ∗ h(𝑡 ) 𝑋 ( 𝑠 ) 𝐻 (𝑠)

Multiplication in the time domain is equivalent to multiplication in

the Laplace (frequency) domain.

𝑥 ( 𝑡 ) h (𝑡 )
1
[ 𝑋 ( 𝑠) ∗ 𝐻 ( 𝑠) ]
2𝜋 𝑗
Laplace and Time Domains analysis of
Systems
The convolution property is very useful in simplifying the system
analysis in the Laplace domain.

Impulse Response y(t)

x(t) h(t)

𝑦 ( 𝑡 )=𝑥 ( 𝑡 ) ∗ h(𝑡 ) Zero-state response. All

initial conditions are zeros

Transfer Function Y(s)

X(s) H(s)
𝑌 ( 𝑠)
𝑌 ( 𝑠 )= 𝑋 ( 𝑠 ) 𝐻 ( 𝑠 ) 𝐻 ( 𝑠 )=
𝑋 ( 𝑠)
 Example: Convolution Property

Find y(t) where

Answer:
Summary of
Laplace
Transform
Properties
 Using Laplace to Solve Differential
Equations and System Analysis
 Solving differential equation of third order and higher in the
time domain to find the output of the system y(t) is
challenging.
 Solving differentials of any order in the Laplace domain is very
easy since Laplace transform transfers differential equation
into algebraic equation that can be easily solved to find Y(s).
 If you set all initial conditions to zero then you will obtain only
the zero-state response of the output y(t).
𝑛 𝑚
𝑛 𝑚
𝑠𝑛 𝑌 ( 𝑠 ) − ∑ 𝑠 𝑛−𝑘 𝑦˙ 𝑘 − 1 ( 0− ) =𝑠𝑚 𝑋 ( 𝑠 ) − ∑ 𝑠 𝑚−𝑘 𝑥˙ 𝑘 −1 (0− )
𝑑 𝑦 𝑑 𝑥
𝑛
= 𝑚
𝑑𝑡 𝑑𝑡
𝑘=1 𝑘=1
𝑚
𝑛 𝑚 𝑠
𝑠 𝑌 ( 𝑠 )=𝑠 𝑋 ( 𝑠 ) 𝑌 ( 𝑠 )= 𝑛 𝑋 ( 𝑠 ) Initial Conditions
𝑠
 System Modeling
2
𝑑 𝑦 (𝑡 ) 𝑑𝑦 ( 𝑡 ) 𝑑𝑥 (𝑡 )
+5 +6 𝑦 ( 𝑡 ) = + 𝑥(𝑡 )
𝑑𝑡 2
𝑑𝑡 𝑑𝑡

Using Laplace to Solve Differential

Equations and System Analysis

𝑠 +1 ( 𝑠+5 ) 𝑦 ( 0 ) + 𝑦
˙ (0)
𝑌 ( 𝑠 )= 2
𝑋 ( 𝑠 ) + 2
𝑠 +5 𝑠+6 𝑠 +5 𝑠+ 6
Zero State Response Zero Input Response
 Example
Find the transfer function H(s) of an LTIC system described by
the equation
d 2 y (t ) dy (t ) dx(t )
2
5  6 y (t )   x(t )
dt dt dt
and the system response y(t) if the input x(t) = 3e-5tu(t) and all the
initial conditions are zero; that is the system is in the zero state
(relaxed).

Answer :
 5t  2t  3t
y (t ) ( 2e e  3e )u (t )
 Example: System Analysis in Laplace
Domain
Solve the differential equation to find y(t)
2
𝑑 𝑦 𝑑𝑦 ( ) 𝑑𝑥
+ 5 +6 𝑦 𝑡 = + 𝑥 (𝑡 )
𝑑𝑡
2
𝑑𝑡 𝑑𝑡

Given that and and the input

Make sure to show the zero-input response, the zero-state
response, and the transfer function of the system H(s).
Answer:
𝑦 ( 𝑡 )=( 7 𝑒− 2 𝑡 − 5 𝑒 −3 𝑡 ) 𝑢 (𝑡 ) + ( − 0.5 𝑒 −2 𝑡 + 2𝑒 − 3𝑡 − 1.5 𝑒 −4 𝑡 ) 𝑢 ( 𝑡 )

zero-input response zero-state response

Frequency Response of LTI System
H(s)
𝑠 +0.1
𝑥 ( 𝑡 )=cos ( 2𝑡 ) 𝑢 (𝑡 ) X(s) 𝑠 +5 Y(s) = H(s)X(s)

𝑠=𝜎 + 𝑗 𝜔

−5 𝑡
𝑦 ( 𝑡 )=0.84 𝑒 𝑢 ( 𝑡 ) + 0.37 cos ( 2 𝑡 + 65.3 ) 𝑢(𝑡 )

