The Laplace Transform in Signal Analysis

There are two kind of systems

1. static system: y(t) = ax(t) => easy (simple processing)
2. Dynamic system:

) ( ...
) (
) ( ...
) ( ) (
) (
0
1
) 1 (
1
) (
t x b
dt
t dx
b t y a
dt
t dy
a
dt
t dy
m
m
m
n
n
n
n
n
+ + + + +

Can we determine y(t) for given u(t) easily

!asier sol"tion method
)) ( ( ) (
) ( ) ( ) (
1
f Y F t y
f X f H f Y

(1) #e have systematic way to o$tain H(f) $ased on the differential e%"ation
(2) #e can o$tain X(f)
Fourier transform: an easier way
& 'ro$lem
(o"rier transform of the inp"t signal:

1 ) )
) ( ) (
2
2

ft j
ft j
e
dt e t x f X

if x(t) does not go to *ero when

t
and
t
X(f) typically does not e+ist, (the e+istence of X(f) if not g"aranteed)
- very strong condition. can not $e satisfied $y many signals,
Why Laplace Transforms?
Dynamic
/ystem
X(f) Y(f )
H(f )
-lge$raic e%"ation. not
differential e%"ation
1) Converts differential e%"ations to alge$raic e%"ations0 facilitates com$ination of m"ltiple
components in a system to get the total dynamic $ehavior (thro"gh addition 1 m"ltiplication)
2) Can gain insight from the sol"tion in the transform domain (2s2)0 inversion of transform not
necessarily re%"ired
3) -llows development of an analytical model which permits "se of a discontin"o"s (piecewise
contin"o"s) forcing f"nction and the "se of an integral term in the forcing f"nction
(important for control)

Definition: 4aplace transform of x(t)

+
0
) ( ) ( ) ( )5 ( 6 j s dt e t x s X t x L
st
#hat is a 4aplace transform of x(t)
- time f"nction 7o. t has $een eliminated $y the integral with respect to t,
- f"nction of s ( s is comple+ varia$le)
The inverse transform

j
j
st
ds e s X
j
t x

) (
2
1
) (
PROBLEMS
(1) x(t)81 and step f"nction x(t)8u(t)

0 (9e(s))
1
)5 ( 6 ) 1 (
1
) 0 sin 0 (cos
1

) 0 ( . ) 0 ( . 0 1 ) ()

) (
1
) ( )5 ( ) ( 6
0
0
0 0
0 0 0
>

< >
+

s
t u L L
s
j
s
if e if e e
e
s
e
e
s
e
e
s
e
s
e
st d e
s
dt e dt e t x t u t x L
j
j j
t
t
t j
t
t
t
st
t
t
st st st

(2)
) ( ) ( t u e t x
t

0
) (
0
)5 ( 6 dt e dt e e t u e L
t s st t t
Define a new comple+ varia$le
+ s s
:

0
:
dt e
t s
we know
0 ) 9e(
1
0
>

s
s
dt e
st

+

> > +
+

> +
+

>

s
e L
s or s
s
t u e L
s
s
dt e
s
s
dt e
t
t
t s
t s
1
5 6
) 9e( ) 9e( 0 ) 9e(
1
)5 ( 6
0 ) 9e(
1
0 ) 9e(
1
0
) (
:
:
0
:
(3)
) ( ) ( t t x
1
) sin (cos
) ( )5 ( 6
0
0 0
0

t
t
t
t j t
t
st
st
t j t e
e e e
dt e t t L

7o constraint on s.
Convergence of the Laplace Transform
(1) To ass"re

0 0
) ( ) ( dt e e t x dt e t x
t j t st
converge.
) 9e(s
m"st $e positive
eno"gh s"ch that
t
e t x

) (
goes to *ero when t goes to positive infinite
(2) 9egion of a$sol"te convergence and pole
PROPRT!S
"# Linearity
-ss"me
) ( ) ( ) (
2 2 1 1
t x a t x a t x +
(
1
a
and
2
a
are time independent)

)5 ( 6 ) ( )5. ( 6 ) (
2 2 1 1
t x L s X t x L s X
then
) ( ) ( )5 ( 6 ) (
2 2 1 1
s X a s X a t x L s X +
$ample
(1) (ind
) (cos
0
t L
e+press
) (cos
0
t
as linear com$ination of
) (t
.
) (t u
.
and;or
t
e

:

s
e L
s
t u L
t L
t
1
5 6
1
)5 ( 6
1 )5 ( 6
let
0
j
0
1
5 6
0

j s
e L
t j
+

let
0
j
0
) (
1
5 6 5 6
0 0

j s
e L e L
t j t j

Can we "se
t j
e
0

and
t j
e
0

to e+press
) cos(
0
t

) sin( ) cos(
) sin( ) cos(
) sin( ) cos(
0 0
0 0
0 0
0
0
t j t e
t j t
t j t e
t j
t j

