DSP

Laplace Transform
The Laplace transform is a generalization of the
Fourier transform.
We need it for two reasons:
1. there are some functions of interest, such as the
ramp function r(t) = t U(t) ,
which do not have a Fourier
transform.
2. we wish to determine a systems response from a
specific time, t=0 , and also include any initial
conditions in the systems response.
Need of Laplace transform
Forward Laplace Transform
Decompose a signal f(t) into complex sinusoids of the
form e
s t
where s is complex: s =
Forward (bilateral) Laplace transform is defined as

f(t): complex-valued function of a real variable t
F(s): complex-valued function of a complex variable s
Bilateral means that the extent of f(t) can be infinite in
both the positive t and negative t direction (also known
as two-sided)
( ) ( )
}

= dt e t f s F
t s


When evaluated along the axis
(i.e., ), the Laplace transform
reduces to the Fourier transform .

e j s =
0 = o
( ) ) ( ) (
0
t d e t x w X
t j
}

=
e
The Laplace transform can therefore be viewed
as a generalization of the Fourier transform from
the imaginary line (a simple frequency axis) to the
complex plane.
Forward Laplace Transform contd..
One benefit of the more general Laplace transform is
the ability to transform signals which have no Fourier
transform. To see this, we can write the Laplace
transform as

We can thus interpret the Laplace transform as the
Fourier transform of an exponentially enveloped input
signal. For (the so-called ``right-half plane''
(RHP)), this exponential weighting forces the Fourier-
transformed signal toward zero as
( )
) ( ) (
) ( ) (
0
0
) (
t d e e t x
t d e t x s X
t j t
t j
}
}

+
=
=
e o
e o
0 > o
t
Forward Laplace Transform contd..
A function x(t) has a Laplace transform
whenever it is of exponential order. That is,
there must be a real number B such that

0 ) ( lim =

Bt
t
e t x
As an example, every exponential
function has a Laplace transform for all finite
values of and
t
Ae
o
A
o
Forward Laplace Transform
continued..
Laplace Transform Properties
Linearity
Time shifting
Frequency shifting
Differentiation in time
( ) ( ) ( ) ( ) s F a s F a t f a t f a
2 2 1 1 2 2 1 1
+ +
( ) ( ) ( ) 0
0 0 0
0
>

t e s F t t u t t f
st
( ) ( )
0
0
s s F e t f
t s

| |
( ) ( )
( ) ( ) ( )
( ) ( ) ( )
( )
( )

=
=
=
=
0 0 0 ) (
0 0 ) ( ' '
0 ) ( '
) 0 ( )) ( ( )} ( {
1 2 1
2
2
2
'
n n n n
n
n
n
f f s f s s F s
dt
f d
t f
f sf s F s
dt
f d
t f
f s sF
dt
df
t f
f t f L s t f L

Initial Value Theorem

=
+
0
| | ) 0 ( )) ( ( )} ( {
'
f t f L s t f L =
| |
| | ) 0 ( ) ( ) (
) 0 ( ) ( )} ( {
0
'
'
f s sF lt dt e t f lt
f s sF lt t f L lt
s
st
s
s s
=
=

}
| |
| |
( ) ( ) t f lt s F s lt
Theref ore
f s sF lt
dt e t f lt e i
zero becomes LHS the of tegration in the s as
t
s
s
st
s
+

=
=
=

}
0
0
'
,
0 ) 0 ( ) (
0 ) ( . .
,
Similarly the final values theorem states that if f(t) and its
derivative f(t) are both Laplace transformable, then

( ) ( ) s F s lt t f lt
s t 0
=
Laplace Transform Properties
Differentiation in frequency

Integration in time
E.g.. f(t) = o(t)





Integration in frequency
( ) ( ) s F
ds
d
t f t
( ) ( )
}

t
s F
s
d f
0
1
t t
( ) ( )
( )
( ) { } ( )
s
s F
s
d f L
s F
t u d f
t
t
1 1

1

0
0
= =
=
=
}
}

t t
t t
( ) ( )
}

s
ds s F t f
t

1
Laplace Transform Properties
Scaling in time/frequency
Under integration,





Convolution in time

( ) 0 for
1
>
|
.
|

\
|
a
a
s
F
a
at f
( ) ( )
( ) { }
( )
| |
1
| |
1
| |
1

| |
1

a
t
a
L
a
t a L
t
a
t a
=
)
`

=
=
o
o
o o
( ) ( ) ( ) ( ) s F s F t f t f
2 1 2 1
-
t
f(2 t)
1 -1
t
f(t)
2 -2
Area reduced
by factor 2
Time-varying and time-invariant property




Convolution in frequency
( ) ( )
) ( ) (
1 0 1
1 1
0
s F e t t f
s F t f
st
L

( ) ( ) ( ) ( ) | | s F s F
j
t f t f
2 1 2 1
2
1
-
t
Convergence of Laplace transform
The condition Re{s} > -Re{a} is the region of
convergence, which is the region of s for which
the Laplace transform integral converges
Re{s} = -Re{a} is not allowed (see next slide)

