Chapter

Laplace Transform
4
Introduction
In the next chapter we will study Fourier transform X() of a time domain signal x(t) where we can get
information of frequency component of signal x(t). Before studying Fourier transform we will study Laplace
transform to make understanding of fourier transform simpler.
Lets see what is Fourier transform of signal x(t).

X() =  x(t ) e jt dt ...(1)


The Fourier transform X() is valid only when

• Signal is absolutely integrable.
• Signal has finite number of discontinuities in a finite interval of time and each discontinuity should
be finite.
• Signal should have finite number of maxima and minima in a finite interval of time.
These conditions are Dirichlet conditions (these conditions are sufficient but not necessary). So if a
signal is absolutely integrable then we can do Fourier analysis for getting it’s frequency analysis. If signal is not
absolutely integrable then we use Laplace transform to get frequency analysis. Examples are eat u(t), a > 0; e–
at, – < t < ; tu(t); and other signals that are not absolutely integrable. The Fourier transform does not exist

for signals that are not absolutely integrable, so the Fourier transform based methods cannot be employed in this
class of problems. The difficulty could be resolved by generalizing the Fourier transform so that the signal x(t) is
expressed as a sum of complex exponential est, where s =  + j and thus is not restricted to the imaginary axis
only. This is equivalent to multiplying the signal by an exponential convergent factor. For example, e–t eat u(t)
satisfies Dirichlet’s conditions for  > .
Another importance of Laplace transform is: generally a linear time invariant (LTI) system is described
by differential equations. The response of a system for a given input is obtained by solving the differential
equations relating its output and input signals. We know that the solution of higher order differential equations
is quite tedious and time consuming compared to the solution of algebraic equations. So the Laplace transform
is used to solve the differential equations. Laplace transform is a powerful mathematical tools used to convert
the differential equations into algebraic equations. It is a simple and systematic method which provides the
complete solution in one stroke by taking into account the initial conditions in a natural way at the beginning of
the process itself. For solving the differential equations using Laplace transform, we take the Laplace transform
of the differential equations (i.e. convert the differential equations in time domain into algebraic equations in

GATE MASTERS PUBLICATION

 laplace Transform 145

frequency domain), insert the initial conditions, solve the resultant algebraic equations (i.e. get the solution in
s-domain) and take the inverse Laplace transform of the solution (to get the solution in time domain).
The Laplace transform comes in two varieties: Bilateral, or two-sided Laplace transform, and unilateral,
or one-sided Laplace transform.

Study Note
The bilateral Laplace transform offers insight into the nature of system characteristics such as stability, causality,
and frequency response. The unilateral Laplace transform is a convenient tool for solving differential equations
with initial conditions.

4.1 The Bilateral Laplace Transform

Consider a case when eCt is input to LTI system.

LTI
Ct
Input = e System Output
Impulse
response = h(t )

Fig. 4.1 : LTI system with eCt as input


Ct
If H( s)   h(t ) e st dt then output is e  H (s )
s C
. This we have proved in previous chapter. The variable


s in H(s) is  + j,
s =  + j ...(2)
H(s) is called Laplace transform of h(t). Thus for any arbitrary signal x(t), Laplace transform can be
defined as,

L(x(t)) = X (s )   x (t ) e st dt ...(3)



 X (s ) =  x(t ) e(   j) t dt



=  x(t ) et e jt dt



Thus we can see that X(s) is Fourier transform of x(t) e–t. Thus when  = 0 we can say that,
L[x(t)] = f [x(t)] ...(4)
Since X(s) is Fourier transform of x(t) e–t, thus x(t) e–t should follow all Dirichlet conditions. That is
x(t) e–t should be absolutely integrable.

Study Note
This means Laplace transform X(s) is defined when x(t) e–t is absolutely integrable.


That is,
 x(t ) et dt   ...(5)


GATE MASTERS PUBLICATION

 146 Signals and Systems

In equation(5) we can see that  (Real part of s) is the variable . The values of s for which equation (5)
is valid for a signal x(t) is called Region of convergence (ROC) of X(s). Since X(s) is defined only when
equation (5) is satisfied. This means X(s) is defined for those values of  where equation (5) is satisfied.

Remember: Laplace transform is valid only when value of  lie within ROC.
For example: Let’s take Laplace transform of e–at u(t). Thus,

 at
X (s ) = e u(t ) e st dt {we assume a is real.}


 
( a  s ) t
= e dt   e( a    j) dt
0 0

1
= [e( a  j)   1]
( a    j )
Now for finite X(s) we want e–(a + +j < .
Since e–j will always have magnitude equal to 1, the condition will be satisfied when, e–(a + )< .
This will be satisfied when (a + ) > 0 because e–(a + ) will be 0 when (a + ) > 0.

Thus X(s) is finite when, a +  > 0    a

1
 X (s ) = [e( a    j)   1]
( a    j )

1 1
= 
a    j as

1
 X (s ) = ,   a
as
In above example Laplace transform is valid any when  > –a.

Example 4.1

Find Laplace transform of x ( t ) = e –(3 + 5j) u ( t ).

Solution 4.1
 
Now applying formulae, X (s ) =  x(t ) e st dt   e(3  5 j ) t e(   j)t dt
 0


[3    5 j  j ] t
= e dt
0

1
= [e(3   5 j  j)   1]
(3    5 j  j )
Now for finite X(s) we want (3 + ) > 0   > –3
1
 X(s ) = ,  > –3
(3  5 j  s)

GATE MASTERS PUBLICATION

 laplace Transform 147

Poles and Zeros of Laplace Transform

The most commonly encounted form of the Laplace transform in engineering is a ratio of two polynomials
in s, i.e.,

bms m  bm  1 sm1  ...  b1s  b0

X (s ) = ...(6)
a n sn  a n  1 sn  1  ...  a1s  a0
It is often convenient to factor the polynomials in the numerator and denominator, and to write the
transfer function in terms of those factors.

N( s) ( s  z1 ) ( s  z2 )...( s  zm  1 ) ( s  zm )
X (s ) = K ...(7)
D( s) ( s  p1 )( s  p2 )...( s  pn  1 ) ( s  pn )

where the numerator and denominator polynomials, N(s) and D(s), have real coefficients and K = bm/an.
As written in the above equation, the zi’s are the roots of the equation N(s) = 0, and are defined as zeros, and
the pi’s are the roots of the equation D(s) = 0, and are defined as poles. In equation (7) the factors in the
numerator and denominator are written so that, when s = zi the numerator N(s) = 0 and the Laplace transform
i.e.,
Lim X( s) = 0
s  zi

and similarly when s = pi the denominator polynomial D(s) = 0 and the value of the Laplace transform
becomes unbounded, that is
Lim X( s) = 
s  pi

Remember: The pole of a Laplace transform is where X(s) tends to  and Zero of a Laplace transform is where
X(s) tends to 0.

Example 4.2

For each of the following integrals, specify the values of the real parameter  which ensures
that the integral converges:
 0
1.  e5t e(   j  ) t dt 2.  e5t e(   j  ) t dt
0 

5 

3.  e5t e(   j  ) t dt 4.  e5t e(   j  ) t dt

5 


5.  e5 t e(   j  ) t dt


Solution 4.2
1 . Consider the given integral,
 
(   5) t
e
5 t
e(   j  ) t dt = e e jt dt
0 0

1 
= e(   5) t e jt
(5    j ) 0

GATE MASTERS PUBLICATION

148 Signals and Systems

 j t 5)t
Note that, e  1 regardless of the value of t. Therefore, as t  , e–( + e–jt  0 only if
( + 5) > 0 or,  < –5. Thus,

 0, (   5)  0    5
Lim e(   5)t e jt = 
t   , (   5)  0    5


(   5)t 1 1
Therefore, e e jt dt =  [0  1]  for  > –5
0
(5    j ) (5    j )

2 . Consider the given integral,

0 0

 e5t e(   j)t dt = e

(   5) t
e jt dt
 

1 0
=  e(   5) t e jt
(5    j ) 

 j t
Note that, e  1 regardless of the value of t. Therefore, t –, e–( + 5)t e–jt  0 only if ( + 5) < 0

or  < –5, and e–( + 5)t e–jt  only if ( + 5) > 0 or,  > –5. Thus,
0
( 5)t 1 1
Therefore, e e  j t dt =  [1  0]   for  < –5

(5    j ) (5    j )

5
3 . The given integral may be written as,
 e(5   )t e jt dt .
5

Since this integral has a finite range of limit, it has a finite value for all finite values of .
4 . The given integral may be written as,
 0 

 e(5   )t e jt dt =  e(5   )t e jt dt  e

(5   )t
e jt dt
  0

The first integral diverges for  < –5. The second integral diverges for  > –5. If  = 5, then both the
integrals still diverge. Therefore, the integral does not converge for any value of .
5 . The given integral may be written as,
0 

 e(5   )t e jt dt  e
(5   )t
e jt dt
 0

The first integral converges for  < 5. The second integral converges for  > –5. Therefore, the
integral converges for   5.

Example 4.3

Find Laplace transform of – e – at u (– t ).

Solution 4.3
 
X(s ) =  x(t ) e st dt  e
 at
u( t ) e st dt
 

GATE MASTERS PUBLICATION

 laplace Transform 149

0 0
 at  st ( a    j  )
=   e e dt   e dt
 

1
= [1  e(   a  j) ]
( a    j )
For finite X(s) we need a +  < 0   < –a
1
 X (s ) = , <–a
s a

4.2 Properties of ROC

Note that the Laplace transforms for the signals e–at u(t) and –e–at u(–t) are identical except for their
ROCs. Therefore, for a given X(s), there may be more than one inverse transform, depending on the ROC. In
other words, unless the ROC is specified, there is no one-to-one correspondence between X(s) and x(t). In this
section, we explain properties of the ROC of various classes of signals.
1. The ROC of X(s) consists of strips parallel to the j-axis in the s-plane.
2. If X(s) is rational, then the ROC must not contain any poles.

3. If x(t) is of finite duration and is absolutely integrable (i.e.  x(t ) dt   ), then the ROC is the


entire s-plane.
4. If x(t) is right-sided and of infinite duration (i.e., x(t) = 0 for all t < T1 for some finite T1), then the
ROC is the region in the s-plane to the right of the right most pole. That is ROC,  > real part of
right most pole.
5. If x(t) is left-sided and of infinite duration (i.e., x(t) = 0 for all t > T2 for some finite T2), then the
ROC is the region in the s-plane to the left of the left most pole, that is ROC,  < real part of
smallest pole.
6. If x(t) is two-sided and infinite duration (i.e. the signal is of infinite extent for both t < 0 and t > 0),
then the ROC will consist of a strip in the s-plane.

Example 4.4

Find Laplace transform of

( a ) x ( t ) = e –2 t u ( t ) – e –3 t u ( t )
( b ) x ( t ) = – e –2 t u (– t ) + e –3 t u (– t )

 e at , 0  t  T
( c ) x( t )  
 0 , elsewhere

Solution 4.4
(a) x(t ) = e–2t u(t) – e–3t u(t)
Signal x(t) has two components.
1
1. e–2t u(t) has Laplace transform = .
s2
So, pole is at –2, signal is right sided.

GATE MASTERS PUBLICATION

150 Signals and Systems

 ROC is  > –2,

Laplace transform 1
 e2t u(t )   ,   2
s2
1
2. e–3t u(t) has Laplace transform = .
s3
So, pole is at –3, signal is right sided so ROC is  > –3.
Laplace 1 1 1
 For x (t )    , (  2)  (   3) = 2 , (   2)
Transform s 2 s 3 s  5s  6
(b) x ( t ) = e–2t u(–t) + e–3t u(–t)
Signal x(t) has two signals.
1
1. e–2t u(–t) has Laplace transform , and pole is at –2, signal is left sided so ROC will be  < –2.
s2
1
2. e–3t u(–t) has Laplace transform  and pole is at –3, signal is left sided so ROC will be  < –3.
s3
Laplace 1 1 1
 x (t ) 
Transform
  , (   2)  (   3)  2
, (   3)
s 2 s 3 s  5s  6

 e at , 0  t  T
(c) x(t ) = 
 0, elsewhere
 T
 st
 X (s ) =  x(t ) e dt   e at e st dt
 0

1
= (1  e (a  s )T )
(s  a )
Now if we will look at X(s) we can see that at s = –a, X(s) is not  so X(s) has no pole at s = –a.
d ( a  s )T 
 dt (1  e )
Lim X( s) = Lim  
  a   a
 d
(s  a ) 
 dt 
 X(–a) = T
So, ROC can be entire s-plane because no pole exist for X(s) at any location. Also at s = 0 and s = 
X(s) is finite. So ROC is entire is s-plane.

Example 4.5

Determine the Laplace transform of

( a ) a unit impulse function x ( t ) = ( t )
( b ) a unit step function x ( t ) = u ( t )
( c ) a unit ramp signal x ( t ) = r ( t )
Solution 4.5
( a ) Given that, x ( t ) = ( t )
 
By definition, X(s ) =  x(t ) e st dt   st
 (t ) e dt  e st 1
  t 0

Therefore, ( t )  1 , ROC is the entire s-plane

GATE MASTERS PUBLICATION

 laplace Transform 151

( b ) Given that, x ( t ) = u( t )
  
By definition, X (s ) =  x(t ) e st dt   u(t ) e st dt   e st dt
  0


1 1
=  e st   [0  1], 0
s 0 s

1
X (s ) = , 0
s
1
Therefore, u(t )  , 0
s
( c ) Given that, x ( t ) = r(t) = t u(t), by definition
  
X (s ) =  x(t ) e st dt   t u(t ) e st dt   t e st dt
  0
  
1 1 1 1
=  t e st   e st dt  0  e st =  [0  1],   0
s 0 s 0 s2 0 s2
1
X (s ) = 0
s2
1
Therefore, r(t )  , 0
s2

Example 4.6

Find Laplace transform of signal x ( t ) = e – a  t  when a > 0 and a < 0.