+

+

2
0
2
0 0
0 0
0 0
0
0
0

) )( (
) ( ) (
2
1

1 1
2
1

)5 ( ) ( 6
2
1
)5 6cos(
2
) cos(
) cos( 2
0 0
0 0
0 0

1
]
1

+
+ +

1
]
1

+
+

+
+

+

s
s
j s j s
j s j s
j s j s
e L e L t L
e e
t
t e e
t j t j
t j t j
t j t j
<omw work : (ind
5 6sin
0
t L
%# Transforms of Derivatives
-ss"me
)5 ( 6 ) ( t x L s X
Then
) 0 ( ) (
) (


,
_

x s sX
dt
t dx
L
&# Laplace Transform of an integral
-ss"me
)5 ( 6 ) ( . ) ( ) ( t x L s X d x t y
t

Then
s
y
s
s X
d x L
t
) 0 ( ) (
) (


+
1
]
1

where

0
) ( ) 0 ( d x y
'# Comple$ Fre(uency shift )s*shift+ Theorem
-ss"me
t
e t x t y

) ( ) (
)5 ( 6 ) ( )5 ( 6 ) ( t y L s Y t x L s X
Then
) ( ) ( + s X s Y

+


s
e t u L
s
t u L
t
1
5 ) ( 6 .
1
)5 ( 6

2
0
2
0
0
2
0
2
0
5 6sin 5 6cos

s
t L
s
s
t L
8=
2
0
2
0
0
2
0
2
0
) (
5 6sin
) (
5 6cos

+ +

+ +
+

s
e t L
s
s
e t L
t t
$ample ,*' Fin-
13 >
?
)5 ( 6 ) (
2
1 1
+ +
+


s s
s
L s X L t x
/ol"tion:
2 2 2 2
2 2
2 ) 3 (
2 ) 2 ; @ (
2 ) 3 (
3
A B >
@ ) 3 (
13 >
?
) (
+ +

+
+ +
+

+ + +
+ +

+ +
+

s s
s
s s
s
s s
s
s X
) 0 ( 2 sin
2
@
2 cos )5 ( 6 ) (
3 3 1
> +

t t e t e s X L t x
t t
,# !nitial .alue Theorem
) ( lim ) 0 (
)5 ( 6 ) (
1
s sX x
s X L t x
s
+

THE INVERSE LAPLACE TRANSFORM

Pro/lem Statement: Civen a 4aplace transform ( ) { }
f t L
. find the inverse. that is.
find ( ) f t
. Do" may "se only the following known 4aplace transforms:
1
,
E F
n
n
n
t
s
+
L . 0 s > . n a nonnegative integerG
1
E F
at
e
s a

L . s a > G
2 2
Esin F
a
at
s a

+
L . 0 s > G
2 2
Ecos F
s
at
s a

+
L . 0 s > G
(1)
1
1 1
, ,
E F
n n
n n
n n
t t
s s

+ +


' )

L L
(2)
1
1 1
E F
at at
e e
s a s a

' )


L L
(3)
1
2 2 2 2
Esin F sin
a a
at at
s a s a

' )
+ +

L L
(A)
1
2 2 2 2
Ecos F cos
s s
at at
s a s a

' )
+ +

L L
Example 10 Fin- the inverse Laplace transform for
2
3 H
2 @
s
s s

+
#
First method
Since the denominator is an irreducible quadratic com!lete the square"

( )
2
2
3 H 3 H
2 @ 1 A
s s
s s s

+ +

Since the numerator contains s and the denominator contains 1 s introduce the
term 1 s in the numerator"
( )
( )
2
3 1 A
1 A
s
s

+

Se!arate the fraction into t#o fractions"
( ) ( )
2 2
1 2
3 2
1 A 1 A
s
s s

+ +
( ) ( ) 3 1 2 1 F s $ s .
where ( ) { }
2
cos 2
A
s
F s t
s

+
L
and ( ) { }
2
2
sin2
A
$ s t
s

+
L
.
%sin& (') (() and ())
3 E cos 2 F 2 E sin2 F
t t
e t e t L L
%sin& (*)
{ }
3 cos 2 2 sin2
t t
e t e t L
+y the basic idea the solution is
1
2
3 H
3 cos 2 2 sin2
2 @
t t
s
e t e t
s s

' )
+

L
Se!o"d method
Since the denominator is an irreducible quadratic com!lete the square"
( )
1 1
2
2
3 H 3 H
2 @ 1 A
s s
s s s

' ) ' )
+ +


L L
Since the numerator contains s and the denominator contains 1 s introduce the
term 1 s in the numerator"
( )
( )
1
2
3 1 A
1 A
s
s

' )
+


L
Se!arate the fraction into t#o fractions"
( ) ( )
1
2 2
1 2
3 2
1 A 1 A
s
s s

' )
+ +


L
( ) ( ) { }
1
3 1 2 1 F s $ s

L .
where ( ) { }
2
cos 2
A
s
F s t
s

+
L
and ( ) { }
2
2
sin2
A
$ s t
s

+
L
.
%sin& (*)
( ) { } ( ) { }
1 1
3 1 2 1 F s $ s

L L
%sin& (')
( ) { } ( ) { }
1 1
3 2
t t
e F s e $ s

L L
%sin& (() and ())
3 cos 2 2 sin2
t t
e t e t
So the ans#er is
1
2
3 H
3 cos 2 2 sin2
2 @
t t
s
e t e t
s s

