Mathematical Foundation

& System Modelling

Mohamad Abdul Hady

Outline
• Signal & System Analysis
• LTI System
• Convolution Integral
• Laplace Transform
Signal & System Analysis
Analysis

Signal System

Continuous Discrete Continuous Discrete

Continuous Discrete Fourier Use Laplace

Use Z-Transform
Fourier Series Series Transform

Continuous
Discrete Fourier
Fourier
Transform
Transform
Linear Time Invariant (LTI) System
• Linearity: follows the properties of superposition
x1 (t ) + x2 (t ) → y1 (t ) + y2 (t )
ax1 (t ) → ay1 (t )
then
ax1 (t ) + bx2 (t ) → ay1 (t ) + by2 (t )
• Time Invariant: time shifting in the input signal
causes time shifting in the output signal
x(t ) → y (t ) 
 Time invariant
x(t + a) → y (t + a)
Convolution Integral

x(t ) h(t ) y (t )

y (t ) = x(t )  h(t )

y (t ) =  x( )h(t −  )d
−
Or, by commutative properties:

y (t ) =  h( ) x(t −  )d
−
 Example of Integral Convolution
• Given the system impulse response is: h(t) = e-at, if the
input signal is unit step, find the system output (y(t))!
x(t ) h(t )
y (t ) = ...?
t t

• Solution:
1 , t  0 y (t ) = x(t )  h(t )
x(t ) = u (t ) =  
0 , t  0
y (t ) =  x( )h(t −  )d
−
− at
h(t ) = e
 Solution
• Graphical Representation: y(t) for t > 0 :
x( ) h(t −  ) = h(− + t ) t


y (t ) = x(t )h(t −  )d
0
t
 
t 
= 1  e −(t − ) d
0
• Overlapping area: t
  =t
 e d = e
−t  −t
t < 0 y(t)=0 t>0 =e [e ]
 =0
0
x( ), h(t −  ) x( ), h(t −  ) = e − t (e t − e 0 ) = 1 − e − t
y (t )
 
t t
No overlapping t
Laplace Transform

X ( s ) =  x(t )e − st dt Laplace transform
−

1  + j
x(t ) = 
st
X ( s ) e ds Inverse Laplace transform
2j  − j

L
s =  + jw
x(t )  X ( s )
damping Complex frequency

• Used to understand the frequency characteristics

of a System
• Used to solve ODEs because of their convenient
calculus and convolution properties (today)
• Equivalent to the Fourier transform when s=jw
 Similar with Other Transformation

x(t ) ⎯⎯→ X ( s ) =  x(t )e − st dt
L
Laplace :
−


x(t ) ⎯⎯→ X ( jw ) =  x(t )e − jwt dt
F
Fourier – Continuous :
−


Fourier – Discrete : x[n] ⎯⎯→
F
X ( j ) =  x[ n
n = −
]e − jn


Z Transform : x[n] ⎯
⎯→ X ( z ) =
Z

n = −
x[ n ] z −n
Laplace Transform Theorems
Laplace
Transform
Table
Properties Proof
• Time Shifting

f (t − a) ⎯⎯→ F ( s ) =  f (t − a)e − st dt
L
−
misal p = t − a  t = p + a

= −
f ( p)e − s ( p + a ) dp a = const

= f ( p)e − sp e − sa dp
−

= e − as F ( s )
Properties Proof
• Time Scaling
1 s
f (at ) ⎯⎯→ F2 ( s ) = F ( )
L
a a
misal :
−t 1
f1 (t ) = e u (t ) → F1 ( s ) =
s +1
1 1 1
f 2 (t ) = e − 4t u (t ) → F2 ( s ) = =
4 s / 4 +1 s + 4
Properties Proof

• Convolution Y (s) = 
−
y (t )e − st dt


= 
−
x( )h(t −  )e − st dt d


= 
−

x( ) h(t −  )e − st dt d


= 
−
x( )e −s d H ( s )

