0% found this document useful (0 votes)
28 views7 pages

2018midterm1 Solution

The document contains 5 problems related to linear time-invariant (LTI) systems and Fourier analysis. Problem 1 asks to check the truth of 5 statements about LTI systems and provides justifications. Problem 2 asks to find the output of an LTI system with a given impulse response for a specified input. Problem 3 asks to find a signal satisfying 6 given conditions. Problem 4 asks to solve a linear system relating an input and output vector. Problem 5 asks Fourier transform questions about 2 given signals.

Uploaded by

김명주
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
0% found this document useful (0 votes)
28 views7 pages

2018midterm1 Solution

The document contains 5 problems related to linear time-invariant (LTI) systems and Fourier analysis. Problem 1 asks to check the truth of 5 statements about LTI systems and provides justifications. Problem 2 asks to find the output of an LTI system with a given impulse response for a specified input. Problem 3 asks to find a signal satisfying 6 given conditions. Problem 4 asks to solve a linear system relating an input and output vector. Problem 5 asks Fourier transform questions about 2 given signals.

Uploaded by

김명주
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 7

Problem 1)[30pt] Check whether each of the following statements is true or false.

If you do
not justify your answer, you will not get any credit.

(a)[6pt] The system whose input-output relationship is given by y(t) = dx(t)


dt is invertible.

(b)[6pt] y(t) = cos 2πt + π2 could be the output signal for some time-invariant system corre-


sponding to the input signal x(t) = cos (4πt + π).


2
(c)[6pt] The LTI system whose frequency response function H(jw) is given by H(jw) = 1+w 2
satisfies the BIBO stability.
n
(d)[6pt] y[n] = 41 u[n] + cos(2n)u[n − 1] could be the output signal for some causal LTI system
corresponding to the input signal x[n] = e3n u[n − 1].

(e)[6pt] y(t) could be the output signal for some LTI system corresponding to the input signal
x(t), where x(t) and y(t) are given as follows:

1 1

0.8 0.8

0.6 0.6

0.4 0.4

0.2 0.2
x(t)

y(t)

0 0

-0.2 -0.2

-0.4 -0.4

-0.6 -0.6

-0.8 -0.8

-1 -1
-3 -2 -1 0 1 2 3 -3 -2 -1 0 1 2 3
t t

(a) x(t) (b) y(t)


Solution) (a)False.
x1 (t) = t and x2 (t) = t + 1 yield the same output.

(b)False.
Suppose that y(t) could be the output signal for some time-invariant system corresponding to
the input signal x(t). Then, y(t + 21 ) should be the output signal corresponding to the input
signal x(t + 12 ). Then, since x(t + 12 ) = cos(4πt + 2π + π) = cos(4πt + π) = x(t), y(t + 21 ) = y(t),
which is a contradiction.

(c) True.
2 −|t| . Note that ∞ |h(t)|dt = 2 < ∞, which means that
R
Since H(jw) = 1+w 2 , we have h(t) = e −∞
h(t) is absolutely integrable. Thus, this system satisfies the BIBO stability.

(d) False.
Suppose that y[n] could be the output signal for some causal LTI system corresponding to the
input signal x[n]. Observe that x[n] = 0 for n < 1. Then, since this system is causal and LTI,
y[n] = 0 for n < 1. However, y[0] = 1, which is a contradiction.

(e) False.
Observe that x(t) and y(t) are the sinusoidal signals of different frequencies. Noting that an LTI
system preserves the frequency of a sinusoidal signal, y(t) cannot be the output signal for some
LTI system corresponding to the input signal x(t).
Problem 2)[10pt] Consider the LTI system whose impulse response is given by h(t) = ejw0 t .
When the input is
x(t) = u(t + 0.5) − u(t − 0.5),
find out the output y(t).

Solution) Observe that 


1 if |t| < 0.5
x(t) =
0 if |t| > 0.5
Since the system is LTI, the output signal y(t) satisfies

y(t) = x(t) ∗ h(t)


Z ∞
= x(τ )h(t − τ )dτ
−∞
Z 0.5
= ejw0 (t−τ ) dτ
−0.5
(
1 if w0 = 0
=
ejw0 t sin(w 0 /2)
w0 /2 6 0
if w0 =
w0
= ejw0 t sinc( ).

Problem 3)[20pt] Find out the signal x(t) satisfying the following conditions:

1. x(t) is a real signal.

2. x(t) is periodic with period T = 6 and has Fourier series coefficients ak .

3. ak = 0 for k = 0 and k > 2.

4. x(t) = −x(t − 3).


R3
5. 16 −3 |x(t)|2 dt = 12 .

6. a1 is a positive real number.

You should justify which properties of the continuous-time Fourier series you use.