Solution 4.6
Case-1: When a > 0

So, x(t ) = e a t  e at u(t )  e at u( t )

Laplace 1
e at u(t ) 
Transform
 ,   a
sa
Laplace 1
eat u( t ) 
Transform
 , a
sa
Laplace 1 1
So for x(t ) 
Transform
  , (  a )  (  a )
sa sa
1 1
 , ( a    a )
=
s a s a
So, when a > 0 then Laplace transform exist as ROC is –a <  < a.
x (t)
1.0

eat e –at

Figure: x(t) when a is positive

GATE MASTERS PUBLICATION

152 Signals and Systems

Case-2: When a is negative then also we get same expressions,

Laplace 1 1
x (t ) 
Transform
  , ( a    a )
s a s a
Here if we look at ROC since a is negative ROC will be null, ROC =  thus Laplace do not exist.
x (t)

eat e–at

Figure: x(t) when a is negative

Study Note
Laplace transform of a signal x(t) will not exist when ROC is a null, ROC =  because in that case for no value of
 Laplace transform will be finite.

Example 4.7

Find Laplace transform of x ( t ) = 1.

Solution 4.7
x ( t ) = 1 = u(t) + u(–t)
Laplace transform of u(t) is 1/s, since it is right sided signal and pole is at 0 so ROC is  > 0.
Now, the Laplace transform of u(–t) is –1/s and it is left sided signal. Since pole is at 0 so ROC is  < 0.
The ROC of x(t) is ( < 0)  ( > 0) = .
So, Laplace transform of x(t) = 1 do not exist.
x (t)

1.0

REMEMBER We can see that why Laplace transform of x(t) =1 do not exist because x(t) is not absolutely
integrable and when we try to find x(t) e–t we can see that it will also be not integrable for any
value of .
And clearly we can see that when x(t) = u(t) for  > 0 we can find x(t) e–t which is absolutely
integrable. Similarly in example 4.6 when a < 0, then x(t) is rising in both directions of t. So
here also x(t) e–t is not absolute integrable for any value of .
Thus we can say that if signal is rising both side or signal is constant from – < t <  then
Laplace transform do not exist.

GATE MASTERS PUBLICATION

 laplace Transform 153

1.0
1.0
t t

Fig. : e–t when  > 0 Fig. : e–t when  < 0

4.3 Properties of Laplace Transform

4.3.1 Linearity
Laplace transform
If x1(t )   X1( s), ROC = R1

Laplace transform
and x 2 (t )  X 2 (s ), ROC = R2
then, ax1(t) + bx2(t)  aX1(s) + bX2(s) with ROC containing R1  R2. ...(6)
As indicated, the ROC of X(s) is at least the intersection of R1 and R2, which could be empty, in which
case X(s) has no ROC, i.e., x(t) has no Laplace transform.
Proof : The Laplace transform of ax1(t) + bx2(t) is given by,

 st
L[ax1(t) + bx2(t)] =  [ax1(t )  bx2 (t )] e dt


 
= a  x1(t ) e st dt  b  x2 ( t ) e
 st
dt
 

= aX1(s) + bX2(s)

Note: In case if R1  R2 =  then Laplace transform of x1(t) + x2(t) do not exist.

Example 4.8

In this example, we illustrate the fact that the ROC for the Laplace transform of a linear
combination of signals can sometimes extend beyond the intersection of the ROCs for the
individual terms.
Solution 4.8
Consider, x ( t ) = x1(t) – x2(t)
where the Laplace transform of x1(t) and x2(t) are, respectively,
1
X1( s ) = , Re( s)  1
s 1
1
and X2( s ) = , Re( s)  1
( s  1) ( s  2)
The pole-zero plot, including the ROCs X1(s) and X2(s), is shown in figure (a) and (b).
1 1 s 1 1
X(s ) =   
s 1 ( s  1) ( s  2) ( s  1) ( s  2) s  2

GATE MASTERS PUBLICATION

 154 Signals and Systems

Thus, in the linear combination of x1(t) and x2(t), the pole at s = –1 is canceled by a zero at s = –1. The
pole-zero plot of X(s) = X1(s) – X2(s) is shown in figure (c). The intersection of the ROCs for X1(s) and
X2(s) is Re{s} > –1. However, since the ROC is always bounded by a pole or infinity, for this example the
ROC for X(s) can be extended to the left to be bounded by the pole at s = –2, as a result of the pole-zero
cancellation at s = –1.
Im Im Im

s-plane s-plane s-plane

R1 R2 R3

Re Re Re
–1 –2 –1 –2

(a) (b) (c)

Figure : Pole-zero poles and ROCs for (a) X1(s); (b) X2(s); (c) X1(s) –X2(s)

REMEMBER When even we are applying any property to find Laplace transform of a signal then find the
expression of Laplace transform using properties and do not find ROC using properties because
we can easily get ROC from expression of Laplace transform using properties of ROC
(Section 4.2).So learn how property effects the expression of laplace transform and no need to
learn the effect on ROC.

4.3.2 Time Shifting

If, x ( t )  X(s) with ROC = R
then, x(t – t0)  X(s) e–st0 with ROC = R ...(7)
Proof : The Laplace transform of x(t – t0) is given by

 st
L[x(t – t0)] =  x(t  t0 ) e dt


A change of variables is performed by letting  = t – t0, which also yields d = dt,  – as t  –,
and   –, and  as t . Therefore,
 
L[x(t – t0)] =  x( ) e s(   t0 ) d  e st0  s
 x() e d
 

L[x(t – t0)] = X( s) e st0

1 e as
The laplace transform of u(t) is then laplace transform of u(t  a) will be .
s s

1 e as
The laplace transform of r(t) is then laplace transform of r(t  a) will be .
s2 s2

t  T  T A sT / 2  sT / 2
The laplace transform of A rect   = u  t    u  t    ( e e ).
T  2  2 s

GATE MASTERS PUBLICATION

 laplace Transform 155

Study Note
We can see that shifting operation by a in time domain lead to multiplication of e–as in laplace domain. Since
multiplication of e–as do not change the location of poles in laplace transform, also the shifting operation donot
change the format of the signal that is left sided signal remain left sided, right sided signal remain left sided signal,
finite duration signal remain finite duration. Thus ROC will not change by shifting operation.

4.3.3 Shifting in the s-domain

If, x ( t )  X(s) with ROC = R
then, x ( t ) e so t  X(s – so) with ROC = R + Re{s0} ...(8)

Proof : The Laplace transform of x(t) e so t is given by

 
s t  st ( s  so )t
L[x(t) e so t ] =  [ x (t ) e 0 ] e dt   x (t ) e dt = X(s – so)
 

REMEMBER • If X(s) has a pole or zero at a then X(s – so) will have pole or zero at a + so. So we can see

that shifting in s-domain by s0 leads to multiplication of e so t in time domain. So format of

signal do not change.
• If ROC of X(s) is R then X(s – s0) will have ROC = R + Re{s0}, because pole of X(s – s0) will
shift by s0 as seen in above point.
For example:
Laplace transform 1
u(t )   , ROC is  > 0
s
Laplace transform 1
 e at u(t )   , ROC is  > – Re{a}
sa
Laplace transform 1
 e j u(t )   , ROC is  > 0
s  j

Laplace transform 1
 e j u(t )   , ROC is  > 0
s  j

Laplace transform 1 1 1  s
 cos( 0 t ) u(t )     ,   0  , 0
2  s  j s  j  s  2

Laplace transform 1  1 1  
 sin( 0 t ) u(t )     ,   0  2 , 0
2 j  s  j  s  j  s  2

Study Note
We can see the in time shifting,

x (t ) X(s), x (t – a ) X(s) e–as

The shifting goes with same sign, but in frequency shifting

s0t
x(t) X(s), x (t ) e X(s – s0)
The shifting goes with opposite sign.

GATE MASTERS PUBLICATION

 156 Signals and Systems

4.3.4 Time Scaling

If x ( t )  X(s) with ROC = R
1 s
then, x ( at )  X   with ROC = aR ...(9)
a a
Aside from the amplitude factor of 1/a, scaling in time domain by a factor of a corresponds to scaling in
the s-domain by a factor of 1/a. Also, for any value of s in the ROC of x(t), the value s/a will be in the ROC of
x(at); that is, the ROC associated with x(at) is a compressed (a > 1) or expanded (a < 1)version of the ROC of
x(t).
Proof : The Laplace transform of x(at) is

 at
L[x(at)] =  x( at ) e dt


Case-I: For a positive real constant ‘a’


 st
L[x(at)] =  x( at ) e dt


A change of variables is performed by letting  = at, which also yields d = a dt, – as t  –, and
  as t . Therefore,

1 s 
L[x(at)] = 1  x ( ) e ( s /a )  d  = X  
a  a a 

Case-II: For a negative real constant ‘a’


 st
L[x(at)] =  x ( at ) e dt


A change of variables is performed by letting  = at, which also yields d = a dt,  as t  –, and
 – as t . Therefore,
 
1 s
L[x(–at)] =  1  x ( ) e ( s / a )  d   1  x ( ) e ( s / a )  d  = X   
a  a  a  a
Combining the two cases, we obtain,
1 s
L[x(at)] = X   , with ROC R1 = aR
a a
That is, for any value s in R [which is illustrated in Fig. 4.2 (a)], the value as will be in R1, as illustrated
in Fig. 4.2 (b) for a positive value of a < 1. Note that, for 0 < a < 1, there is a compression in the size of the
ROC of X(s) by a factor of a, as depicted in Fig. 4.2 (b), while for a > 1, the ROC is expanded by a factor of a.
Also, equation (9) implies that if a is negative, the ROC undergoes a reversal plus a scaling. In particular, as

1 s
depicted in Fig. 4.2 (c), the ROC of X   for 0 > a > –1 involves a reversal about the j-axis, together with
a a

a change in the size of the ROC by a factor of a . Thus, time reversal of x(t) results in a reversal of the ROC.
That is,
L
x( t )  X(  s); with ROC = –R ...(10)

GATE MASTERS PUBLICATION

 laplace Transform 157

Im Im Im

s-plane s-plane s-plane

R

r2 r1 Re Re Re
ar2 ar1 ar1 ar2

(a) (b) (c)

Fig. 4.2 : Effect on the ROC of time scaling:

(a) ROC of X(s); (b) ROC of (1 / a ) X( s / a ) for 0 < a < 1; (c) ROC of (1 / a ) X( s / a ) 0 > a > –1

REMEMBER If x(t) has poles and zeros of it’s laplace transform at pi and zi then x(at) will have pole and zeros
of it’s laplace transform at api and azi .

Example 4.9

Find Laplace transform of

( a ) u (–2 t – 7)
( b ) e3t – 7 u (3 t – 9)
Solution 4.9
( a ) Since we know that,

1
u(t )  ,   0
First shifting,
s
1
u(t  7)  e7 s
s
Then scaling,
1 1
u(2t  7)  e7 s / 2  e7 s / 2
 s s
2 
Then inversion, 2
1
u( 2t  7)   e7 s / 2
s

1
So we get expression of Laplace transform  e7s / 2 .
s
Since, Laplace transform has pole at origin and signal is left sided so ROC is  < 0.
( b ) To find Laplace transform of e3t – 7 u(3t – 9)
1
u(t ) 
s
1
et u(t ) 
s 1
1
et  9 u(t  9)  e9s
s 1
1 1 1
e3t  9 u(3t  9)  e3s  e3 s
3 s 1 s3
3

GATE MASTERS PUBLICATION

 158 Signals and Systems

1
e2 e3t  9 u(3t  9)  e2  e3 s
s3
1
e3t  7 u(3t  9)  e23 s 
s3
Since signal is right sided and Laplace transform has a pole at 3 so ROC is  > 3.

Study Note
Equation (9) show that expansion in time domain that is if a  1 then Laplace domain contract and vice-versa. So
scaling in time domain lead to scaling in Laplace domain and vice-versa.

4.3.5 Differentiation in Time Domain

If x(t)  X(s) with ROC = R
dx(t )
then,  sX( s) with ROC containing R ...(11)
dt
Proof : If x(t) has it’s Laplace transform X(s), that is

X (s ) =  x(t ) e st dt


d
and let Y(s) is Laplace transform of x(t ) that is
dt

d
Y(s) =  x(t ) e st dt

dt



= x(t ) e st   st
 (s) x(t ) e dt




 st
= 0s  x(t ) e dt


= sX ( s )
Similarly differentiation property can be extended to yield,

d n x(t )
 sn X ( s ) ...(12)
dt n
Consequently, dx(t)/dt is the inverse Laplace transform of sX(s). The ROC of sX(s) includes the ROC of
X(s) and may be larger if X(s) has a first-order pole at s = 0 that is canceled to the multiplication by s. For
example, if x(t) = u(t), then X(s) = 1/s, with an ROC that is Re{s} > 0. The derivative of x(t) is an impulse with
an associated Laplace transform that is unity and an ROC that is the entire s-plane.

