MAE 143A

Signals and Systems Prof. M. Krstic

Solutions for Homework 3

Problem 1. Compute the impulse response of the following systems:
1
(a) H(s) = e−5s
(s + 4)2 + 2

1
(b) H(s) =
(2s + 3)5

Solution:
(a) Impulse response is given by L−1 {H(s)}. Since
√
1 −5s 2
H(s) = √ e √ 2
2 (s + 4)2 + 2

and
( √ )
√ 
2
L−1 √ 2 = e−4t sin 2t ,
(s + 4)2 + 2
we get
1 −4(t−5) √ 
L −1
{H(s)} = √ e sin 2(t − 5) 1(t − 5) .
2

(b)

1 t4 − 3 t
   
1 1
L −1
=L −1
= · e 2.
(2s + 3)5 32(s + 3/2)5 32 4!

Problem 2. Compute the step response of the following systems:

s2 + s + 1
(a) H(s) = 2
s2 + 3s + 2

1
(b) H(s) =
(s + 1)3

2
(c) H(s) =
(s + 1)(s2 + 4)
 Solution: (a)
1 s2 + s + 1 s2 + s + 1
F (s) = H(s) = 2 2 =2
s s(s + 3s + 2) s(s + 1)(s + 2)
C1 C2 C3
F (s) = + +
s s+1 s+2
s2 + s + 1  s2 + s + 1 
 
2
C1 = 2 = 1, C2 = 2 = = −2
(s + 1)(s + 2) s=0 s(s + 2) s=−1 −1


s2 + s + 1 

6
C3 = 2 = =3
s(s + 1) s=−2 (−2)(−1)


1 2 3
F (s) = − +
s s+1 s+2


f (t) = 1 − 2e−t + 3e−2t 1(t) .

(b)
1 1
F (s) = H(s) =
s s(s + 1)3
A B C D
F (s) = + + 2
+
s s (s + 1) (s + 1)3
 
1  1 
A=  = 1, D =  = −1
(s + 1)3 s=0 s s=−1
  
d 1  1 
C= = − 2 = −1
ds s s=−1 s s=−1
1 d2 1 
   
1 2 
B= = = −1
2 ds2 s s=−1 2 s3 s=−1
1 1 1 1
F (s) = − − 2
−
s s + 1 (s + 1) (s + 1)3
t2 −t
 
f (t) = 1 − e − te − e
−t −t
1(t)
2
(c)
1 2
F (s) = H(s) = ,
s s(s + 1)(s2 + 4)

and doing partial fraction decomposition,

2 2 C1 C2 C3 C4
2
= = + + + .
s(s + 1)(s + 4) s(s + 1)(s + j2)(s − j2) s s + 1 s + j2 s − j2
Now,

H(s)  2 1
C1 = s = = ,
s s=0 4
 2

H(s)  2 −2
C2 = (s + 1) = = ,
s s=−1 (−1)5
 5

H(s)  2 −1
C3 = (s + j2) = = ,
s s=−j2 −j2(−j2 + 1)(−j4)
 4(1 − j2)

H(s)  2 −1
C4 = (s − j2) = = = C̄3 .
s  j2(j2 + 1)(j4)
s=j2 4(1 + j2)

Then,
2 11 2 1 1 1 1 1
2
= − − − .
s(s + 1)(s + 4) 2 s 5 s + 1 4(1 + j2) s + j2 4(1 + j2) s − j2

Working on the last two fractions, we have that

1 1 1 1 (1 + j2)(s − j2) + (1 − j2)(s + j2) s+4

− − = − =−
4(1 − j2) s + j2 4(1 + j2) s − j2 4(1 + j2)(1 − j2)(s + j2)(s − j2) 10(s2 + 4)
 
1 s 2
= − + 2
10 s2 + 4 s2 + 4

Hence,

    
H(s) 11 2 1 1 s 2
L −1
= L −1
− − +2 2
s 2 s 5 s + 1 10 s2 + 4 s +4
 
1 2 −t 1
= − e − (cos(2t) + 2 sin(2t)) 1(t)
2 5 10
√   !
1 2 −t 5 1
= − e − sin 2t + arctan 1(t)
2 5 10 2

A maybe easier alternative, using matching coefficients in the partial fraction decomposition,
2 C1 C2 As + B
2
= + + 2 .
s(s + 1)(s + 4) s s+1 s +4
1
As before, C1 = 2
and C2 = −2
5
. Then, expanding the fraction:

2 C1 C2 As + B
2
= + + 2
s(s + 1)(s + 4) s s+1 s +4
C1 (s + s + 4s + 4) + C2 (s3 + 4s) + (As + B)(s2 + s)
3 2
=
s(s + 1)(s2 + 4)
s3 (C1 + C2 + A) + s2 (C1 + A + B) + s(4C1 + 4C2 + B) + 4C1
= ,
s(s + 1)(s2 + 4)
hence,

C1 + C2 + A = 0,
C1 + A + B = 0,
4C1 + 4C2 + B = 0,
4C1 = 2,

from where we get C1 = 1/2. Then, A = −C1 − C2 = −1/2 + 2/5 = −1/10, B = −A − C1 =

1/10 − 1/2 = −4/10, and the third equation is verified thus ensuring that our solution is correct.
We get
 
2 11 2 1 1 s 4
= − − +
s(s + 1)(s2 + 4) 2 s 5 s + 1 10 s2 + 4 s2 + 4

as before.

