Sabancı University

Faculty of Engineering and Natural Sciences

ENS 211 - SIGNALS
Problem Set No. 6
Fall 2021 - 2022

Issued: Wednesday, 16 December 2021 Due: Friday, 31 December 2021, 23:59

Problem 6.1 Find the Fourier transform or inverse Fourier transform for the following ex-
pressions. The result of part (d) shall be stated in terms of X(jω).

FT jω
(a) ←→ e−jw2
5 + jω
FT 1
(b) ←→
ω2 + 2.25
FT sin2 (10w)
(c) ←→ 2w2

d2 x(0.5t − 3) F T
(d) ←→
dt2
FT
(e) ea(t−4) u(4 − t) sin(ω0 t) , when a > 0 ←→

Problem 6.2 Determine the inverse Fourier transform of the following signal:

2 cos(2ω) + 2ω sin(2ω)
X(jω) =
1 + ω2

Problem 6.3 A continuous-time LTI system H is described by the following differential

equation:

d2 y(t) dy(t) dx(t)

2 2
− − y(t) = + x(t)
dt dt dt
(a) Find H(jω), the frequency response of this system.

(b) If the following signal x(t) is given as an input to this system, find y(t), the correspond-
ing output signal.
x(t) = −2e−t + 6e−5t u(t)

(c) [MATLAB] Fill out the following code to calculate and plot the magnitude and phase
of the filter frequency response H(jω), Fourier transform of the input signal X(jω),
and Fourier transform of the output Y (jω) for ω ∈ [−8π, 8π].
 clear all
close all

% Define the frequency vector

w = -8*pi:0.01:8*pi;

% Define the filter’s frequency response

H = ... ./... ;

% Define the Fourier transform of the input signal

X = ... ;

% Define the Fourier transform of the output signal

% using its formula or properties of Fourier Transform
Y = ... ;

% Plot the magnitude and phase responses of the above

% functions versus normalized angular frequency
figure;
subplot(321); plot(w/(2*pi), ... ); grid
title(’Magnitude Response of X(j\omega)’);
subplot(322); plot(w/(2*pi), ... ); grid
title(’Phase of X(j\omega)’);

subplot(323); plot(w/(2*pi), ... ); grid

title(’Magnitude Response of H(j\omega)’);
subplot(324); plot(w/(2*pi), ... ); grid
title(’Phase of H(j\omega)’);

subplot(325); plot(w/(2*pi), ... ); grid

xlabel(’Frequency (Hz)’);
title(’Magnitude of Y(j\omega)’);
subplot(326); plot(w/(2*pi), ... ); grid
xlabel(’Frequency (Hz)’);
title(’Phase of Y(j\omega)’);

(d) What kind of filter does system H represent? Analyze the plots and explain how
the system acts on the gain and phase of the CTFT of the input signal to generate
the corresponding CTFT components of the output signal. Here, CTFT stands for the
continuous-time Fourier transform.

sin(wco t)
Problem 6.4 A continuous-time LTI system H with an impulse response of h(t) =
πt
receives the following input:
n=+∞
X n
x(t) = δ(t − )
n=−∞
4
In other words, x(t) is an impulse train with a period of T0 = 0.25 seconds.

(a) Find X(jω), the Fourier transform of this input. Then plot it over −26π ≤ ω ≤ +26π.
(b) If wco is equal to 20π rad/s, find and sketch H(jω), the frequency response of this
system. What kind of filter does it represent?
(c) Find y(t), the corresponding output of this system using results and plots obtained
from parts (a) and (b).
(d) Find the range of all possible values for the cutoff frequency wco for which the output
signal would only have a DC component, i.e. y(t) would be a constant signal.

Problem 6.5 [No Submission] (Ch 12: LTI Systems) Suppose that an ideal bandpass filter
has a passband from ωco1 = 100π to ωco2 = 300π. Show that the impulse response of this
filter can be represented as the convolution of the impulse responses of an ideal highpass
filter and an ideal lowpass filter as in
 
sin(100πt) sin(300πt)
hbp (t) = δ(t) − ∗ (1)
πt πt

Clearly provide formula and plots of all the steps in your work.

Problem 6.6 [No Submission] (Ch 12: Sampling and Reconstruction) The following block
diagram shows an impulse train used for sampling from the continuous-time signal x(t)
followed by an LTI system used for reconstruction.