GATE MASTERS PUBLICATION

 laplace Transform 159

Example 4.10

Using the bilateral time-shift and differentiation properties find the Laplace transform of
d 2 3(t  2)
x(t) = (e u (t  2))
dt 2
Solution 4.10
We know that,
1
e3t u(t )  ,   3
s3
Using the time-shifting property, we obtain
1
e3( t  2) u(t  2)  e2 s ,   3
s3
Now, using differentiation in the time-domain property, we obtain
d2 3( t  2) s2 2 s
e u( t  2)  e ,   3
dt 2 s3

4.3.6 Differentiation in the s-Domain

If x(t )  X( s) with ROC = R

dX( s)
then, t x(t )   with ROC = R ...(13)
ds
Proof: By definition,

 st
X (s ) =  x(t ) e dt


Differentiating both sides w.r.t. s, we get


dX( s)  st
ds
=  [t x(t )] e dt


dX( s)
= L[–t x(t)]
ds
dX( s)
Therefore, t x(t )  
ds
Similarly, differentiation in the s-domain property can be extended to yield,

d n X( s)
t n x(t )  ( 1)n ...(14)
dsn

Example 4.11

Determine the L aplace transform ROC for following signals:

( a ) t e – at u ( t )
( b ) t sin( 0 t ) u ( t )
( c ) t cos( 0 t ) u ( t )

GATE MASTERS PUBLICATION

 160 Signals and Systems

Solution 4.11
Laplace transform 1
( a ) We know that, e at u(t )  
sa
Now if we apply differentiation property, we get
Laplace transform d  1  1
t e at u (t )  1 
ds  s  a  (s  a )2
Since, Laplace transform has pole at –a and signal is right sided signal so ROC is  > – a.

Laplace Transform   
( b ) Since, sin( 0t ) u (t )   2 0 2  , 0
s 
 0 
To find Laplace transform of t sin(0t) u(t), we apply differentiation in s-domain property

d  0  2 0 s
 t sin(0t ) u(t )   1   2
ds  s2  02  s  02
Since, poles of Laplace transform are at ±j0, and signal is right sided so ROC is  > 0.
Laplace s
( c ) Since, cos( 0 t ) u(t )   , 0
Transform
s  02
2

To find Laplace transform of t cos(0t) u(t), we apply differentiation in s-domain property.

d  s  s 2  02
 t cos(0t ) u (t )   1  2 2
  2 2
ds  s  0  (s  0 )
Since, Laplace transform has poles at ±j0 and signal is right sided so ROC is  > 0.

Study Note
We can find Laplace transform of sin(0t) u(t) and cos(0t) u(t) but we cannot find Laplace transform of cos(0t)
and sin(0t), because they exist for – < t <  and when we multiply signal with e–t we cannot find any value of
 for which x(t) e–t is absolutely summable. Thus there is no valid ROC for these signals.
For example:
sin(0t) = sin(0t)  [u(t) + u(–t)]
= sin(0t) u(t) + sin(0t) u(–t)

Laplace 0
Now, sin( 0 t ) u( t )   , 0
Transform
s2  20

Laplace 02
and sin(0t ) u( t )   , 0
Transform
s2  20
Since, ROC = ( > 0)  ( < 0) = 
so Laplace transform of sin(0t) do not exist.

4.3.7 Convolution Property

If x 1( t )  X1(s) with ROC = R1
and x 2(t)  X2(s) with ROC = R2
x1(t)  x2(t) X2(s) X2(s) with ROC containing R1  R2 ...(15)

GATE MASTERS PUBLICATION

 laplace Transform 161

The Laplace transform maps the convolution of two signals into the product of their Laplace transforms.
The ROC of X1(s) X2(s) includes the intersection of the ROCs of X1(s) and X2(s) and may be larger if pole-zero
cancellation occurs in the product.
Proof : The Laplace transform of x1(t)  x2(t) is given by

 st
L[x1(t)  x2(t)] =  [x1(t )  x2 (t )] e dt



   st
=    x1( ) x2 (t  ) d  e dt
   
Interchanging the order of integration and noting that x1() does not depend on t, we obtain
  
L[x1(t)  x2(t)] =  x 1(  )   x 2 (t  ) e st dt  d 
 
   
Using the time-shifting property the bracketed term is X2(s) e–st. Substituting this into the above equation
yields,

 s
L[x1(t)  x2(t)] =  x1( ) ( X2 (s) e ) d



= X2 ( s)  x1( ) e s d


L[x1(t)  x2(t)] = X2(s) X1(s) = X1(s) X2(s)

Therefore, x1(t)  x2(t)  X1(s) X2(s)
The convolution property states that convolution in the time domain corresponds to multiplication in
the s-domain.

Example 4.12

Consider a signal y ( t ) which is related to two signals x 1 ( t ) and x 2 ( t ) by,

y ( t ) = x 1 ( t – 2)  x 2 (– t + 3)
where, x 1 ( t ) = e –2 t u ( t ) and x 2 ( t ) = e –3 t u ( t )
Use properties of the Laplace transform to determine the Laplace transform Y ( s ) of y ( t ).
Solution 4.12
1
We know that, x 1( t ) = e2t u(t )  ,   2
s2
Using the time-shifting property, we obtain
e2s
x1(t – 2) = e2(t  2) u(t  2)  ,   2
s2
Similarly, we obtain
1
x 2( t ) = e 3t u (t )  ,   3
s 3
Applying time shifting property we get
e 3s
x2(t + 3) = e 3(t 3) u (t  3)  ,   3
s 3

GATE MASTERS PUBLICATION

 162 Signals and Systems

Now, using the time-reversal property, we obtain

e3s
x2(–t + 3) = e3( t  3) u( t  3)  , 3
s  3
Now, using the convolution property, we obtain
L[y(t)] = L[x1(t – 2)] L[x2(–t + 3)]
 e2 s   e3 s 
Y(s) =     , 2    3
 s  2   s  3 

e5s
Y(s) = , 2    3
( s  2) (3  s)

REMEMBER Suppose in any question we have to find convolution of two signal x1(t) and x2(t). Then first of
all find X1(s) and X2(s) and inverse Laplace transform of X1(s)  X2(s). This will be much easier
than what we have learnt in Chapter-2.

In equation (15) we saw that ROC of Laplace transform of convolution of two signals will be R1  R2 but
this can become in valid in some cases where pole-zero cancellation take place.
For example:
( s  5)
1. X 1( s ) = , 67
( s  6) ( s  7)

( s  6)
and X2( s ) = ,  8    5
( s  5) ( s  8)
Here we can see that R1  R2 =  then also we can find x1(t)  x2(t) and its Laplace transform will
exist because here pole-zero cancellation take place.

Laplace transform 1
x 1(t )  x 2 (t )  , 8    7
(s  7) (s  8)

s5
2. X1( s ) = , 67
( s  6) ( s  7)

1
and X 2( s ) = ,  8    5
( s  5) ( s  8)

Laplace transform 1
So, x1( t )  x2 (t ) 
 , 67
( s  6) ( s  7) ( s  8)

1
3. X1( s ) = , 67
( s  6) ( s  7)

1
and X2( s ) = ,  8    5
( s  5) ( s  8)

Here no pole-zero cancellation, so Laplace transform of x1(t)  x2(t) will not exist.

GATE MASTERS PUBLICATION

 laplace Transform 163

4.3.8 Integration in Time-Domain

If x ( t )  X(s) with ROC = R
t
1
then,  x( ) d  X( s) with ROC containing R  { > 0} ...(16)

s
Proof: If a signal x(t) is convolved with a unit step function u(t), we get

x(t)  u(t) =  x( ) u(t  ) d


1,   t
Since, u(t – ) = 
0,   t
t
we have, x(t)  u(t) =  x( ) d

The convolution of a signal with a unit step function is the same as the cumulative integral of the signal.
Now we can prove the integration property of the Laplace transform,
t

 x( ) d = x(t)  u(t)


t 
L   x( ) d = L[x(t)  u(t)]
  
Using the convolution property, we obtain
t 
L   x( ) d = X( s ) U( s)  1 X( s)
   s
t
1
Therefore,  x( ) d 
s
X( s)


4.3.9 Conjugation Property

If x ( t )  X (s )
then, x  ( t )  X (s) ...(17)
Proof : The Laplace transform of x(t) is
 
  
L[x(t)] =  x  (t ) e st dt    x (t )e s t dt   [ X (s  )]  X (s  )
   
Therefore, if x(t) is real, that is, x(t) = x*(t), then
X(s ) = X ( s  ) ...(18)

X (s) = X(s )  ...(19)
 
Thus, when signal is real then X(s) = [X(s )] . That is if X(s) has a pole or zero at a then it must have
pole or zero at a. That is
Thus in rational Laplace transform of real signal the coefficients of s will be real. If pole or zero in
Laplace transform of real signal is complex then there will be a pole or zero at it’s conjugate location also that
is complex pole or zero always occur in conjugate pairs, it pole or zero is real then they can sit alone without any
pair.

GATE MASTERS PUBLICATION

 164 Signals and Systems

4.3.10 Reversal in Time

If, x ( t )  X(s), ROC is R
then, x (– t )  X(–s), ROC is –R
We have already proved this property. If signal is even then,
x ( t ) = x (– t )
 X(s ) = X(–s)
Thus, X(s) will have only even power of s. The poles and zeros will be at mirror image with respect to
imaginary axis. That is if pole or zero is at a then it will be at –a also.
If signal is odd then, x ( t ) = –x (– t )
 X(s) = –X(–s)
Thus X(s) will have only odd power of s. For odd signals also if pole or zero is at a then it will be
at –a also.

 st
REMEMBER Since Laplace transform of signal x(t) is X(s) =  x (t )e dt , and for odd signals we know that


 x (t )dt  0 thus we can say that X(0) =0. That is Laplace transform of odd signal always

have zero at s = 0.

Property Signal Transform ROC

x (t) X(s) R

x 1(t) X1(s) R1

x 2(t) X2(s) R2

Linearity a 1x 1(t) + a 2x 2(t) a 1X1(s) + a 2X2(s) R  R1  R2

– s0 t
Time shifting x (t – t0) e X(s) R = R
Shifting in s es0t x (t) X(s – s0) R = R + Re(s0)
1 s
Time scaling x (at) X( ) R = aR
a a

Time reversal x (–t) X(–s) R = –R

dx(t ) sX(s)
Differentiation in t R  R
dt
dX( s)
Differentiation in s –tx (t) R = R
ds
t 1
Integration  x() d X( s) R  R1  {Re(s) > 0}
s
Convolution x 1(t)  x 2(t) X1(s) X2(s) R  R1  R2

4.4 Inverse Laplace Transform

Inverse Laplace transform means to find out signal x(t) from it’s Laplace transform X(s). Symbolically
it is represented by,
x ( t ) = L –1{X(s)} ...(20)

GATE MASTERS PUBLICATION

 laplace Transform 165

Method-1: Inverse Formulae

This is the procedure that can be used for all type of Laplace transform, this include evaluation of a line
integral in complex s-plane; that is
C  j
1
x(t ) =  X( s) est ds ...(21)
2j C  j

In the above integral the integration is performed for any value of C within ROC.

Note: Generally this method is not used.

2. Partial Fraction Expansion

If X(s) is a rational function, that is, of the form
N( s)
X (s ) = ...(22)
D( s)
For partial fraction the X(s) should be a proper function that is degree of numerator < degree of
denominator. If in any question this condition is not satisfied then divide numerator with denominator and
convert X(s) into
R( s)
X(s) = Q( s)  ...(23)
D( s)

R( s)
Now solve using partial fraction.
D( s)
Case-1: When denominator has only single order roots or X(s) has single order poles.
N (s ) (s  Z 1 ) (s  Z 2 ) .... (s  Z m )
So, X (s ) = K ...(24)
D (s ) (s  P1 ) (s  P2 ) .... (s  Pn )
In this case when all poles are of single order,
C1 C2 Cn
X (s ) =   ....  ...(25)
(s  P1 ) (s  P2 ) (s  Pn )

Here, Ck = X (s )  (s  Pk ) s  P ...(26)
k

Each pole in X(s) can give a right sided or left sided signal

Left sided Pk t
–Ck e u(–t),  < Re{Pk }
Signal
Ck
s – Pk
Right sided Pk t
Ck e u(t),  > Re{Pk}
Signal

If in a question ROC of X(s) is known then

• If ROC is in right side of pole then pole correspond to right sided signal.
• If ROC is in left side of pole then pole correspond to left sided signal.
• If in a question ROC of X(s) is not given then we have to find all possible x(t) that can correspond
to X(s) as Laplace transform.

GATE MASTERS PUBLICATION

166 Signals and Systems

REMEMBER If ROC of X(s) is not given and we have to find number of time domain signals that will correspond
to X(s), the method to find the number of time domain signals the procedure is as follows.

First of all mark all poles of X(s) on s-plane and draw lines parallel to imaginary axis and find
out into how many regions the s-plane is divided. Each region will correspond to a different x(t).
So number of time domain signal is equal to number of regions in which s-plane is divided.

Example 4.13

( s  10) ( s  15)
If, X( s )  then find number of time domain signal that correspond
( s  16) ( s2  5s  6) ( s2  7)
2

to X ( s ) as Laplace transform.
Solution 4.13
The poles of X(s) are at  4 j,  7 j ,  3,  2 let us mark all of these on s-plane.