Problem 3. For the system with a transfer function

s+1 1
H(s) = 2
−
3s + 7 3s + 2
find the differential equation governing the relationship between the input u(t) and the
output y(t), assuming zero initial conditions.

Solution:

Y (s) = H(s)U(s)

(3s2 + 7)(3s + 2)Y (s) = [(s + 1)(3s + 2) − (3s2 + 7)]U(s)

(9s3 + 6s2 + 21s + 14)Y (s) = (5s − 5)U(s)

...
9 y + 6ÿ + 21ẏ + 14y = 5u̇ − 5u

Problem 4.
(a) Let the persistent forcing signal u(t) = sin(2t)1(t) drive the system

2s2 + 8
Y (s) = U(s) .
s(s2 + 2s + 15)

Does this system, despite persistent forcing, reach a steady state? If so, what is limt→∞ y(t)?
(b) Same questions as in (a) for the system

2s2 + 8
Y (s) = U(s) .
s(s2 + 2s − 15)
 2
Solution: (a) Since U(s) = s2 +4
,

2s2 + 8 4 4
Y (s) = 2
U(s) = 2
= ,
s(s + 2s + 15) s(s + 2s + 15) s((s + 1)2 + 14)
4
and then sY (s) = (s+1)2 +14
has all poles in the LHP, so the FVT can be applied and

4 4
lim y(t) = lim sY (s) = = .
t→∞ s→0 12 + 14 15
(b)

2s2 + 8 4 4
Y (s) = 2
U(s) = 2
= ,
s(s + 2s − 15) s(s + 2s − 15) s(s + 5)(s − 3)
4
and then sY (s) = (s+5)(s−3) has a pole in the RHP, so the FVT cannot be applied: the system
does not have a steady state (actually, it diverges to infinity).

Problem 5. Using the Laplace transform, solve the following differential equations:
...
(a) y + 2ÿ + 2ẏ = 0, y(0) = 1, ẏ(0) = 2, ÿ(0) = 3.

(b) ÿ + 2ẏ + 4y = (1 − t)e−2t , y(0) = 1, ẏ(0) = 1.

Solution:
(a) Applying the Laplace transform, we get

s3 Y (s) − s2 y(0) − sẏ(0) − ÿ(0) + 2(s2 Y (s) − sy(0) − ẏ(0)) + 2(sY (s) − y(0)) = 0

s2 + 4s + 9 s2 + 4s + 9 C1 C2 s + C3
Y (s) = = = +
s(s2 + 2s + 2) s((s + 1)2 + 1) s (s + 1)2 + 1

s2 + 4s + 9 

9
C1 = 2
=
(s + 1) + 1 s=0 2


9 2 7
(s + 2s + 2) + C2 s2 + C3 s = s2 + 4s + 9 ⇒ C2 = − , C3 = −5
2 2

9/2 (−7/2)s − 5 9/2 7/2(s + 1) 3/2

Y (s) = + 2
= − −
s (s + 1) + 1 s (s + 1) + 1 (s + 1)2 + 1
2

 
9 7 3
y(t) = − cos t + sin t e−t .
2 2 2
 (b)
 1 1 s+1
L (1 − t)e−2t = − 2
=
s + 2 (s + 2) (s + 2)2
Applying the Laplace transform to the ODE we get
s+1
s2 Y (s) − sy(0) − ẏ(0) + 2(sY (s) − y(0)) + 4Y (s) =
(s + 2)2

s+1
(s2 + 2s + 4)Y (s) = +s+3
(s + 2)2

s+1 s+3
Y (s) = + 2
(s2 + 2s + 4)(s + 2)2 s + 2s + 4

s+1 C1 C2 C3 s + C4
= + + 2
(s2 + 2s + 4)(s + 2)2 s + 2 (s + 2)2 s + 2s + 4

s2 + 2s + 4 − (s + 1)(2s + 2) 
  
d s+1  1
C1 = 2
 = 2 2
=
ds s + 2s + 4 s=−2
 (s + 2s + 4) 
s=−2 8

s + 1  1
C2 = 2
=−
s + 2s + 4 s=−2
 4

Matching coefficients gives

1 1
(s + 2)(s2 + 2s + 4) − (s2 + 2s + 4) + (C3 s + C4 )(s + 2)2 = s + 1
8 4

1 1 1
+ C3 = 0, 4C4 = 1 ⇒ C3 = − , C4 =
8 8 4
Going back to Y (s),
1
− 14
8
− 18 s + 14 s+3
Y (s) = + 2
+ 2 + 2
s + 2 (s + 2) s + 2s + 4 s + 2s + 4
1 1 7
8
−4 8
s + 134
= + +
s + 2 (s + 2)2 (s + 1)2 + 3
1
8
− 14 7
8
(s + 1) + 198
= + 2
+ 2
s + 2 (s + 2) (s + 1) + 3

1 −t
 √  19 √  1
y(t) = e 7 cos 3t + √ sin 3t + (1 − 2t)e−2t
8 3 8