H( jω)
𝑗 𝜔+ 0.1
X( jω ) 𝑗 𝜔+ 5
Y( jω) =H( jω)X( jω)
 Frequency Response of a LTI System
Transfer Function 𝑠=𝜎 + 𝑗 𝜔 Frequency Response
2s If  = 0 2 𝑗𝜔
H (s)  𝐻 ( 𝑗 𝜔 )=
s 2  2s  5
2
− 𝜔 +2 𝑗 𝜔+5
2 √ 𝜎 2 +𝜔 2 2𝜔
|𝐻 ( 𝑠 )|= |𝐻 ( 𝜔 )|=
√ ( 𝜎 2 − 𝜔 2+ 2 𝜎 +5 ) +( 2 𝜎𝜔+ 2 𝜔 )2
2
√ 2
( 5 − 𝜔2 ) + 4 𝜔 2

|H(s)|
𝜎

𝜔 2.24
 Laplace and Frequency Response of a
LTI System
𝑥 𝑡 =∑ 𝑋 (𝑠𝑖 )𝑒 𝑦 𝑡 =∑ 𝐻(𝑠 𝑖 ) 𝑋 (𝑠 𝑖)𝑒
𝑠𝑡 𝑠𝑡
( ) 𝑖
( ) 𝑖

𝑖 H(s) 𝑖
X(s)
1
𝑠 +1 Y(s) 𝑌 ( 𝑠𝑖 )

𝑠=𝜎 + 𝑗 𝜔
If σ = 0 in the complex frequency s, so s = jω then system analysis
is simplified further, and it becomes frequency response.
𝑦 ( 𝑡 )=∑ 𝐻( 𝑗𝜔 𝑖 ) 𝑋 ( 𝑗 𝜔𝑖 )𝑒
𝑗𝜔 𝑖𝑡
𝑥 ( 𝑡 )=∑ 𝑋 ( 𝑗 𝜔𝑖 )𝑒
𝑗𝜔 𝑖 𝑡
H( jω) 𝑖
𝑖 1
X( jω ) 𝑗 𝜔+ 1
Y( jω)
𝑥 ( 𝑡 )= Acos ⁡( 𝜔 𝑜 𝑡 +45 𝑜 )
𝑦 ( 𝑡 ) =| 𝐻 ( 𝑗 𝜔 𝑜 )| Acos ¿ ¿

Amplitude Response Phase Response

 Example: Frequency Response
For the system described by the transfer function

Find the frequency response H(jω), and the system response

y(t) for input x(t) = cos2t and x(t) = cos(10t - 50 )
Solution

𝑗 𝜔+ 0.1
Replace s by jω 𝐻 ( 𝑗 𝜔)=
𝑗 𝜔+5

 2  0.01
Find the Magnitude H ( j ) 
 2  25
   1   
Find the Phase  ( j ) H ( j ) tan  1    tan  
 0.1  5
Magnitude and Phase of the Frequency
Response
𝑗 𝜔+ 0.1
𝐻 ( 𝑗 𝜔)=
𝑗 𝜔+5
 2  0.01
H ( j )     1   
  25
2  ( j ) H ( j ) tan  1    tan  
 0.1  5

x(t) = cos2t & x(t) = cos(10t - 50 )

 Example: Frequency Response

For input x(t)=cos2t, we have:

22  0.01  2  1  2 
H ( j 2)  0.372  ( j 2) tan  1    tan   65.3


22  25  0.1   5

Therefore y (t ) 0.372 cos(2t  65.3 )

 Example: Frequency Response

For input x(t)= cos(10t - 50), we will use the amplitude and
phase response curves directly:
H ( j10) 0.894
 ( j10) H ( j10) 26
Therefore
∘ ∘ ∘
𝑦 (𝑡)=0.894cos(¿10𝑡 −5 0 +26 )=0.894 cos(¿10𝑡 −2 4 )¿ ¿
 Frequency Response of Ideal Delay
x(t) Delay y(t)=x(t-T)
X(s) T Y(s)= X(s) e-sT
H(s) of an ideal T sec delay is:

The magnitude
The phase
An ideal delay system does not effect the
amplitude of the input but phase shift the
input with –T gradient.

The group delay is the time delay of all

frequencies
 Frequency Response of an Ideal
Differentiator
x(t) 𝑑 𝑑𝑥
𝑦 ( 𝑡 )= Y(s)= sX(s)
X(s) 𝑑𝑡 𝑑𝑡
H(s) of an ideal differentiator is:

The magnitude
The phase is /2
If x(t) = cosωt what is the output
d
(cos t )   sin t  cos(t   / 2)
dt
Differentiators amplify inputs with high
frequencies such as noise and for that
it is avoided in system design.
Frequency Response of an ideal Integrator
x(t)
X(s) ∫ 𝑥 (𝑡 ) 𝑑𝑡 𝑦 ( 𝑡 )=∫ 𝑥 ( 𝑡 ) 𝑑𝑡 X(s)

H(s) of an ideal integrator is:

The magnitude
The phase is -/2
If x(t) = cosωt what is the output

1 1
cos t dtsupress
Integrator t  with
 sininputs
 as noise
cos( t   / 2)
high
 and for that it
frequencies such
is used in system design.
 Internal Stability
• Internal Stability (Asymptotic)
– Stable if all the poles are in the LHP
– Unstable if one or both of the following
conditions exist:
• At least one pole is in the RHP
• There are repeated poles on the imaginary axis
– Marginally stable if there are no poles in the
RHP, and there are some unrepeated poles
on the imaginary axis.
 External Stability BIBO
The transfer function H(s) can only indicate the external
stability of the system BIBO.
b0 s M  b1s M  1  ...  bM
H ( s)  N
s  a1s N  1  ...  a N
BIBO stable if M  N and all poles are in the LHP

Example
Is the system below BIBO and asymptotically (internally) stable?

1 S1
x(t) y(t)
S1 S 1