38 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

3A 4.1 Using the graphical method, for each pair of functions x and h given in the figures below, compute x ∗ h.

x(t) h(t)
2 2

1
1 1

t t
−1 0 1 2 3 −1 0 1 2 3
(a)

h(t)
2
x(t)
1 1

− 12
t 1 t
−1 1 −1 2 1

−1
−1

−2

(b)

x(t) h(t)

2 1

t
−1 1
t −1
0 1
(c)

h(t)
x(t) 2

1
1
t
0 1 2 3
t
−1 0
(d)

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 39

h(t)

x(t)
1

1
t
−1 1
t
0 1 2
−1

(e)

x(t) h(t)
5 5
4 4
3 3
2 2
1 1
t t
−3 −2 −1 0 1 2 −2 −1 0
(f)

3A Answer (e).
To assist in the convolution computation, we first plot x(τ) and h(t − τ) versus τ as shown below.

(a)
(b)

(d)
(c)

From the above plots, we can deduce that there are six cases (i.e., intervals of t) to be considered, which
correspond to the scenarios shown in the graphs below.

Edition 2022-09-01 Copyright © 2012–2022 Michael D. Adams

40 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

(a) (b)

(c) (d)

(f)
(e)

In the case that t < −1, which corresponds to Figure (a), we trivially have

x ∗ h(t) = 0.

In the case that −1 ≤ t < 0, which corresponds to Figure (b), we have

Z t+1
x ∗ h(t) = (τ)(t − τ + 1)dτ.
0

In the case that 0 ≤ t < 1, which corresponds to Figure (c), we have

Z t Z 1 Z t+1
x ∗ h(t) = (τ)(t − τ − 1)dτ + (τ)(t − τ + 1)dτ + (2 − τ)(t − τ + 1)dτ.
0 t 1

In the case that 1 ≤ t < 2, which corresponds to Figure (d), we have

Z 1 Z t Z 2
x ∗ h(t) = (τ)(t − τ − 1)dτ + (2 − τ)(t − τ − 1)dτ + (2 − τ)(t − τ + 1)dτ.
t−1 1 t

In the case that 2 ≤ t < 3, which corresponds to Figure (e), we have

Z 2
x ∗ h(t) = (2 − τ)(t − τ − 1)dτ.
t−1

In the case that t ≥ 3, which corresponds to Figure (f), we trivially have

x ∗ h(t) = 0.

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 41

Combining the above results, we have that

t+1
R
(τ)(t − τ + 1)dτ −1 ≤ t < 0
R0t


− τ − 1)dτ + t1 (τ)(t − τ + 1)dτ + 1t+1 (2 − τ)(t − τ + 1)dτ

0≤t <1
R R
(τ)(t

R0


1 Rt R2
x ∗ h(t) = t−1 (τ)(t − τ − 1)dτ + 1 (2 − τ)(t − τ − 1)dτ + t (2 − τ)(t − τ + 1)dτ 1≤t <2
R2
t−1 (2 − τ)(t − τ − 1)dτ 2≤t <3





0

otherwise.

3A Answer (f).

To assist in the convolution computation, we first plot x(τ) and h(t − τ) versus τ as shown below.

(b)
(a)

(c)
(d)

From the above plots, we can deduce that there are five cases (i.e., intervals of t) to be considered, which
correspond to the scenarios shown in the graphs below.

(a)
(b)

Edition 2022-09-01 Copyright © 2012–2022 Michael D. Adams

42 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

(c) (d)

(e)

In the case that t < −5, which corresponds to Figure (a), we trivially have

x ∗ h(t) = 0.

In the case that −5 ≤ t < −4, which corresponds to Figure (b), we have
Z t+2
x ∗ h(t) = (2 − τ)(τ − t − 1)dτ.
−3

In the case that −4 ≤ t < 0, which corresponds to Figure (c), we have

Z t+2
x ∗ h(t) = (2 − τ)(τ − t − 1)dτ.
t+1

In the case that 0 ≤ t < 1, which corresponds to Figure (d), we have

Z 2
x ∗ h(t) = (2 − τ)(τ − t − 1)dτ.
t+1

In the case that t ≥ 1, which corresponds to Figure (e), we trivially have

x ∗ h(t) = 0.

Combining all of the above results, we have

R t+2
R−3 (2 − τ)(τ − t − 1)dτ
 −5 ≤ t < −4
 t+2 (2 − τ)(τ − t − 1)dτ

−4 ≤ t < 0
x ∗ h(t) = Rt+12
 t+1

 (2 − τ)(τ − t − 1)dτ 0≤t <1
0 otherwise.


Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 43

3A 4.3 Using the graphical method, compute

( x ∗ h for each pair of functions x and h given below.
t − 1 1≤t <2
(a) x(t) = et u(−t) and h(t) =
0 otherwise;
1 1
(b) x(t) = e−|t| and h(t) = rect(
( 3 [t − 2 ]);
t −1 1 ≤ t < 2
(c) x(t) = e−t u(t) and h(t) =
0 otherwise;
1 2−t
(d) x(t) = rect 2 t and h(t)(= e u(t − 2);

t + 2 −2 ≤ t < −1
(e) x(t) = e−|t| and h(t) =
0 otherwise;
(
t −1 1 ≤ t < 2
(f) x(t) = e−|t| and h(t) =
0 otherwise;
1
( (
1 − 4t 0 ≤ t < 4 t −1 1 ≤ t < 2
(g) x(t) = and h(t) =
0 otherwise 0 otherwise;
(
2−t 1 ≤ t < 2
(h) x(t) = rect 41 t and h(t) = and


0 otherwise;
(
−t t −2 2 ≤ t < 4
(i) x(t) = e u(t) and h(t) =
0 otherwise.