Im

4j

7j

Re
–3 –2

 7j

–4j
R1 R2 R3 R4

So, 4 regions are created in the plane  > 0,  < 3, 3 <  < 2, 2 <  < 0. So, 4 time domain signals
can be created.

Example 4.14

2s  4
If, X( s )  ,  3    1 then find the time domain signal corresponding to it.
s2  4 s  3
Solution 4.14
2s  4 C1 C2
Using partial fraction, X (s ) = 2
 
s  4s  3 ( s  3) ( s  1)

 C 1 = ( s  3) X( s) s  3  1

 C 2 = ( s  1) X( s) s  1  1

1 1
So, X(s ) = 
( s  3) ( s  1)

GATE MASTERS PUBLICATION

 laplace Transform 167

If we plot the ROC and mark poles on s-plane, we get

Im

Re
–3 –1

Since ROC is left of –1 so pole at –1 correspond to left sided signal and ROC is right of –3 so pole at –3
correspond to right sided signal.
1 1
Since, X (s ) = 
( s  3) ( s  1)
 x(t ) = e–3t u(t) – e–t u(–t)
We can see that ROC is a strip and the corresponding time domain signal is both sided.

Example 4.15

Find inverse Laplace transform of

s2  6 s  7
(a) X(s) = 2 ,   1
s  3s  2
s3  2 s 2  6
(b) X(s) = , 0
s2  3 s
Solution 4.15
( a ) Since X(s) is not proper function so first of all we divide numerator by denominator.
3s  5
 X (s ) = 1 
( s  1) ( s  2)
3s  5
Applying partial fraction as ,
( s  1) ( s  2)
3s  5 C C
X 1( s ) =  1  2
( s  1) ( s  2) s  1 s  2

 C 1 = X1( s) ( s  1) s  1  2

and C 2 = X1( s) ( s  2) s   2  1

2 1
 X1( s ) = 
s 1 s2
2 1
 X (s ) = 1  
( s  1) ( s  2)

GATE MASTERS PUBLICATION

168 Signals and Systems

Now plotting ROC and marking poles of X(s) on s-plane we get

Im

Re
–2 –1

Since ROC is right of both poles, so both correspond to right sided signal.
2 1
Since, X (s ) = 1  
s 1 s  2
x(t ) = (t) + 2e–t u(t) + e–2t u(t)
( b ) Here also X(s) is not proper, so
s3  2s2  6 3s  6
X (s ) = 2
 s 1 
s  3s s( s  3)

3s  6
Let, X 1( s ) = ,
s( s  3)
Now applying partial fraction at X1(s) we get
C1 C2
X 1( s ) = 
s s3
 C 1 = X1( s) s s  0  2

and C 2 = X1( s) ( s  3) s  3  1

2 1
so, X 1( s ) = 
s s3
2 1
 X (s ) = s  1  
s s3
Since ROC [ > 0] is right to both poles [0, –3].
 Both pole given right sided signals.
1 2
Since, X(s ) = s  1  
s s3
We know that laplace transform of (t) is 1 thus laplace transform of (t) will be s (using differentiation
in time domain property)
So, x ( t ) = (t) – (t) + 2u(t) + e–3t u(t)

GATE MASTERS PUBLICATION

 laplace Transform 169

Example 4.16

(3 s  7)
Find time domain signal corresponding to X ( s ) = 2
.
( s  5s  6)
Solution 4.16
Since ROC is not given so first to all we apply partial fraction on X(s),
3s  7 C C
X (s ) =  1  2
( s  3) ( s  2) s  3 s  2

 C1 = ( s  3) X( s) s  3  2

and C 2 = ( s  2) X( s) s  2  1
1 2
So, X (s ) = 
s2 s3
There are two poles –2 and –3, there are three regions in which s-plane is divided.
Im

R1 R2 R3

Re
–3 –2

• So signal corresponding to ROC = R1 i.e.  < –3. Since ROC is left of both poles so both poles
correspond to left sided signal.
So, x(t ) = –e–2t u(–t) –2 e–3t u(–t)
• Now, when ROC = R2 i.e. –3 <  < –2 then pole –2 given left sided signal and pole –3 given right sided
signal.
So, x(t ) = e–2t u(–t) + 2 e–3t u(t)
• Now, when ROC = R3 i.e.  > –2 then both pole given right sided signal.
So, x(t ) = e–2t u(t) + 2 e–3t u(t)
In previous case we analysed the Laplace transform where poles were of single order

Case-2 : Laplace Transform with non single order poles.

Consider a case where D(s) or denominator of X(s) has roots which are not of single order. Let us
assume one root of D(s) has order r, that is
N (s ) (s  Z 1 ) (s  Z 2 )....( s  Z m )
X(s)= K
D (s ) (s  P1 ) (s  P2 )....(s  Pi )r ....(s  Pn )
The partial fraction for all poles will be same as in case-1 except for pole Pi,

C1 C2  1 2 r  Cn
X(s) =  ....   ....   ....
(s  P1 ) (s  P2 )  (s  Pi ) (s  Pi )2 (s  Pi )r  s  Pn

GATE MASTERS PUBLICATION

170 Signals and Systems

Here, Ck = (s  Pk ) X (s ) s  P ...(27)
k

1 dk
and r – k = [(s  Pi )r X (s )] ...(28)
k ! ds k s  Pi

It depend on ROC of X(s) that pole will correspond to right side signal or left side signal.

REMEMBER While finding inverse Laplace transform in case-2, this formulae can be helpful.
When pole correspond to right sided signal, when  > Real {Pi}

A For right A  t m  1 Pi t
  e u (t )
(s  Pi )m Sided signal ( m  1)!
When pole correspond to left sided signal, when  < Real {Pi]

A For ledt t m  1 Pi t
m

Sided signal
A e u ( t )
(s  Pi ) ( m  1)!

Example 4.17

s2  2 s  5
Find the inverse Laplace transform of X( s )  ,   3 .
( s  3) ( s  5)2

Solution 4.17
We see that, X(s) has one simple pole at s = –3 and one multiple pole at s = –5 with multiplicity 2. Then by
we have,
c1 1 2
X (s ) =  
s 3 s  5 (s  5)2
we have, c 1 = ( s  3) X( s) s  3

s2  2 s  5
= 2
( s  5)2 s  3
2
 2 = ( s  5) X( s)
s  5

s2  2s  5
=  10
s  3 s  5

d
1 = [( s  5)2 X( s)]
ds s  5

d  s2  2 s  5  s2  6s  1
=     1
ds  s  3  s  5 ( s  3)2 s  5

2 1 10
Hence, X (s ) =  
s  3 s  5 ( s  5)2
The ROC of X(s) is Re(s) > –3. Thus, x(t) is a right-sided signal and we obtain
x ( t ) = 2e–3t u(t) – e–5t u(t) – 10t e–5t u(t)
= [2e–3t – e–5t – 10t e–5] u(t)

GATE MASTERS PUBLICATION

 laplace Transform 171

Example 4.18

Find the inverse Laplace transform of

2  2 s e2 s  4 e4 s
X(s) = ,   1
s2  4s  3
Solution 4.18
We see that, X(s) is a sum
X(s ) = X1(s) + X2(s) e–2s + X3(s) e–4s
2
where, X 1( s ) = 2
s  4s  3
2s
X 2( s ) = 2
s  4s  3
4
X 3( s ) =
s2  4s  3
If x1(t)  X1(s) x2(t)  X2(s) x3(t)  X3(s)
then by the linearity property and the time-shifting property we obtain,
x(t ) = x1(t) + x2(t – 2) + x3(t – 4)
Next, using partial fraction expansions and from table, we obtain
1 1
X 1( s ) =   x 1(t )  (e t  e 3t ) u (t )
s 1 s  3
1 3
X 2( s ) =    x 2 (t )  ( e t  3e 3t ) u (t )
s 1 s  3
2 2
X 3( s ) =   x3 (t )  2( et  e3t ) u(t )
s 1 s  3
Thus, x(t ) = (e–t – e–3t) u(t) + [–e–(t – 2) – 3 e–3(t – 2)] u(t – 2) +
2[e–(t – 4) – e–3(t – 4)] u(t – 4)

4.5 Analysis and of LTI System Using Laplace Transform

Suppose we have an LTI system with impulse response h(t) and input x(t), the output will be
y(t) = x(t)  h(t).
Output = Input  Impulse response ...(29)
If we take Laplace transform of above equation we get
Y(s) = X(s)  H(s) ...(30)

h(t)
x (t) y (t) = x (t)  h(t)

X(s) Y(s) = X(s) H(s)

H(s)

Fig. 4.3 : Impulse response and System function

GATE MASTERS PUBLICATION

 172 Signals and Systems

In equation (30) H(s) is Laplace transform of h(t), here

Y( s) Laplace transform of output
H (s ) =  ...(31)
X( s) Laplace transform of input

Study Note
• H(s) is commonly referred as the system function or transfer function.
• For s = j, H(s) is frequency response of LTI system.
• Equation (31) is valid only when initial condition are zero.

The Transfer Function and Differential Equation System Description:

The transfer function may be related directly to the differential-equation description of an LTI system by
using the bilateral Laplace transform. The relationship between the input and output of an Nth order LTI system
is described by the differential equation,
N
dk y(t ) M
dk x(t )
 ak dtk
=  bk dt k
k 0 k 0

Taking the Laplace transform of the above equation, we obtain

N N
 ak sk Y(s) =  bk sk X( s)
k 0 k 0

M
 bk sk
Y( s) k 0
H( s)  = N
X( s)
 ak sk
k0

H(s) is a ratio of polynomials in s and is thus termed a rational transfer function.

4.5.1 Step Response and Impulse Response

The h(t) is impulse response of system, that is output when (t) is input to the system.
Laplace transform
h(t )   H( s)

The step response s(t) is output when input is u(t), so

s( t ) = u(t)  h(t)
H( s)
 S(s ) = ...(32)
s
or, H(s) = s(S(s)) ...(33)

4.5.2 Causality of LTI System

For a causal LTI system the impulse response h(t) = 0 for t < 0, thus the signal should be right sided.
In case of right sided signal which is of infinite duration, then ROC is the region in the s-plane to the right of
right most pole.
If H(s) is rational, then we can determine whether the system is causal simply by checking to see if its
ROC is a right-half plane, that is for a system with a rational system function, causality of the system is equivalent
to the ROC being the right-half plane to the right of the right most pole.

GATE MASTERS PUBLICATION

 laplace Transform 173

In an exactly analogous manner, we can deal with the concept of anticausality. For an anticausal LTI
system, the impulse response h(t) = 0 for t > 0 and is thus left-sided. We know that if a signal is left-sided and
of infinite duration, then the ROC is the region in the s-plane to the left of the left most pole. Consequently, the
ROC associated with the system function for an anticausal system is a left-half plane. However, the converse of
this statement is not necessarily true.

Study Note
• In the above explanation we were talking about rational H(s), but if H(s) is not rational that is for example
e5s/(s + 5),  > –5 is irrational function and has right sided ROC but it will be non-causal as here h(t) will
be e–5(t + 5) u(t + 5) that is non zero for t < 0.
• In case of rational H(s) the system will be causal if ROC is  > Real part of right most pole.

4.5.3 Stability of LTI System

An LTI system is stable if the impulse response h(t) is absolutely integrable, that is


 h(t ) dt <  ...(34)



Since, H(s) is transfer function of the system, that is Laplace transform of h(t). So ROC of H(s) is
defined as values of  where h(t) e–t is absolutely integrable. That is


 h(t ) et dt <  ...(35)



If we compare above two equations then we get that if signal h(t) is absolutely integrable then ROC of
it’s laplace transform will contain the j-axis. Now if H(s) is given and we know it’s ROC then to find that h(t)
is absolutely integrable the ROC of H(s) must contain  = 0 or j-axis or imaginary axis. Thus to have stable LTI
system the ROC of transfer function must contain j-axis.

REMEMBER Suppose in any question for example it is given that x(t) e–4t is absolutely integrable. If X(s) is
Laplace transform of x(t) then it’s ROC is values of  where x(t) e–t is absolutely integrable. So
if we compare the given information with definition of ROC we can say that  = 4 is present in
ROC.

4.5.4 Stability of a Causal LTI System

If system is causal as well as stable and H(s) is a rational function then it must satisfy two conditions
together:
(i ) ROC should be  > Real part of right most pole.
( ii ) ROC should certain j-axis.
If we want both conditions then it means that right most pole of H(s) lie left of j-axis. That is all poles
of H(s) should lie in left of j-axis.

Study Note
• System with rational H(s) will be causal and stable if and only if all poles lie in left of j-axis.
• Causal and stable system will have ROC,  > Real part of right most pole.