{ }
{ }

<
>
=

0 Re if
0 Re if 0
lim
s
s
e
st
t
( )
{ } { } { }
{ } { } { }

< < +
> > +
=
+
a s a s
a s a s
e
t a s
t Re Re 0 Re if
Re Re 0 Re if 0
lim
( ) { } { } a s
a s
s F Re Re for
1
>
+
=

Regions of Convergence
What happens to F(s) = 1/(s+a) at s = -a?
= (1/0) (
Unbounded..)
-e
a t
u(-t) and e
-a t
u(t) have same transform
function but different regions of convergence
( ) { } { } a s
a s
t u e
L at
Re Re if
1
>
+

Im{s}
Re{s} = -Re{a}
Re{s}
-1
t
f(t)
f(t) = -e
a t
u(-t)
anti-causal
1
t
f(t)
f(t) = e
-a t
u(t)
causal
e.g..
Review of 0
-
and 0
+

o(t) not defined at t = 0 but has unit area at t
= 0
0
-
refers to an infinitesimally small time
before 0
0
+
refers to an infinitesimally small time after
0
( ) ( ) ( ) 1 lim but 0
0
0
0
0
0
0 0
= = =
} } }
+

+

dt t dt t dt t o o o
c
c c
Unilateral Laplace Transform

Forward transform: lower limit of integration is 0
-

(i.e. just before 0) to avoid ambiguity that may
arise if f(t) contains an impulse at origin

Unilateral Laplace transform has no ambiguity in
inverse transforms because causal inverse is
always taken:
No need to specify a region of convergence
Disadvantage is that it cannot be used to analyze non
causal systems or non causal inputs
( ) ( )
}

=
0

dt e t f s F
t s
( ) t u e
a s
t a L

1

1

+


Inverse Laplace Transform
Many Laplace transform expressions are
ratios of two polynomials, also known as
rational functions
Convert complicated rational functions
into simpler forms
Apply partial fractions decomposition
Use lookup tables
Partial Fractions Method:

( )
( )
| | | |
( )
( ) ( ) ( ) ( ) t u t t u e e t f
j s j s
s F
B A B A B A
s
B j Bs Aj As
j s
B
j s
A
s F
j s
s
s
s F
t j t j
cos
2
1
2
1
2
1
2
1
2
1
0 1
: numerators two the Equating
: (poles) r denominato in Roots
0
0 0
2
0
2
0 0
0 0
0
2
0
2
0 0
e
e e
e
e e
e e
e
e
e e
=
|
.
|

\
|
+ =
+
+

=
= = = = +
+
+ +
=
+
+

=
=
+
=

e.g..#1
Partial Fractions Method
( )
( )( )
( ) ( ) | |
( )
( ) ( ) | |
( )
( )
( ) ( ) ( ) t u e e
s s
L t f
s s
s F
s F s k
s F s k
s
k
s
k
s s
s
s F
t t
s
s
3 4
3
3
2
4
3
3
2
4
3
2 3
6 21
3
4
3 2
6 14
2
3 2
3 2
6 7
3 2 1
3 2
2 1
2 1
+ =
)
`

+
+
=

+
+
=
=
+

= =
=


= + =

+
+
=
+

=

=
=
e.g..#2
Partial Fractions Method
( )
( )( )
( )
: division polynomial Performing After
fraction. improper an is
7
11
like fraction improper an is
2 1
5 2
2 3
5 2
2
2
2
s F
s s
s
s s
s
s F
+ +
+
=
+ +
+
=
( )
( )( )
( )
( )( ) ( )
( )
( )( ) ( )
( )
( ) ( ) ( ) ( ) t u e e t t f
s s
s F
s s
s
s k
s s
s
s k
s
k
s
k
s s
s
s F
t t
s
s
13 7 2
2
13
1
7
2
13
1 2
1 12
2 1
1 6
2
7
2 1
1 6
2 1
1 6
1
2 1
2
2 1
1 6
2
2
2
2
1
1
2 1

=
=
+ =
+

+
+ =
=
+
+
=
(

+ +
+
+ =
=
+
+
=
(

+ +
+
+ =
+
+
+
+ =
+ +
+
+ =
o
e.g..#3
1 6
4 6 2
2
5 0 2 2 3
2
2 2
+
+ +
+ + + +
s
s s
s s s s
E.g..#4
Compute y(t) = e
a t
u(t) * e
b t
u(t) , where
a = b
( ) ( ) { }
( ) ( )
( ) ( ) ( ) t u e e
b a
t y
b s a s b a
b s a s
t u e t u e L
t b t a
t b t a