𝑥𝑠 (𝑡) LTI System

𝑥(𝑡) 𝑥𝑟 (𝑡)
𝐻𝑟 (𝑗𝜔)

𝑝 𝑡 = 𝛿(𝑡 − 𝑛𝑇𝑠 )
𝑛=−∞

𝑋(𝑗𝜔)and reconstruction system

Figure 1: Block diagram of a sampling
𝐴

The input signal for this system is defined as

x(t) = 2 cos(150πt + π/5) + cos(400πt − π/3)

−80𝜋 0 80𝜋 𝜔
for −∞ < t < ∞.
The frequency response of the lowpass reconstruction filter is also given as
(
Ts |ω| ≤ π/Ts
Hr (jω) =
0 |ω| > π/Ts
 where Ts is the sampling period.

(a) Determine the Fourier transform X(jω) and plot the Fourier transform Xs (jω) for
−2π/Ts < ω < 2π/Ts when ωs = 2π/Ts = 1000π. Carefully label your sketch. What is
the output xr (t) in this case?

(b) Now assume that ωs = 2π/Ts = 500π. Plot Xs (jω) again and determine an equation
for the output signal xr (t).

(c) Is it possible to choose the sampling rate so that

xr (t) = A + 2 cos(150πt + π/5)

where A is constant? If so, what are the values of Ts and A?

Problem 6.7 [No Submission] (Ch 12: Sinewave Amplitude Modulation) Consider the
double-sidebanded AM demodulation system in Figure 2 below:

w(𝑡) LTI System

y(𝑡) v(𝑡)
ℎ 𝑡 , 𝐻(𝑗𝜔)

cos (𝜔𝑐 + Δ 𝑡)

Figure 2: A DSBAM Demodulation System

𝑋(𝑗𝜔)
𝐴
where the filter’s frequency response is given as
(
2 |ω| ≤ ωco
H(jω) =
0 |ω| > ωco
−80𝜋 0 80𝜋 𝜔
and the input signal is equal to
y(t) = x(t) cos(ωc t)
where x(t) has a Fourier transform shown in Figure 3.
We have shown that if x(t) has a bandlimited Fourier transform such that X(jω) = 0 for
|ω| ≥ ωb , and if ωc > ωb , ∆ = 0, and ωb < ωco < (2ωc − ωb ), then the DSB AM signal
y(t) = x(t) cos(ωc t) can be demodulated by the system in Figure 2. That is, for the perfect
adjustment of the demodulator frequency and phase, v(t) = x(t).
 𝑋(𝑗𝜔)
1

−𝜔𝑏 0 𝜔𝑏 𝜔

Figure 3: Fourier transform of x(t)

In the following parts, assume that ∆ << ωc and the input signal x(t) has a bandlimited
Fourier transform represented by the following plot:

(a) Suppose that ∆ ̸= 0. Use Euler’s formula for the cosine function to show that

w(t) = y(t) cos((ωc + ∆)t)

= x(t) cos(ωc t) cos((ωc + ∆)t)
1 1
= x(t) cos((2ωc + ∆)t) + x(t) cos(∆t) (2)
2 2

(b) Use the result in Equation 2 and the modulation property of the Fourier Transform
from Table 11-3 in the book to obtain W (jω) in terms of X(jω). Then plot W (jω).

(c) Assuming that ωco = ωb + ∆, find V (jω) in terms of X(jω). Relate your result to the
plot of W (jω).

(d) From the equation of V (jω) and modulation property of the Fourier transform, obtain
an equation for v(t) in terms of x(t) and ∆.

Problem 6.8 [No Submission] (Ch 12: Sampling and Reconstruction) Refer to Figure 4 for
sampling with an impulse train from an input signal x(t).

𝑥𝑠 (𝑡) LTI System

𝑥(𝑡) 𝑥𝑟 (𝑡)
𝐻𝑟 (𝑗𝜔)

𝑝 𝑡 = 𝛿(𝑡 − 𝑛𝑇𝑠 )
𝑛=−∞

𝑋(𝑗𝜔)
𝐴
Figure 4: Block diagram of a sampling with impulse train and reconstruction system

2π
(a) Use the sampling theorem to choose the sampling rate ws = Ts so that no aliasing
occurs for the input in Figure 5.
−80𝜋 0 80𝜋 𝜔
 Figure 5: Fourier transform of x(t)

(b) Plot the Fourier transform of xs (t) given the sampling rate in part (a).

(c) Design the ideal low-pass filter Hr such that xr (t) = x(t).

(d) If the sampling period was Ts = 0.2s, what would the spectrum of the reconstructed
signal be, given the filter you found in part (c)?