GATE MASTERS PUBLICATION

174 Signals and Systems

Example 4.19

The unit step response of an LTI system is s ( t ) = 2 e –t u ( t ). Determine its system function
and the impulse response.
Solution 4.19
Given that the unit step response,
s (t ) = 2 e–t u(t)
L[s(t)] = L[2 e–t u(t)]
2
S (s ) =
s 1
The relationship between L[h(t)] = H(s) and L[s(t)] = S(s) is given by
2s 2s  2  2 2( s  1) 2
h(s) = s S(s) =   
s 1 s 1 s 1 s 1
2
H (s ) = 2 
s 1
The inverse transform of this equation yields,
h(t) = 2(t) – 2 e–t u(t)

Example 4.20

For the following check whether corresponding system is causal and stable.
1
( a ) H1( s )  2
, 3
s s6

1
( b ) H 2 (s )  ,   2
s2 s 6

1
( c ) H3 ( s)  2
, 2 3
s s6

Solution 4.20
( a ) Since transfer function is rational and ROC is  > Real part of right most pole. So it is causal but since
ROC do not contain j-axis, so unstable.
H1(s)  Causal, unstable system
( b ) In this system ROC is  < –2, ROC is  < Real part of left most pole. So system is anticausal and
unstable as ROC do not contain j-axis.
( c ) In this system, ROC is –2 <  < 3, since ROC is a strip in s-plane thus time domain signal will be both
sided signal this it is non-causal, ROC contain j-axis so it is stable.

Example 4.21

For the following system function, check whether the corresponding LTI system is causal
and stable,
1
H (s )  2
,   2
s  5s  6

GATE MASTERS PUBLICATION

 laplace Transform 175

Solution 4.21
Since H(s) has all poles in left side of j and  > –2 so it is stable and causal.
1 1 1
H (s ) = 2
 
s  5s  6 s  2 s  3
Since both poles are in left of ROC so both give right sided signal.
 h(t) = e–2t u(t) – e–3t u(t)

Example 4.22

4s2  15s  8
Find the inverse Laplace transform of H( s)  , assuming that (a) h ( t ) is causal
( s  2)2 ( s  1)

and (b) the Fourier transform of h ( t ) exists, i.e., h ( t ) is absolutely integrable.

Solution 4.22
4s2  15s  8
Given that, H (s ) =
( s  2)2 ( s  1)
Using partial fraction expansion, we obtain
1 2 3
H (s ) =  2

s  2 ( s  2) s 1
The system has a double pole at –2 and a single pole at 1.
( a ) For this system to be causal, the ROC lies to the right of the right most pole, i.e.  > 1. Since the ROC,
 > 1, is to the right of the right most pole, all the poles correspond to causal (right-sided) signals.
Therefore,
1
e2t u(t ) 
s2
1
2t e2t u( t ) 
( s  2)2

3
3 et u(t ) 
s 1
and hence we obtain, h(t) = e–2t u(t) + 2 t e–2t u(t) + 3 et u(t)
( b ) For h(t) to be Fourier transformable, i.e., for h(t) to be absolutely summable (stable system), the ROC
must include the j-axis. To include the j-axis, its ROC is in the region –2 <  < 1. The pole of the
first and second term is at –2. The ROC lies left of pole s =1 thus it correspond to left sided signal and
ROC lie right of the pole s=2 thus this pole give right sided signals. Therefore,
1
e2t u(t ) 
s2
1
2t e2t u(t ) 
( s  2)2

3
3 et u( t ) 
s 1
and hence we obtain, h(t) = e–2t u(t) + 2 t e–2t u(t) –3 et u(–t)

GATE MASTERS PUBLICATION

176 Signals and Systems

Example 4.23

Consider a LTI system for which input is x ( t ) and output is y ( t ) are related by differential
equation,
d2 y ( t ) dy ( t )
  2 y ( t )  x( t )
dt 2 dt
Determine h ( t ) for each of the following case:
1 . The system is stable.
2 . The system is causal.
3 . The system is neither stable nor causal.
Solution 4.23
Taking the Laplace transform of differential equation we get,
s2 Y(s) – sY(s) – 2Y(s) = X(s )
Y( s) 1 1/ 3 1/ 3
  H( s) = 2  
X( s) s  s  2 ( s  2) ( s  1)
So, the system has two poles, –1 and 2.
( a ) For stable system, ROC must contain j-axis, so ROC will be –1 <  < 2. So pole –1 will give right
sided signal and 2 will give left sided signal.
1 1
 h(t) =  e2t u( t )  et u(t )
3 3
( b ) For causal system, ROC must be right of right most pole i.e.  > 2, so both poles give right sided signal,
1 2t 1
e u(t )  et u(t )
h( t ) =
3 3
( c ) For neither stable or causal, ROC must not be right of right most pole and must not certain j-axis, so
ROC is  < –1. So both poles will give left sided signal. So
1 1
h(t) =  e2t u( t )  et u( t )
3 3

Example 4.24

Find the response y ( t ) of a non-causal system with the transfer function,

1
H(s) = , 1
s 1
to the input x ( t ) = e –2 t u ( t ).
Solution 4.24
Given that, x ( t ) = e–2t u(t)
1
L[x(t)] = X( s)  ,   2
s2
1
we knot that, Y(s) = X( s) H( s) 
( s  1) ( s  2)
The ROC of Y(s) is in the region –2 <  < 1. Using partial fraction expansion, we obtain
1 / 3 1 / 3
Y(s) =  , 2   1
s 1 s  2

GATE MASTERS PUBLICATION

 laplace Transform 177

Taking the inverse Laplace transform, we obtain

1 t
y(t) = [e u( t )  e2t u(t )]
3

4.5.5 System Interconnection

Cascade Connection
For two LTI systems [with h1(t) and h2(t), respectively] in cascade [Fig. 4.4(a)], the overall impulse
response h(t) is given by
h(t) = h1(t)  h2(t) ...(36)

x(t) h1(t) h2(t) y(t) x(t) h(t) y(t)

h(t) = h1(t)* h2(t)....

(a)

X(s) H1(s) H2(s) Y(s) X(s) H(s) Y(s)

H(s) = H1(s) H2(s)....

(b)

Fig. 4.4 : Two systems in cascade (a) Time-domain representation (b) s-domain representation
Thus, the corresponding system functions are related by the product,
H(s) = H1(s) H2(s) ...(37)
This relationship is illustrated in Fig. 4.4 (b).

Parallel Connection
The impulse response of a parallel combination of two LTI systems [Fig. 4.5(a)] is given by,
h(t) = h1(t) + h2(t)
Thus, H(s) = H1(s) + H2(s) ...(38)
This relationship is illustrated in Fig. 4.5 (b).

h1(t)

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

h(t) = h1(t) + h2(t)

h2(t)

(a)

H1(s)

X(s) Y(s) X(s) H(s) Y(s)

H(s) = H1(s) + H2(s)

H2(s)

(b)

Fig. 4.5 : Two systems in parallel

(a) Time-domain representation (b) s-domain representation

GATE MASTERS PUBLICATION

 178 Signals and Systems

Feedback Connection
When the output is feedback to the input, as shown in Fig. 4.6, the overall transfer function
H(s) = Y(s)/X(s) can be computed as follows.
+ E(s)
X(s) H1(s)
–
H1( s)
= X(s) H( s)  Y(s)
1  H1( s) H2 (s)

H2(s)

Fig. 4.6 : Elementary connection of block diagram

The input to the adder are X(s) and –H2(s) Y(s). Therefore, E(s), the output of the adder, is
E(s) = X(s) – H2(s) Y(s)
and the output is given by,
Y(s) = H1(s) E(s) = H1(s) [X(s) – H2(s) Y(s)]
Y(s) = H1(s) X(s) – H1(s) H2(s) Y(s)
Y(s) [1 + H1(s) H2(s)] = H1(s) X(s)
Y( s) H1( s)
 H( s) =
X( s) 1  H1( s) H2 ( s)
Therefore, the feedback loop can be replaced by a single block with the transfer function shown above.
The transfer function H1(s) is the forward path transfer function and H2(s) is the feedback path transfer function.

4.5.6 Response of Input (eCt ) from LTI Continuous Time System

When input is of format eCt where C is any constant value, we can see that Laplace transform of eCt do
not exist. [eCt = eCt u(t) + eCt u(–t) and then ROC will be  for this case]. So we cannot use Y(s) = H(s)X(s).

LTI
eCt System
Laplace
h(t )   H( s) Output
Transform

 y ( t ) = h(t)  eCt ...(39)


 st
and H(s ) =  h(t ) e dt ...(40)


Now, if we look at equation (39), we get


C( t   )
y(t) =  h( ) e d



= eCt  h( ) e
C
d


Ct
y ( t ) = e H (s ) ...(41)
s C

GATE MASTERS PUBLICATION

 laplace Transform 179

Study Note
So whenever input is eCt for – < t <  then output is equal to
y ( t )= eCt. Value of Laplace transform at s = C
Using above result we get,

ej 0t H(s) e j0t  H( s)

s  j 0

e–j 0 H(s) e j 0t  H( s)

s   j0

When h(t) is real then H(s) is rational function and H (s ) s  j   (H (s ) s   j  ) , that is for real h(t) we
0 0

have H ( j o )  H (  j o )* . This will be proved in Fourier transform chapter. When nothing is given about h(t)
then we can assume that h(t) is real.

Now, consider a case that LTI system with impulse response h(t) has input cos(0t) [we cannot find
Laplace transform of cos(0t)] then to find output,

e j0t  e j0t
cos(0t) =
2
so, if H(s) is transfer function of the system then output of cos(0t) will be

e j 0t  H (s ) s  j   e  j 0t H (s ) s  j 
0 0
cos(0t ) 
2

e j0t  H( j 0 )  e j0t H(  j 0 )
cos( 0 t ) 
2
Assuming that h(t) is real then
Let, H(j0) = H ( j o ) e j  , where   H ( j 0 )

H(–j0) = H ( j o )*

Thus, H(-j0) = H ( j o ) e  j 

e j0t H( j 0 ) e j  e j0t H( j 0 ) e j
Thus, output =
2

= H( j0 ) cos(0 t  )

Thus, cos(0t) H(s) H(j 0)cos(0t)

or
H( j 0 ) cos( 0 t H( j 0 ))

Similarly, sin(0t) H(s) H(j 0)sin(0t)

or
H( j0 ) sin( 0 t  H( j0 ))

GATE MASTERS PUBLICATION

 180 Signals and Systems

4.6 The Unilateral Laplace Transform

The unilateral (or one-sided) Laplace transform X1(s) of a signal x(t) is defined as

 st
X 1( s ) =  x (t ) e dt ...(42)

0

The lower limit of integration is chosen to be 0– (rather than 0 or 0+) to permit x(t) to include (t) or its
derivatives. Thus, we note immediately that the integration from 0– to 0+ is zero except when there is an
impulse function of its derivative at the origin. The unilateral Laplace transform ignores x(t) for t < 0. Since
x(t) in equation (42) is a right-sided signal, the ROC of X1(s), is always of the form  > max, that is, a right half-
plane in the s-plane.
In case of unilateral Laplace transform it is not necessary to define ROC because it is understood that
signal will be right sided.
If we want to find unilateral Laplace transform of a signal x(t) then it simply means to find bilateral
Laplace transform of x(t) u(t).
For example:
If, x ( t ) = e–a(t + 1) u(t + 1) then,
it’s bilateral Laplace transform is,
1
X (s ) = es ,  > Real{–a}
(s  a)
but to find unilateral Laplace transform we multiply x(t) with u(t) i.e. now,
xu ( t ) = e–a(t + 1) u(t + 1)  u(t)
That is now, xu ( t ) = e–a(t + 1) u(t) = e–a e–at u(t)
Now, bilateral Laplace of xu(t) will be unilateral Laplace transform of x(t)

 1 
 e a e at u(t )   e a 
 s  a 

 e a 
Thus, unilateral Laplace transform is   .
sa

Example 4.25

Find unilateral transform of

( a ) x ( t ) = e –2 t u ( t + 1)
( b ) x ( t ) = ( t + 1) + ( t ) + e –2( t + 3) u ( t + 1)
Solution 4.25
We can use equation (42) or find bilateral Laplace transform of x(t)  u(t). That is for
(a) x(t ) = e–t u(t + 1)

t
Since, X (s ) =  x(t ) e dt

0

  1, t  1
= e
2t
u(t  1) e st dt  u(t  1)   
 0, t  1
0

GATE MASTERS PUBLICATION

 laplace Transform 181


2 t
= e e st dt
0

1
= ,   2
s2
(b) x ( t ) = (t + 1) + (t) + e–(2t+ 3) u(t + 1)

Since, X (s ) =  x(t ) e st dt
0


(2 t  3)
=  [(t  1)  (t )  e u(t  1)] e st dt
0

Here, (t + 1) is at t = –1, (t) is at t = 0, u(t + 1) starts from-1


(2t  3) 1
 X (s ) =  ((t )  e ) e st = 1 
s2
0

s3
= ,   2
s2

4.6.1 Properties of Unilateral Laplace Transform

Most of the properties of the unilateral Laplace transform are the same as for the bilateral transform.
The unilateral Laplace transform is useful for calculating the response of a causal system to a causal input when
the system is described by a linear constant-coefficient differential equation with nonzero initial conditions. The
basic properties of the unilateral Laplace transform that are useful in this application are the time-differential
and time-integration properties which are different from those of the bilateral transform. They are presented in
the following.

1. Linearity:
Unilateral Laplace transform
If x 1(t )   X 1(s )
Unilateral Laplace transform
and x 2 (t )   X 2 (s )
Unilateral Laplace transform
then, ax 1(t )  bx 2 (t )   aX 1(s )bX 2 (s ) ...(43)

2. Time Scaling
Unilateral Laplace transform
If x (t )   X (s )

Unilateral Laplace transform 1 s 

then, x (at )   X   a 0 ...(44)
a a 

Note: This will be invalid if a is negative, because for negative value of a signal will become left sided.