3A Answer (b).

(a) (b)

(c) (d)

(e) (f)

Edition 2022-09-01 Copyright © 2012–2022 Michael D. Adams

44 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

(g)

R t+1
Rt−2 e dτ R t < −1
 τ
0 t+1 −τ
x ∗ h(t) = t−2 e dτ + 0
τ e dτ −1 ≤ t < 2
R t+1 −τ
t−2 e dτ t ≥2


3A Answer (g).

(a) (b)

(c) (d)

(f)
(e)

(g) (h)

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 45

(i)

R t−1
− 41 τ + 1 (t − τ − 1)dτ 1≤t <2


 0
R t−1

1
t−2 − 4 τ + 1
(t − τ − 1)dτ 2≤t <5


x ∗ h(t) = R 4 1
t−2 − 4 τ + 1 (t − τ − 1)dτ 5≤t <6



0 otherwise


Edition 2022-09-01 Copyright © 2012–2022 Michael D. Adams

 46 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

3A 4.5 Let x, y, h, and v be functions such that y = x ∗ h and

Z ∞
v(t) = x(−τ − b)h(τ + at)dτ,
−∞

where a and b are real constants. Express v in terms of y.

3A Answer.
From the definition of v, we have
Z ∞
v(t) = x(−τ − b)h(τ + at)dτ.
−∞

Now, we employ a change of variable. Let λ = −τ − b so that τ = −λ − b and dτ = −dλ . Applying this
change of variable and simplifying, we obtain
Z −∞
v(t) = x(λ )h([−λ − b] + at)(−1)dλ
Z∞∞
= x(λ )h(at − b − λ )dλ
−∞
Z ∞
= x(λ )h([at − b] − λ )dλ
−∞
= x ∗ h(at − b)
= y(at − b).

Therefore, we have that v(t) = y(at − b).

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 47

3A 4.6 Consider the convolution y = x ∗ h. Assuming that the convolution y exists, prove that each of the following
assertions is true:
(a) If x is periodic, then y is periodic.
(b) If x is even and h is odd, then y is odd.

3A Answer (a).
From the definition of convolution, we have

y(t) = x ∗ h(t)
Z ∞
= x(τ)h(t − τ)dτ.
−∞

Suppose that x is periodic with period T . Then, we have x(t) = x(t + T ) and we can rewrite the above integral
as
Z ∞
y(t) = x(τ + T )h(t − τ)dτ.
−∞

Now, we employ a change of variable. Let λ = τ + T so that τ = λ − T and dλ = dτ. This yields
Z ∞
y(t) = x(λ )h(t − [λ − T ])dλ
−∞
Z ∞
= x(λ )h(t + T − λ )dλ
−∞
Z ∞
= x(λ )h([t + T ] − λ )dλ
−∞
= x ∗ h(t + T )
= y(t + T ).

Therefore, y is periodic with period T .

Edition 2022-09-01 Copyright © 2012–2022 Michael D. Adams

 48 Continuous-Time Linear Time-Invariant Systems (Chapter 4)

3A 4.9 Consider a LTI system whose response to the function x1 (t) = u(t) − u(t − 1) is the function y1 . Determine the
response y2 of the system to the input x2 shown in the figure below in terms of y1 .

x2 (t)

t
−3 −2 −1 1 2 3

−1

−2

−3

3A Answer.
First, we express x2 in terms of x1 . This yields

x2 (t) = 2x1 (t + 2) + x1 (t + 1) + x1 (t) − 2x1 (t − 1).

Then, we observe that the system is LTI. This implies that

2x1 (t + 2) + x1 (t + 1) + x1 (t) − 2x1 (t − 1) → 2y1 (t + 2) + y1 (t + 1) + y1 (t) − 2y1 (t − 1).

Therefore, we have

y2 (t) = 2y1 (t + 2) + y1 (t + 1) + y1 (t) − 2y1 (t − 1).

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01

 182 MATLAB (Appendix D)

3A D.103 Let F denote the complex-valued function of a real variable given by

1
F(ω) = .
jω + 1

Write a program to plot |F(ω)| and arg F(ω) for ω in the interval [−10, 10]. Use subplot to place both plots
on the same figure.

3A Answer.
w = linspace ( -10 , 10 , 500);
f = (j * w + 1) .ˆ ( -1);
subplot (2 , 1, 1);
plot (w , abs (f ));
title ( ’ Magnitude ’);
xlabel ( ’\ omega ’);
ylabel ( ’|F (\ omega )| ’);
subplot (2 , 1, 2);
plot (w , unwrap ( angle (f )));
title ( ’ Argument ’);
xlabel ( ’\ omega ’);
ylabel ( ’arg F (\ omega ) ’);

Magnitude
1

0.8

0.6
|F(ω)|

0.4

0.2

0
−10 −8 −6 −4 −2 0 2 4 6 8 10
ω

Argument
1.5

0.5
arg F(ω)

−0.5

−1

−1.5
−10 −8 −6 −4 −2 0 2 4 6 8 10
ω

Copyright © 2012–2022 Michael D. Adams Edition 2022-09-01