= X (s) H (s)
Properties Proof
• Final Value Theorem:

df (t ) − st
lim 
s →0 dt
e dt = lim sF (s) − f (0)
s →0
0

f (t ) 0 =
lim sF (s) − f (0)
s →0

f () − f (0) = lim sF ( s ) − f (0)

s →0

f () = lim sF ( s )
s →0

• Initial value Theorem: f (0) = lim sF (s)

s →
Example: Application for RC Circuit
Modelling
According to the RC Circuit below, find the Transfer
Function! (all initial condition is zero)
deo (t )
RC + eo (t ) = ei (t )
dt
RCsEo ( s ) + Eo ( s ) = Ei ( s )
( RCs + 1) Eo ( s ) = Ei ( s )
Eo ( s ) 1
=
Ei ( s ) ( RCs + 1)
Eo ( s ) 1
=
Ei ( s ) ( RCs + 1)
Invers Laplace Transform
by Partial Fraction
 Cases on Partial Fraction Expansion
There are three cases to consider in doing the partial fraction expansion of F(s)

Case 1: F(s) has simple poles:

Q( s) K1 K2 Kn
F (s) = = + + ... +
( s + p )( s + p )...( s + p ) s + p s + p s+ p
1 2 n 1 2 n

Case 2: F(s) has repeated poles:

Q( s)
F (s) =
( s + s1 )( s + s2 )...( s + sn − r )( s + si ) r
K s1 K K s ( n−r ) A A2 Ar
= + s 2 + ... + + 1 + + ... +
s + s1 s + s2 ( s + sn − r ) s + si ( s + si ) 2 ( s + si ) r

Case 3: F(s) has complex-conjugate poles:

Q( s) K1 K2
F (s) = = +
( s +  − jw )( s +  + jw ) s +  − jw s +  + jw
 Example for Case 1 : Simple Poles

Given:
4( s + 2) K1 K2 K3
F (s) = = + +
( s + 1)( s + 4)( s + 10) ( s + 1) ( s + 4) ( s + 10)

Find K1, K2, K3 :

( s + 1)4( s + 2) ( s + 4)4( s + 2)
K1 = | = 4 27 K2 = =49
( s + 1)( s + 4)( s + 10) s = −1 ( s + 1)( s + 4)( s + 10)
|s =−4
( s + 10)4( s + 2)
K3 = | = − 16 27
( s + 1)( s + 4)( s + 10) s = −10

Find Invers Laplace Transform:

f (t ) = (4 27)e −t + (4 9)e − 4t + (−16 27)e −10t u (t )

 
Coefficient Evaluation Method of Case 2 :
Repeated Poles
Q( s)
F (s) =
( s + s1 )( s + s2 )...( s + sn − r )( s + si ) r
K s1 K K s ( n−r ) A A2 Ar
= + s 2 + ... + + 1 + + ... +
s + s1 s + s2 ( s + sn − r ) s + si ( s + si ) 2 ( s + si ) r

When we have repeated roots we find the coefficients of the terms as follows:


Ar = ( s + s ) F ( s)
i
r  s = −s
i

Ar −1 =
d
ds

( s + s ) r F ( s)
i
s = −s

i

Ar − 2 =
1 d2
2! ds 2

( s + s ) r F ( s)
i  ... A1 =
1 d r −1
(r −1)! ds r −1

( s + s ) r F ( s)
i 
s = −s s = −s
i i
Example for Case 2 : Repeated Poles
Let the System Transfer Function is:
( s + 1) K
1
A
1
A
2
F (s) = = + +
s ( s + 3) 2 s ( s + 3) ( s + 3) 2

K =
1

A =
1

A =
2

 
f (t ) = ___?___  + ___?___ e −3t + ____?____te −3t u (t )
Coefficient Evaluation Method of Case 3
: Complex Conjugate Poles
Q( s) K1 K2
F (s) = = +
( s +  − jw )( s +  + jw ) s +  − jw s +  + jw

K1 & K2 is given by,

K1 = ( s +  − j w ) F ( s ) s = −  + j w
K 2 = ( s +  + j w ) F ( s ) s = − − jw
Example for Case 3 (1)
10
Given: F (s) = 2
s + 4s + 5
Modified to :
10 K1 K2
F (s) = = +
( s + 2 − j )( s + 2 + j ) s + 2 − j s + 2 + j
Find the Coefficients:

K1 = ( s + 2 − j ) F ( s ) s = −2+ jw = 5
j
K 2 = ( s + 2 + j ) F ( s ) s = −2− jw =− 5
j
Example for Case 3 (2)
10
F (s) =
( s + 2 − j )( s + 2 + j )
5 1 1 
=  −
j s + 2− j s + 2+ j 
L-1

5 − 2t jt
f (t ) = e (e − e − jt ) t  0
j

f (t ) = 10e −2t (sin t ) t  0