3. Shifting in the s-domain

Unilateral Laplace transform
If x (t )   X (s )

then, e s0t x (t ) 

Unilateral Laplace transform
 X (s  s 0 ) ...(45)

GATE MASTERS PUBLICATION

 182 Signals and Systems

4. Conjugation
Unilateral Laplace transform
If x (t )   X (s )

Unilateral Laplace transform

then, x  (t )   X  (s  ) ...(46)

5. Differential in s-domain
Unilateral Laplace transform
If x (t )   X (s )

Unilateral Laplace transform d X (s )

then, t x (t )   ...(47)
ds

6. Convolution
If x 1( t ) = x2(t) = 0, for all t < 0, and if
Unilateral Laplace transform
x 1(t )   X 1(s )
Unilateral Laplace transform
and x 2 (t )   X 2 (s )
Unilateral Laplace transform
then, x1(t ) x2(t )   X1(s ) X2(s ) ...(48)
It is important to note that the convolution property for unilateral transform applies only if the signals
x1(t) and x2(t) are both zero for t < 0. That is, while we have see that the bilateral Laplace transform of
x1(t)  x2(t) always equals to the product of the bilateral transform of x1(t) and x2(t), the unilateral transform
of x1(t)  x2(t) in general does not equal to the product of the unilateral transform if either x1(t) or x2(t) is
nonzero for t < 0.

7. Integration in the s-domain


If x(t) satisfies the condition of existence  x(t ) et dt  , and the limit of x(t)/t, as t approaches 0
0

from the right, exists, and

Unilateral Laplace transform
x (t )   X (s )


then,
x (t ) Unilateral Laplace transform ...(49)
   X (s ) ds
t s

8. Time Shifting
Unilateral Laplace transform
If x (t )   X (s )

then, Unilateral Laplace transform

x (t  t 0 )   X (s ) e st0 ...(50)
Observe that x(t) starts at t = 0, and therefore, x(t – t0) starts at t = t0. So, this property may be
restated as follows.
Unilateral Laplace transform
If x (t ) u (t )   X (s )

then, Unilateral Laplace transform

x (t  t 0 ) u (t  t 0 )   X (s ) e st0 ...(51)

Note: This property is valid if signal is shifted right, that is t0 > 0.

GATE MASTERS PUBLICATION

 laplace Transform 183

9. Differentiation in Time-domain
Unilateral Laplace transform
If x (t )   X (s )

d Unilateral Laplace transform

then, x (t )  s X (s )  x (0 ) ...(52)
dt
Proof : From definition,

 st
X (s ) =  x(t ) e dt
0

d d
Let Y(s) is unilateral Laplace transform of x(t ), and y(t )  x(t ).
dt dt

d
 Y(s) =  x(t ) e st dt Applying integration by parts,
dt
0



= x(t ) e st   ( s) x(t ) e
 st
dt
0

0


= [0  x(0 )]  s  x(t ) e st dt
0

= s X(s) – x(0–)

Thus, d 2 x (t ) Unilateral Laplace transform

   s (Y (s ))  y (0  )
dt 2

2  d
= s X( s)  s x(0 )  x(t )
dt t  0

General formulae will be

d n x (t ) Unilateral Laplace transform  n n 

n
   s X ( s )   s n  k x ( k 1) (0  ) ...(53)
dt  k 1 

10. Integration in the time domain

Unilateral Laplace transform
If x (t )   X (s )
0

t
X (s )
 x ( ) d 
then, Unilateral Laplace transform  ...(54)
 x ( ) d  

S

S


t
Unilateral Laplace transform X (s )
and  x ( ) d  

s ...(55)
0

Proof:
t
Let, y(t) =  x( ) d


d
that is, x(t ) = y(t )
dt

GATE MASTERS PUBLICATION

 184 Signals and Systems

Let X(s) and Y(s) are unilateral Laplace transform of x(t) and y(t). So using equation (52) we get
X(s ) = s Y(s) – y(0–)

X (s ) y (0 )
 Y(s) = 
s s
t
when, y(t) =  x( ) d then,


0
y (0 – ) =  x( ) d


0

t  x ()d 
X (s ) 
 Unilateral Laplace transform
 x ( ) d  

s

s


t
and when, y(t) =  x( ) d then,

0

y (0 – ) = 0
t
Unilateral Laplace transform X (s )
  x ( ) d  

s
0
0

11. Initial value theorem

The initial value theorem allows us to determine the initial value x(0+) of x(t) directly from X(s). The
theorem states that if x(t) = 0, for t < 0, and if it contains no impulses or higher order singularities at the
origin, then
x(0+ ) = Lim s X( s) ...(55)
s

The initial value theorem does not apply to rational functions X(s) in which the order of the numerator
polynomial is greater than or equal to that of the denominator polynomial.
Proof : We know that,
d Unilateral Laplace transform
x (t )   s X (s )  x (0  )
dt

d
That is, s X(s) – x(0–) =  x(t ) e st dt
dt
0

Now we put Lim in above equation. As s   then RHS will be zero because of e–st term.
s

So, Lim [s X( s)  x(0 )] = 0

s

 Lim s X( s) = x(0 – )
s

Remember: x(0–) = x(0+) because no impulses exist in x(t) because X(s) is proper function.

GATE MASTERS PUBLICATION

 laplace Transform 185

Example 4.26

Use the initial value theorem to find the initial value of the signal corresponding to the
L aplace transform,
s1
X(s) =
s( s  2)
Solution 4.26
From the initial value theorem, we have
 s 1   s 1 
x(0 )  Lim s X( s) = Lim s    Lim  
s s  s( s  2)  s    s  2 

 1  (1  s) 
x(0+ ) = Lim   1
s  1  (2 / s) 

12. Final value theorem

The final value theorem allows us to determine the final value x() of x(t) directly from X(s). The
theorem states that if x(t) = 0, for t < 0, and if it contains no impulses or higher order singularities at the origin,
then
Lim x(t ) = x(  )  Lim s X( s)
t t

The final value theorem is applicable only if the poles of s X(s) are in the left half of the s-plane, with at
most a single pole at s = 0. If X(s) has a pole in the right half of the s-plane, x(t) contains an exponentially
growing term and x() does not exist. If there is a pole on the imaginary axis, then x(t) contains an oscillating
term and x() does not exist. However, if there is a pole at the origin, then x(t) contains a constant term, and
hence x() exists and is a constant.
Proof : Taking the unilateral Laplace transform of dx(t)/dt, we obtain

 dx(t )  dx(t )  st
Lu   =  e dt
 dt  0  dt

Using differentiation in the time-domain property, we obtain


dx(t )  st
sX(s) – x(0–) =  e dt
 dt
0


Lim[sX( s)  x(0 )] = Lim dx(t )  st 
 e dt  x(t ) 0
s 0 s0 dt

0

Lim[sX( s)  x(0 )] = x() – x(0–)

s 0

Lim s X( s) = x ()
s0

GATE MASTERS PUBLICATION

 186 Signals and Systems

Example 4.27

Determine the initial and final values of x ( t ) if its Laplace transform is given by:

10(2 s  3)
X(s) =
s( s2  2 s  5)

Solution 4.27
From the initial value theorem, we have
10(2s  3)
x(0+ ) = Lim s X( s)  Lim s
s s s( s2  2s  5)

= Lim  10(2s  3)   6
s  
 s2  2s  5 
From the final value theorem, we have
 10(2s  3)   10(2s  3) 
Lim x (t )  x () = Lim sX (s )  Lim s    Lim  2 6
t  2
s 0 s 0
 s (s  2s  5)  s  0  s  2s  5 

Example 4.28

Find the value of output at steady state when input to system is x ( t ) and transfer function is
H(s),
s3
H(s) =
s2  4 s  5
When,
( a ) x(t) = u(t)
( b ) x ( t ) = ( t )
Solution 4.28
Now, steady state value is y(), thus
(a) Y(s) = H(s)  X(s)
s3 1 ( s  3)
= 2
 = 2
s  4s  5 s s( s  4s  5)
[Since all poles of sY(s) are in left of j-axis, so we can apply final value theorem]
3
Y() = Lim s Y( s) 
s0 5
(b) Y(s) = H(s)  X(s)
s3
= 2
1
s  4s  5
 Here also poles of sY(s) lie in left of j-axis so we can apply final value theorem,
Y() = Lim s Y( s)  0
s0

Remember: If in any question system transfer function H(s) is given but it’s ROC is not given then assume that
system is causal and assume ROC accordingly.

GATE MASTERS PUBLICATION

 laplace Transform 187

4.7 Solution of Differential and Integer Differential Equations

One of the most common applications of the Laplace transform is to solve linear constant coefficient
differential equations. Such equations are used to model continuous-time LTI systems. Solving of these equations
depends on the differentiation property of the Laplace transform. Procedure that should be followed to solve is
1. For a given set of initial conditions, take the Laplace transform of both sides of the differential
equation to obtain an algebraic equation in Y(s).
2. Solve the algebraic equation for Y(s).
3. Take the inverse Laplace transform to obtain y(t).

Example 4.29

Solve the second-order linear differential equations,

d2 y ( t ) dy ( t ) dx( t )
2
5  6y(t )   x( t )
dt dt dt
dy ( t )
for the initial conditions y (0 – ) = 2 and  y (0  )  1 and the input x ( t ) = e –4 t u ( t ).
dt t  0 

Solution 4.29
If x(t) = e–4t u(t), then x(0–) = 0 and
1
X (s ) =
s4
Now, consider the given differential equation,
d2 y ( t ) dy(t ) dx(t )
5  6y(t ) =  x(t )
dt 2 dt dt
Taking the Laplace transform of the above equation, we obtain

[s2 Y(s) – sy(0–) – y (0 ) ] + 5[sY(s) – y(0–)] + 6Y(s) = [sX(s) – x(0–)] + X(s)



[s2 Y(s) – 2s – 1] + 5[sY(s) – 2] + 6Y(s) = [sX(s) – 0] + X(s)

(s2 + 5s + 6) Y(s) – (2s + 11) = X(s) (s + 1)
s 1
(s2 + 5s + 6) Y(s) = (2s  11) 
s4
2s2  20s  45
Y(s) =
( s  2) ( s  3) ( s  4)
Expanding the RHS into partial fractions yields,
13 / 2 3 3/2
Y(s) =  
s2 s3 s4
Taking the inverse Laplace transform of the above equation, we obtain
13 3 
y ( t ) =  e2t  3 e3t  e4t  u(t )
2 2 

Study Note
In the above example we can see that output Y(s) has no ROC but since system is causal and input is also causal so
output is always right sided.

GATE MASTERS PUBLICATION

 188 Signals and Systems

4.7.1 Zero-Input Response and Zero-State Response

The Laplace transform method for solving differential equations offers a clear separation between the
zero-input response of the system to initial conditions and the zero-state response of the system associated with
the input. The zero-input response represents the system output when the input is zero, i.e., x(t) = 0, and thus
is the result of internal system conditions (such as energy storage, initial conditions) alone. In contrast, the
zero-state response represents the system response to the external input x(t) when the initial conditions are
zero. The total response is defined as,
Total response = Zero-input response + Zero-state response ...(58)
Natural response and Force response: In total response, the response due to the poles of the
system is called natural response of the system. The response due to the poles of the input is called the forced
response of the system.

Example 4.30

Use the unilateral Laplace transform to determine the output of a system represented by the
differential equation,

d2 y ( t ) dy ( t ) dx( t )
2
5  6 y (t )   6 x( t )
dt dt dt
in response to the input x ( t ) = u ( t ). Assume that the initial conditions on the system are
.
y (0 – ) = 1 and y (0 ) = 2. Identify the zero-state response y zs ( t ), of the system, and the zero
input response y zi ( t ).
Solution 4.30
If x(t) = u(t), then x(0–) = 0 and X(s) = 1/s. Now, consider the given differential equation,
d2 y(t ) dy(t ) dx(t )
2
5  6 y(t )   6 x(t )
dt dt dt
Taking the Laplace transform of the above equation, we obtain
.

[s2 Y(s) – sy(0–) – y (0 ) ] + 5[sY(s) – y(0–)] + 6Y(s) = [sX(s) – x(0–)] + 6X(s)
[s2 Y(s) – s – 2] + 5[sY(s) – 1] + 6Y(s) = [sX(s) – 0] + 6X(s)
(s2 + 5s + 6) Y(s) – (s + 7) = X(s) (s + 6)
(s2 + 5s + 6) Y(s) = ( s  7)

 X( s) ( s  6)

Initial condition terms Input terms

s7 s6
Y(s) = 2
 2
s 
 5s 
6 s( s  5s  

6)
Zero-input component Zero-state component

s7 s6
Y(s) = 
( s  2) ( s  3) s( s  2) ( s  3)
Using partial fraction expansion, we obtain
 5 4  1 2 1 
Y(s) =      
s  2 s  3 s s  2 s  3

GATE MASTERS PUBLICATION

 laplace Transform 189

The inverse transform of this equation yields,

y(t) = [5e–2t u(t) – 4e–3t u(t)] + [u(t) – 2e–2t u(t) + e–3t u(t)]
y(t) = [5 e2t  4 e3t ] u(t )  [1  2 e2t  e3t ] u(t )
   
Zero-input response or natual response Zero-state response or forced response

y(t) = yzi(t) + yzs(t)

where the zero-input response is,
yzi(t ) = [5e–2t – 4e–3t] u(t)
and the zero-state response is
yzs(t) = [1 – 2e–2t + e–3t] u(t)

4.8 System Realization

We now develope a systematic method for realization (or implementation) of an arbitrary Nth-order
transfer function. The most general transfer function with M = N is given by,
b0 sN  b1sN  1  ...  bN  1 s  bN
H(s ) =
sN  a1 sN  1  ...  aN  1 s  aN
Since realization is basically a synthesis problem, there is no unique way of realizing a system. A given
transfer function can be realized in many different ways. A transfer function H(s) can be conveniently represented
in block diagram from using the basic building blocks representing the integrators of differentiators, the multiplier,
the adder, and the pick-off node as shown in Fig. 4.7.