' )
+

L
,xam!le -: !val"ate
2
01
3 2
2 B
2 3
s s
s s s
+
' )
+

L
.
Factor the denominator com!letely"
( ) ( ) ( )
2 2 2
3 2 2 1
2 B 2 B 2 B
2 3 2 3 1 3
s s s s s s
s s s s s s s s s
+ + +

+ + +
.erform !artial fractions decom!osition on the fraction"
( ) ( )
2
2 B
1 3 1 3
s s / + 0
s s s s s s
+
+ +
+ +
( ) ( ) ( ) ( )
2
2 B 1 3 3 1 s s / s s +s s 0s s + + + + +
( ) ( ) ( )
2 2 2
2 3 3 / s s + s s 0 s s + + + +
( ) ( )
2
2 3 3 / + 0 s / + 0 s / + + + +
Settin& corres!ondin& coefficients equal
2 / + 0 + +
2 3 1 / + 0 +
3 B /
(rom the last e%"ation. we have 3 / .
/et 1 s :

( ) ( ) ( ) ( )
2 3
2 1 1 B 1 A
2
+ + +
/et 3 s :
( ) ( ) ( ) ( )
2 1
2 3 3 B 3 A
2
0 0 +
/o
( ) ( )
2
3 1
2 B 3
2 2
1 3 1 3
s s
s s s s s s
+

+ +
.
1eturnin& to the !roblem"
( ) ( )
2 2
3 2
2 B 2 B
2 3 1 3
s s s s
s s s s s s
+ +

+ +
3 1
3
2 2
1 3 s s s

+
1 3 1 1 1
3
2 1 2 3 s s s
_ _ _


+
, , ,
%sin& (2) and (-)
{ } { } { }
3
3 1
3 1
2 2
t t
e e

L L L
%sin& (*)
3
3 1
3
2 2
t t
e e

' )

=L
+y the basic idea the solution is
2
01 3
3 2
2 B 3 1
3
2 3 2 2
t t
s s
e e
s s s

+

' )
+

L
Se!o"d method
Factor the denominator com!letely"
( ) ( ) ( )
2 2 2
01 01 01
3 2 2 1
2 B 2 B 2 B
2 3 2 3 1 3
s s s s s s
s s s s s s s s s

+ + +

' ) ' ) ' )
+ + +


L L L
.erform !artial fractions decom!osition on the fraction"
( ) ( )
2
2 B
1 3 1 3
s s / + 0
s s s s s s
+
+ +
+ +
( ) ( ) ( ) ( )
2
2 B 1 3 3 1 s s / s s +s s 0s s + + + + +
( ) ( ) ( )
2 2 2
2 3 3 / s s + s s 0 s s + + + +
( ) ( )
2
2 3 3 / + 0 s / + 0 s / + + + +
Settin& corres!ondin& coefficients equal
2 / + 0 + +
2 3 1 / + 0 +
3 B /
(rom the last e%"ation. we have 3 / .
/et 1 s :
( ) ( ) ( ) ( )
2 3
2 1 1 B 1 A
2
+ + +
/et 3 s :
( ) ( ) ( ) ( )
2 1
2 3 3 B 3 A
2
0 0 +
/o
( ) ( )
2
3 1
2 B 3
2 2
1 3 1 3
s s
s s s s s s
+

+ +
1eturnin& to the !roblem"
( ) ( )
2 2
01 01
3 2
2 B 2 B
2 3 1 3
s s s s
s s s s s s

+ +

' ) ' )
+ +


L L
01
3 1
3
2 2
1 3 s s s



' )
+


L
%sin& (*)
01 01 01
1 3 1 1 1
3
2 1 2 3 s s s


' ) ' ) ' )
+

L L L
%sin& (2) and (-)
8
3
3 1
3
2 2
t t
e e

So the ans#er is
2
01 3
3 2
2 B 3 1
3
2 3 2 2
t t
s s
e e
s s s

+

' )
+

L
Assignments
1.
2
3 2 1
2
) 2 ( 2 ?
) ? ( ) 2 (
10
) (
+
+
+
+
+

+ +

s
/
s
/
s
/
s s
s
s Y
2.
3
A
2
3 2 1
3
) 2 ( ) 2 ( 2 ?
) ? ( ) 2 (
10
) (
+
+
+
+
+
+
+

+ +

s
/
s
/
s
/
s
/
s s
s
s Y
3. Comple+ 0 ConI"gate (actors

j s
/
j s
/
s
/
s s
s s
s s s
s s
s Y
+
+

+
+

+ + +
+ +

+ + +
+ +

1 1 2 5 1 ) 1 )6( 2 (
> > 2
) 2 2 )( 2 (
> > 2
) (
3 2 1
2
2
2
2
A. 9epeated J"adratic (actors

2 2
2 2
2
1 1 1
2 2
2 3 A
) 1 ( 1 2
) 1 )( 2 (
1@ H 12 @
) (
+
+
+
+
+
+
+

+ +
+ + + +

s
c s +
s
c s +
s
/
s s
s s s s
s Y