Solution) Since x(t) = −x(t − 3) and x(t) is periodic with period T = 6, we have

ak = −ak e−jk 6
3
= −ak (−1)k .

From this result, we obtain ak = 0 if k is an even number. In particular, a2 = a−2 = 0.


Furthermore, since x(t) is a real signal, we have ak = a∗−k . Combining this with the condition 3
and 6, we have ak = 0 for |k| > 2, a1 = a−1 > 0, and a0 = 0. Finally, by using the Parseval’s
π π
relation, we have a1 = a−1 = 21 . Therefore, x(t) = 21 (ej 3 t + e−j 3 t ) = cos π3 t.
Problem 4)[20pt] Solve the linear system y = Ax, where
   
1 1 1 1 16 + j4
1 1 −j −1

 and y = −2 − j2 .
j   
A= 
4 1 −1 1 −1 −4 − j4
1 j −1 −j −6 + j6

 T  T
Solution) Let y = y0 y1 y2 y3 and x = x0 x1 x2 x3 . Observe that

1 1 1 1
 
−j π2 π π
1 1 e
 (e−j 2 )2 (e−j 2 )3 
A=  ,
4 1 e−jπ (e−jπ )2 (e−jπ )3 
3π 3π 3π
1 e−j 2 (e−j 2 )2 (e−j 2 )3

which implies that y0 , y1 , y2 , y3 are the Fourier series coefficients of the periodic signal x[n],
where x[n] = x[n + 4] and x[n] = xn for 0 ≤ n ≤ 3. Therefore, we have
π 3π
x[n] = y0 + y1 ej 2 n + y2 ejπn + y3 ej 2
n
. (1)

Then, by putting n = 0, 1, 2, 3 to (1), we have


 
4 + j4
28 + j12
 20 − j4  .
x= 

12 + j4
Problem 5)[30pt] Please answer the following questions. You should justify which properties
of the continuous-time Fourier transform you use.

(a)[10pt] Find out the Fourier transform of the signal x(t) given by

(sin(2t))2 /πt2 t 6= 0

x(t) =
4/π t=0

(b)[20pt] Find out the Fourier transform of the signal x(t) given as

1.5
x(t)

0.5

0
-2 -1 0 1 2 3 4
t

FT
Solution) (a)Note that x(t) = π4 · sinc2 ( π2 t) and sinc(t) ←→ rect( 2π
w
). Then, by using the
multiplication property and the time scaling property of the continuous Fourier transform, we
obtain
w
X(jw) = 2Λ( ),
4

w + 1
 if − 1 ≤ w ≤ 0
where Λ(w) = −w + 1 if 0 ≤ w ≤ 1 .

0 if |w| ≥ 1

(b)Observe that x(t + 1) = Λ(t + 2) + 2Λ( 21 t) + Λ(t − 2). Hence, by using the time shifting
property and the time scaling property of the continuous Fourier transform, we have

F{x(t + 1)}(jw) = ej2w F{Λ(t)}(jw) + 4F{Λ(t)}(j2w) + e−j2w F{Λ(t)}(jw)


w w w
= ej2w sinc2 ( ) + 4sinc2 ( ) + e−j2w sinc2 ( ).
2π π 2π
Finally, since F{x(t + 1)}(jw) = ejw X(jw), we have
w w w
X(jw) = ejw sinc2 ( ) + 4e−jw sinc2 ( ) + e−j3w sinc2 ( ).
2π π 2π
Problem 6)[30pt] Consider the causal LTI system described by the differential equation

d2 y(t) dy(t)
+6 + 9y(t) = −x(t).
dt dt

(a)[10pt] Find out the frequency response H(jw) and the impulse response function h(t) for this
system.

(b)[10pt] Find out the magnitude response |H(jw)| of this system and plot it. You should
indicate asymptote, peak, and the value when w = 0 (if they exist).

(c)[10pt] Find out the phase response ∠H(jw) of this system and plot it. You should indicate
asymptote, peak, and the value when w = 0 (if they exist).

Solution) (a)Taking the continuous Fourier transform on each side, we obtain

(jw)2 + 6jw + 9 Y (jw) = −X(jw).




Then, since the system is LTI, we have

Y (jw) 1
H(jw) = =− . (2)
X(jw) (jw + 3)2
FT
Note that e−3t u(t) ←→ 1
jw+3 . Then, by using the ‘differentiation in frequency’ property of the
FT j
continuous Fourier transform, we have te−3t u(t) ←→ d 1
dw ( jw+3 ) = − (jw+3)2 . Hence,

h(t) = −jte−3t u(t).

1
(b)From (2), we have |H(jw)| = w2 +9
.

(c)From (2), we have ∠H(jw) = π − 2 arctan( w3 ).

You might also like