X( s) A
X(s) 1/s Y( s)  X(s) Y(s) = AX(s)
s
(a) (b)

+ Y(s) = X1(s) + X2(s)

X1(s) X(s) X(s)

X(s)
X2(s)
(c) (d)

Fig. 4.7 : Basic building block representing

(a) Integrator, (b) Multiplier, (c) Adder and (d) Pick-off

4.8.1 Direct Form I Realization

An Nth order transfer function is characterized by 2N + 1 unique coefficients and, in general, requires
2N + 1 multipliers and 2N two input adders for implementation. Structures in which the multiplier coefficients
are precisely the coefficients of the transfer functions are called direct form structures.
b0 s3  b1s2  b2 s  b3
H(s ) =
s3  a1s2  a2 s  a3

b0  ( b1 / s)  ( b2 / s2 )  ( b3 / s3 )
H(s ) =
1  ( a1 / s)  ( a2 / s2 )  ( a3 / s3 )

 b b b  1 
We can express H(s) as, H(s) =  b0  1  2  3   
 s 
s s  1  a1 / s  ( a2 / s )  ( a3 / s ) 
2 3
  
H1( s ) H2 ( s )

GATE MASTERS PUBLICATION

 190 Signals and Systems

We can realize H(s) as a cascade of transfer function H1(s) followed by H2(s), as depicted in Fig. 4.8 (a),
where
W( s) b b b
H 1( s ) =  b0  1  22  33 ...(59)
X( s) s s s

 b b b 
W(s) =  b0  1  22  33  X( s) ...(60)
 s s s 
Y( s ) 1
and H 2( s ) =  ...(61)
W( s ) 1  ( a1 / s )  ( a 2 / s2 )  ( a3 / s3 )

a1 a2 a3 
Y(s) = W( s)      Y( s) ...(62)
 s s2 s3 
We shall first realize H1(s) given by equation (59). Equation (60) shows that the output W(s) can be
synthesized by adding the input b0 X(s) to b1X(s)/s, b2X(s)/s2, and b3X(s)/s3. Because the transfer function of an
integrator is 1/s, the signals X(s)/s, X(s)/s2, and X(s)/s3 can be obtained by successive integration of the input
x(t). The left-half section of Fig. 4.7 (b) shows the realization of H1(s).
W(s)
X(s) H1(s) H2(s) Y(s)

(a)

X(s) b0 W(s)
Y(s)

1 1
s s

b1 –a 1

1 1
s s

b2 –a 2

1 1
s s
b3 –a 3

(b)
Fig. 4.8 : (a) Realization of a function in two steps and
(b) Direct form I realization of a third- order continuous-time LTI system
We next consider the realization of H2(s) given by, equation (61). Equation (62) shows that the output
Y(s) can be synthesized by subtracting a1Y(s)/s, a2Y(s)/s2, and a3Y(s)/s2 from W(s). To obtain signals Y(s)/s,
Y(s)/s2, and Y(s)/s2, we assume that we already have the desired output Y(s). Successive integration of Y(s)
yields the needed signals Y(s)/s, Y(s)/s2, and Y(s)/s3. The right-half section of Fig. 4.8 (b) shows the realization
of H2(s).
We can generalize this procedure, known as the direct form I realization, for any value of N. This
procedure requires 2N integrators to realize an Nth-order transfer function. This realization is noncanonic since
it employs six integration to implement a third-order transfer function. A realization is canonic if the number of
integrators used in the realization is equal to the order of the transfer function realized. Thus, canonic realization
has no redundant integrators.

GATE MASTERS PUBLICATION

 laplace Transform 191

4.8.2 Direct Form II (or Canonic) Realization

In Direct form I, we realize H(s) by implementing H1(s) followed by H2(s), as shown in Fig. 4.9 (a). We
can also realize H(s), as shown in Fig. 4.9, where H2(s) is followed by H1(s). This procedure is known as the
direct form II realization.
V(s)
X(s) H2(s) H1(s) Y(s)

(a)

Study Note
The direct form II realization Fig. 4.9 (b) implements zeros first [the left-half section represented by H1(s)] followed
by realization of poles [the right-half section represented by H2(s)] of H(s). In contrast, the direct form II realization
implements poles first followed by zeros.

Again consider for simplicity a third-order system characterized by a transfer function,

b0 s3  b1s2  b2 s  b3
H(s ) =
s3  a1s2  a2 s  a3

b0  ( b1 / s)  ( b2 / s2 )  ( b3 / s3 )
H(s ) =
1  ( a1 / s)  ( a2 / s2 )  ( a3 / s3 )

 1  b1 b2 b3 
We can express H(s) as, H(s) =  2 3   b0  s  2  3 
 1  ( a1 / s)  ( a2 / s )  ( a3 / s )  

s s 
H2 ( s ) H1( s )

We can realize H(s) as a cascade of transfer function H2(s) followed by H1(s), as depicted in Fig. 4.8(a),
where,

V( s) 1
H2(s) =  ...(63)
X( s ) 1  ( a1 / s)  ( a 2 / s2 )  ( a3 / s3 )

 a1 a2 a3 
V(s) = X( s)      V( s) ...(64)
 s s2 s3 

Y (s ) b b b
and H 1( s ) =  b0  1  22  33 ...(65)
V (s ) s s s

 b b b 
Y(s) =  b0  1  22  33  V( s) ...(66)
 s s s 
The left-half section of Fig. 4.9(b) shows the realization of H2(s) and the right-half section shows the
realization of H1(s).

GATE MASTERS PUBLICATION

 192 Signals and Systems

X(s) V(s) b0 Y(s)

1 1
s s

–a1 b1
1 1

1 1
s s
–a2 b2
2 2

1 1
s s
–a3 b3
3 3

(b)

X(s) V(s) b0
Y(s)

1
s

–a 1 b1

1
s
–a 2 b2

1
s
–a 3 b3

(c)

Fig. 4.9 : Direct form II realization of a third-order LTI system

We observe that in Fig. 4.9 (b), the signal variables at nodes 1 and 1’ are the same, and hence the two
top integrators can be shared. Likewise, the signal variables at nodes 2 and 2’ are the same, which permits the
sharing of the two middle integrators. Following the same argument, we can share the integrators, leading to
the final structure shown in Fig. 4.9 (c).

Study Note
This implementation halves the number of integrators to N, and is thus more efficient in hardware utilization. This
is the direct form II realization. This realization is canonic since it employs N integrators to implement an Nth order
transfer function.

GATE MASTERS PUBLICATION

 laplace Transform 193

Example 4.31

4 s  28
Find the canonic realization of the transfer function, H( s )  2
.
s  6s  5
Solution 4.31
Y( s) V( s) Y( s)
Given that, H (s ) = 
X( s) X( s) V( s)

 28  4
 2 
4s  28
 s   s 1   4 28 
= 2    2 
s  6s  5 1   6   5  1  (6 / s)  (5 / s )  s
2
s 
  2    
s s H2 ( s ) H1( s )

V( s) 1
where, H 2( s ) = 
X( s) 6  5 
1     2 
 s s 

V(s) = X( s)  6 V( s)  5 V(2s)
s s X(s)

Y( s) 4 28
and H 1( s ) =  
V( s) s s2 1
s

Y(s) = 4 V (s )  28 V (s ) –6 4
Y(s)
s s2
4s  28 1
The given transfer function H( s)  is of the second s
2
s  6s  5
–5 28
order therefore, we need only two integrators for its realization.
(d)

4.8 Laplace Transform of Causal Period Signal

The Laplace transform of a causal periodic signal can be determined from the knowledge of the Laplace
transform of its first cycle (period). Consider a causal periodic signal x(t) with period T0 as shown in Fig. 4.10.
x(t)

x 1(t) x 2(t) x 3(t)

t
0 T0 2T0 3T0 4T0

Fig. 4.10 : Causal periodic signal

Let x1(t), x2(t), x3(t)..., be the signals representing the 1st, 2nd, 3rd, ...., cycles of a causal periodic
signal x(t). Therefore, the causal periodic signal x(t) can be written as,
x(t ) = x1(t) + x2(t) + x3(t) + x4(t) +....
= x1(t) + x1(t – T0) + x1(t – 2T0) + x1(t – 3T0) +....
Assume x1(t)  X1(s). Using the time-shifting property, the Laplace transform of the above equation
becomes,

GATE MASTERS PUBLICATION

194 Signals and Systems

X(s ) = X1( s)  X1( s) e sT0  X1( s) es2T0  X1( s) e s3T0  ....

= X1( s) [1  e sT0  es2T0  e s3T0  ....]

 
= X1( s)  e snT 0  X1( s)  ( esT 0 )n
n0 n0

1
X(s ) = X1( s)
1  e sT0
Transform, the Laplace transform of a periodic signal x(t) is given by,
X1( s)
X (s ) = ...(67)
1  e sT0

Example 4.32

Find the Laplace transform of the square wave shown in Fig. (a).
Solution 4.32
We know that the Laplace transform of a causal periodic signal is given by equation,
X1( s)
X (s ) =
1  e sT0
x(t) x 1(t)

A A

t t
0 1 2 3 4 5 0 1 2

–A –A

(a) (b)
Figure : (a) Causal periodic sequence wave (b) first cycle of x(t)
where X1(s) is the Laplace transform of the first cycle x1(t) of the causal periodic signal x(t). The given
square wave is a causal periodic signal with period T0 = 2. The first cycle x1(t) of x(t) is as shown in
Fig. (b). Using the step function u(t), x1(t) can be written as,
x 1( t ) = Au(t) – 2 Au(t – 1) + Au(t – 2)
Taking the Laplace transform of the above equation, we obtain
L[x1(t)] = AL[u(t)] – 2 AL[u(t – 1)] + AL[u(t – 2)]
A 2 A  s A 2s
X 1( s ) =
e  e
s s s
A A
= [1  2 e s  e2 s ]  (1  e s )2
s s
Substituting X1(s) and T0 = 2 in the expression of X(s), we get
X 1(s ) A (1  e s )2 A (1  e s )2
X (s )  sT0 = 
1e s 1  e2 s s (1  e s ) (1  e s )

A 1  e s A e s / 2 ( es / 2  e s / 2 )
= 
s 1  e s s e s / 2 ( e s / 2  e s / 2 )

A  es / 2  e s / 2  A s
X(s ) =  s / 2 s/2   tan h  
s  e e  s 2

GATE MASTERS PUBLICATION

 laplace Transform 195

Example 4.33

1
An LTI system is described by H( s )  2
. Find the system response for the input:
s  3s  1
( i ) x ( t ) = 2 e –2 t
( i i ) x ( t ) = 2 cos(2 t + 20°)
Solution 4.33
Ct
( i ) When the input is exponential eCt , the response of system with transfer function is e H (s ) s C
Given, x ( t ) = 2e–2t
(This is real exponential with C = –2)

1
H(–2) =  1
4  6 1
Hence, ycs(t) = –2e–2t
( i i ) x(t) = 2 cos(2t + 20°), we know that output for A cos(o t  ) is A cos( o t  )H ( j o ) .
Here o = 2, thus response will be
yss ( t ) = 2 H( j 2) cos(2t  20  H( j 2) )
= 0.4 cos(2t + 20° + 53.1°)
= 0.2 cos(2t + 73.1°)

Example 4.34

An absolutely integrable signal is known to have a pole at s = 2. Answer the following

questions:
( i ) Could x ( t ) be of finite duration?
( i i ) Could x ( t ) be left sided?
( i i i ) Could x ( t ) be right sided?
( i v ) Could x ( t ) be two sided?
Solution 4.34
(i) No we know that for a finite duration signal, ROC is the entire s-plane. Therefore, there can no poles
in finite s-plane for a finite duration signal. Since, the given signal x(t) has a pole at s = 2, this is not
the case.
( i i ) Yes since the signal is absolutely integrable, the ROC must include the j-axis. Furthermore, X(s) has
a pole at s = 2. Therefore, one valid ROC for signal would be  < 2. Signal it would correspond to a
left-sided.
( i i i ) Since the signal is absolutely integrable, the ROC must include the j-axis. Furthermore, X(s) has a
pole at a s = 2. For a right-sided signal ROC is of the form  > a for some constant a and ROC does
not contain any pole. Therefore, x(t) cannot be right-sided signal.
( i v ) Yes since the signal is absolutely integrable, the ROC must include the j-axis. Furthermore, X(s) has
a pole at s = 2. Therefore, a valid ROC for the signal could be a strip in the region a < R{s} < 2 such
that a < 0 and we know that this would correspond to a two-sided signal.