1
1 1 1
1 1

=
(

= -
Partial Fractions Method
Partial Fractions Method
Given:
) 10 ( ) 4 ( ) 1 ( ) 10 )( 4 )( 1 (
) 2 ( 4
) (
3 2 1
+
+
+
+
+
=
+ + +
+
=
s
A
s
A
s
A
s s s
s
s F
27 4
) 10 )( 4 )( 1 (
) 2 ( 4 ) 1 (
|
1
1
=
+ + +
+ +
=
= s
s s s
s s
A 9 4
) 10 )( 4 )( 1 (
) 2 ( 4 ) 4 (
|
4
2
=
+ + +
+ +
=
= s
s s s
s s
A
27 16
) 10 )( 4 )( 1 (
) 2 ( 4 ) 10 (
|
10
3
=
+ + +
+ +
=
= s
s s s
s s
A
) ( ) 27 16 ( ) 9 4 ( ) 27 4 ( ) (
10 4
t u e e e t f
t t t
+ + =
Find A
1
, A
2
, A
3

E.g..#5
Key Transform Pairs
( ) { } ( ) 1
0 0
= = =
}

t
st st
e dt e t t L o o
( ) { } ( )
{ } 0 Re for
1
1 1
lim
1
0
0
0
> =
+ =
(

=
=
=

}
}
s
s
s
e
s
e
s
dt e
dt e t u t u L
st
t
st
st
st
( ) ( )
a s
t u e
s
t u e t u
at t
+
=

1
) ( since
1
0



Key Transform Pairs
( ) ( ) { } | | ( )
( ) { } { }
{ }
rad/s. in is and Hz, in is So, . 2 Here,
0 Re for
2
1

2
1
2
1

2
1
) (
2
1

2
1

2
1
2 cos
0 0 0 0
2
0
2
0 0
2 2
2 2
0
0 0
0 0
e t e
e
t t
t
t t
t t
f f
s
s
s
f j s f j s
t u e L t u e L
t u e e L t u t f L
t f j t f j
t f j t f j
=
>
+
=
+
+

=
+ =
)
`

+ =

Relation to the z Transform

The Laplace transform is used to analyze
continuous-time systems. Its discrete-time
counterpart is the Z- transform:

If we define , the Z- transform
becomes proportional to the Laplace transform
of a sampled continuous-time signal:

=
0
) ( ) (
n
n
z nT x z X
sT
e z =

=
0
) ( ) (
n
nTs sT
e nT x e X
As the sampling interval T goes to zero,
we have
) (
) (
) ( lim ) ( lim
0
0
0 0
s X
dt e t x
e t x e X
st
n
nTs
t
sT
T
=
=
=
}

A
Relation to the z Transform continued..

The Laplace Transform Pair
( ) ( )
( ) ( )
}
}

=
=
i
i
st
st
ds e s H
i
t h
dt e t h s H
t 2
1
0
Example 1: decaying exponential
( )
( )
( )
( )
( ) o
o
o
o
o
o
>
+
=
+

=
=
=

}
s
s
e
s
dt e s H
e t h
t s
t s
t
Re ;
1
1
0
0
Re(s)
Im(s)
x
-o
t 0
Transform:
The Laplace Transform Pair
Inverse Transform:
( )
t
t
t
i
i
st
e
d i
i
e
d i e
i
ds
s
e
i
t h
o
t
t
o
t
t
o
u
t
u
t
o t

=
=
=
+
=
}
}
}
2
1
2
1
1
2
1
1
Re(s)
Im(s)
x
-o
u

o
o
u
u
u
d e i ds
e
s
e s
i
i
i
=
=
+
+ =
1
1

u
Cauchy Principal Value Theorem
If F(s) is analytic then ( ) ( ) a iF ds
a s
s F t 2
1
=

}
Example 1 continued, decaying exponential
The Laplace Transform Pair
Example 2: Damped sinusoid
( ) ( )
( )
( ) ( )
( )
( ) ( ) o
e o e o
e
e o e o
e e o
o
>
|
.
|

\
|
+ +
+
+
=
+ =
+ =
=
+

s
i s i s
s H
e e
e e e
t e t h
t i t i
t i t i t
t
Re ;
1 1
2
1
2
1
2
1
cos
Re(s)
Im(s)
x
-o
x
-e
e
t
0
Laplace Transform Pairs
h(t)
H(s)
unit step
0 t
s
1
(like lim
o0
e
-ot
)

t
e
o
o + s
1
( ) t e
t
e
o
cos
|
.
|

\
|
+ +
+
+ e o e o i s i s
1 1
2
1
( ) t e
t
e
o
sin

|
.
|

\
|
+ +

+ e o e o i s i s i
1 1
2
1
delayed step
0 t T
s
e
sT
unit pulse
0 t T
s
e
sT
1
1
Laplace Transform Properties (1)
( ) ( ) { } ( ) ( ) s G s H t g t h L | o | o + = +
( ) { } ( ) s H e T t h L
sT
= +
( ) { } 1 = t L o
( ) ( ) ( ) ( ) s G s H t d t t g t h L
t
=
)
`

' ' '

' ' '

}
'
0
Linearity
Time-shift
Dirac delta function transform
(sifting property)
Convolution
Impulse response
Frequency response

EXERCISE

1) Use Matlab to compute the Laplace
transform of the following functions
sin(2t), exp(3t)cos(t), and t^5.

2) Then use Matlab to compute the
inverse Laplace transform of the three
results you found.