GATE MASTERS PUBLICATION

196 Signals and Systems

Example 4.35

How many signals have a Laplace transform that my be expressed as,

1
X( s ) 
( s  2) ( s  3) ( s2  s  1)
in its region of convergence?
Solution 4.35
We may find different signals with the given Laplace transform by choosing different region of convergence.
Consider the given Laplace transform,
1
X (s ) =
( s  2) ( s  3) ( s2  s  1)

1
=
 1 j 3 1 j 3
( s  2) ( s  3)  s     s  
 
 s 2  2 2 

The poles of the given Laplace transform are

s 1 = –2, s2 = –3

1 3 1 3
s3 =   j ; s4    j
2 2 2 2
The pole-zero diagram is shown in figure. Based on the location of these poles, we may choose from the
following ROCs:
1 1
(i )  ( ii ) 2    
2 2
(iii) –3 <  < –2 (iv )  < –3
Therefore, we may find four different signals with the given Laplace transform.

Example 4.36

Let x ( t ) be a signal that has a rational Laplace transform with exactly two poles located at
s = –1 and s = –3. If g ( t ) = e 2t x ( t ) and G () = F [ g ( t )] converges, determine whether x ( t )
is left sided, right sided, or two sided.
Solution 4.36
It is given that g(t) = x(t) e2t is absolutely integrable, since Laplace transform X(s) has two poles –1 and –3.
So X(s) can be have three possibilities of ROC.
1.  < –3 2. –1 <  < –3 3.  > –1
From the definition of ROC, ROC is value of  where x(t) e t is absolutely integrable. Since x(t) e2t is
–

absolutely integrable so ROC must contain  = –2. So ROC of X(s) is –3 <  < –1. So, x(t) is double-sided
signal.

Example 4.37

Suppose the following facts are given about the signal x ( t ) with Laplace transform X ( s ):
1 . x ( t ) is real and even.
2 . X ( s ) has four poles and no zeros in the finite s-plane.

GATE MASTERS PUBLICATION

 laplace Transform 197

1  j / 4
3 . X ( s ) has a pole at s  e .
2

4.  x( t ) dt  4 .


Determine X ( s ) and it’s ROC.

Solution 4.37
Given that, x(t) is real and even. So poles will be in conjugate
pairs and will be at mirror image with respect to j-axis also.
Im
1
1st : Pole at P1  e j / 4 , 1
j
2 2 2

1  j / 4
2nd : Pole will be P2  P1  e ,
2 Re
1 1

1 2 2 2 2
3rd : Pole will be mirror image of P1 i.e. P3  P1   e j  /4 and
2
1  j / 4 1
 j
4th : Pole will be at mirror image of P2 i.e. P4  P2   e 2 2
2
Thus, there are three ROC is
1 1 1 1
1.  2.   3.  
2 2 2 2 2 2 2 2
and x(t) is absolutely integrable so ROC must certain j-axis so ROC is
A
X (s ) =
( s  P1 ) ( s  P2 ) ( s  P3 ) ( s  P4 )

 st
Since, X (s ) =  x(t ) e dt


 X(0) =  x(t ) dt  4

So, X(0) = 4 will give value of A
1
so we can find, A =
4
1/ 4 1 1
 X (s ) = ,  
( s  P1 ) ( s  P2 ) ( s  P3 ) ( s  P4 ) 2 2 2 2

Example 4.38

Determine the impulse response h ( t ) of the system H ( s ) of an LTI system from the following
facts:
1 . When the input to the system is x ( t ) = e 2 t and the output is y ( t ) = 1/6 e 2t .
dh( t )
2 . When h ( t ) satisfies the differential equation,  2h( t )  e4t u( t )  b u( t ) where b is an
dt
unknown constant. Your answer must not contain any unknown constant.

GATE MASTERS PUBLICATION

198 Signals and Systems

Solution 4.38
We know that if we apply a complex exponential input x(t) = eCt to an LTI system with impulse response
h(t), the system output will be y(t) = H(C) eCt, etc.,
x(t )  y ( t )
eCt  H(C) eCt
1 2t
Given that, e2t  e = H(2) e2t
6
Taking the Laplace transform of the equation given in question, we obtain
1 b
sH(s) + 2H(s) = 
s4 s
s  b( s  4)
(s + 2) H(s) =
s( s  4)

Also, given that, H( s) s  2 = H(2)  2  6 b  1

2 46 6
Thus, b = –1
s  ( s  4) 2( s  2)
Therefore, H (s ) = 
s( s  2) ( s  4) s( s  2) ( s  4)

2
H (s ) =
s( s  4)
Using partial fraction expansion, we obtain
1/ 2
1/ 2
H(s ) = 
s s4
The inverse Laplace transform of the above equation yields,
1
h( t ) = [1  e4t ] u(t )
2

Example 4.39

The input x ( t ) and output y ( t ) of an LTI system are related through above representation
shown in figure. Find differential equation that represent the system and find all properties
of system.

x(t) y (t)

1
s
–2 –1

1
s

–1 –6

GATE MASTERS PUBLICATION

 laplace Transform 199

Solution 4.39
Let us draw the system again, we can see that,
V(s)
x(t) y (t)

1 1
s s

–2 –1

1 1
s s

–1 –6

V( s) V( s) X( s)
X(s ) = 2  2  V( s)  V(s) =
s s 2 1
1  2
s s
V( s) 6 V( s)
and Y ( s ) = V( s )  
s s2
 1 6
X( s) 1   2 
1 6  s s 
 Y(s) = V( s) 1   2   Y(s) = 2 1
 s s  1 
s s2
Y( s)  s2  s  6 
 = H( s)   2 
X( s)  s  2s  1 
Now, in such systems since system is realizable so it is causal and since pole of H(s) are at –1, –1 so ROC is
 > –1 so system is stable.
Now we can easily differential equation.
 (s2 + 2s + 1) Y(s) = X(s) (s2 – s – 6)
d2 y(t ) dy(t ) d2 x(t ) dx(t )
  2  y ( t ) =   6 x(t )
dt2 dt dt 2 dt

Study Note
• A finite duration signal has ROC entire s-plane thus it will never have a finite location pole. If pole is at s = 
in that case ROC is entire s-plane except s = . Example (t) which has Laplce transform ‘s’.
(t )  s , ROC entire s-plane except s = 
• The signal which is both sided must have minimum of 2 poles with different real values.
• If in any question transfer function of a system, H(s) is given and no information about it then assume it
to be causal.
• If an LTI system has transfer function H(s) then it is always invertible and invertible system will have
1
transfer function HI ( s)  .
H( s)



GATE MASTERS PUBLICATION

 200 Signals and Systems

1. x(t) e–3t is absolutely integrable.

2. x(t)  (e–t u(t)) is absolutely integrable.
3. x(t) = 0, t > 1.
x(t) = 0, t < –1.
1 Subjective Practice Problems
4.
Im
2j
Im
2j

1 . Consider the signal,

x(t) = e–5t u(t) + e–t u(t) –2 2
Re
–2 2
Re

and denote its Laplace transform by X(s). What are

the constraints placed on the real and imaginary parts –2j –2j

 if the region of convergence of X(s) is Re{s} > –3?

Im Im
2 . Suppose we are given the following three facts about
2j 2j
the signal x(t):
1. x(t) = 0 for t < 0
Re Re
2. x(k/80) = 0 for k = 1, 2, 3.... –2 2 –2 2
3. x(1/160) = e–120
–2j –2j
Let X(s) denote the Laplace transform of x(t), and
determine which of the following statements is
consistent with the given information about x(t):
6 . We are given the following five facts about a real
(a) X(s) has only one pole in the finite s-plane.
signal x(t) with Laplace transform X(s):
(b) X(s) has only two poles in the finite s-plane.
1. X(s) has exactly two poles.
(c) X(s) has more than two poles in the finite s-plane.
2. X(s) has no zeros in the finite s-plane.
3 . Consider two right-sided signals x(t) and y(t) related 3. X(s) has a pole at s = –1 + j.
through the differential equations, 4. e2t x(t) is not absolutely integrable.
dx(t ) 5. X(0) = 8.
= –2 y(t) + (t) Determine X(s) and specify its region of convergence.
dt
dy(t ) 7 . Consider an LTI system for which the system function
and = 2 x(t)
dt H(s) has the pole-zero pattern shown in figure.
Determine Y(s) and X(s), along with their regions of (a) Indicate all possible ROCs that can be associated
convergence. with this pole-zero pattern.
4 . A causal LTI system S has the block diagram (b) For each ROC identified in part (a), specify
representation shown in figure. Determine a whether the associated system is stable and/or
differential equation relating the input x(t) to the causal.
output y(t) of this system. Im

x(t) 2/s y (t)

–4 Re
–2 –1 +1 +2

1/s

–2 8 . Determine whether each of the following statements

is true or false. If a statement is true, construct a
5 . For each of the following statements about x(t), and
convincing argument for it. If it false, give a counter
for each of the four pole-zero plots in figure,
example:
determine the corresponding constraint on the ROC:

GATE MASTERS PUBLICATION

 laplace Transform 201

(a) The Laplace transform of t 2 u ( t ) does not 12. Determine whether or not each of the following
converge anywhere on the s-plane. statements about LTI system is true. If a statement is
2
(b) The Laplace transform of e t u (t ) does not true, construct a convincing argument for it. If it
false, give a counter example.
converge anywhere on the s-plane.
(a) A stable continuous-time system must have all
(c) The Laplace transform of e j0t does not converge
its poles in the left half of the s-plane
anywhere on the s-plane.
(d) The Laplace transform of e j0t u(t) does not [i.e. Re{s} < 0].
converge anywhere on the s-plane. (b) If a system function has more poles than zeros,
(e) The Laplace transform of t  does not converge and the system is causal, the step response will
anywhere on the s-plane. be continuous at t = 0.
(c) If a system function has more poles than zeros,
9 . Let h(t) be the impulse response of a causal and and the system is not restricted to be causal, the
stable LTI system with a rational system function.
step response can be discontinuous at t = 0.
(a) Is the system with impulse response dh(t)/dt
(d) A stable, causal system must have all its poles
guaranteed to be causal and stable?
and zeros in the left half of the s-plane.
t
(b) Is the system with impulse response  h( ) d 13. Suppose that we are given the following information
 about an LTI system:
guaranteed to be causal and unstable?
1. The system is causal.
10. Let x(t) be the sampled signal specified as, 2. The system function is rational and has only two
 poles, at s = –2 and s = 4.
x(t) =  e nT (t  nT ) 3. If x(t) = 1, then y(t) = 0.
n0
where, T > 0. 4. The value of the impulse response at t = 0+ is 4.
(a) Determine X ( s ), including its region of Determine the system function.
convergence. 14. Consider a stable and causal system with impulse
(b) Sketch the pole-zero plot for X(s). response h(t) and system function H(s) is rational,
(c) Use geometric interpretation of the pole-zero plot
contains a pole at s = –2, and does not have a zero
to argue that X(j) is periodic.
at the origin. The location of all other poles and
11. Consider a stable and causal system with a real zeros is unknown. For each of the following
impulse response h(t) and system function H(s). It is statements let us determine whether we can
known that H(s) is rational, one of its poles is at definitely say that is true, whether we can definitely
–1 + j, one of is zeros is at 3 + j, and it has exactly say that it is true, whether we can definitely say that
two zeros at infinity. For each of the following it is false, or whether there is insufficient information
statements, determine whether it is true, whether is to as certain the statement’s truth:
false, or whether there is insufficient information to (a) L[h(t) e3t} converges.
determine the statement’s truth.

(a) h(t) e–3t is absolutely integrable. (b)  h(t ) dt  0 .
(b) The ROC for H(s) is Re{s} > –1. 
(c) The differential equation relating inputs x(t) and (c) t h(t) is the impulse response of a causal and
outputs y(t) for system may be written in a form stable system.
having only real coefficients. (d) dh(t)/dt contains at least one pole in its Laplace
(d) Lim H( s)  1. transform.
s
(e) H(s) does not have fewer than four poles. (e) h(t) has finite duration.
(f) H(j) = 0 for at least one finite value of . (f) H(s) = H(–s).
(g) If the input to system is e3t sint, the output is (g) Lim H( s)  2.
s
e3t cost.

GATE MASTERS PUBLICATION

 202 Signals and Systems

15. Find the inverse Laplace transform of the following 17. The Laplace transform of a signal x(t) that is zero
functions: for t < 0 is
2s  5 s3  2s2  3s  2
(a) X( s)  ;  3  R{s}  2 X( s) 
( s  2) ( s  3) s4  2s3  2s2  2s  2
2s  5 Determine the Laplace transform of the following
(b) X( s)  ; 2  R{s}  3 signals:
( s  2) ( s  3)
t
2s  3 (a) y(t )  3x  
(c) X( s)  ; R{s}  1 3
( s  1) ( s  2)
(b) y(t) = tx(t)
2s  3 (c) y(t) = t x(t – 1)
(d) X( s)  ; R{s}  2
( s  1) ( s  2) dx(t )
(d) y(t ) 
dt
16. An absolutely integrable signal x(t) has a pole at
s = . It is possible that other poles may be present. dx(t )
(e) y(t )  (t  1) x(t  1) 
(a) Can x(t) be left-sided? Explain. dt
(b) Can x(t) be right-sided? Explain.
(c) Can x(t) be Two-sided? Explain. 
(d) Can x(t) be finite duration? Explain.

GATE MASTERS PUBLICATION