Ryerson University

Department of Electrical, Computer, & Biomedical Engineering

ELE635 - Communication Systems - Winter 2022

Midterm Examination
Tuesday, March 1, 2022 Time: 10:00-11:20 am

Write your name & student ID, and sign below.

Student Name Student ID Signature

Note that the complete midterm exam has two parts. Each student has to do (i) multiple
choice questions online in D2L from 9:00-9:50 am and (ii) problem solving questions from
10:00-11:20 am with given instructions below.
This is a closed-book exam. An aid-sheet is provided with the exam paper. Any assumptions
should clearly be stated for any credit. There are two questions with equal weight.

You should write your answers on A4 (11 in x 8.5 in) papers, scan/copy the answer papers,
and then upload it as a single file in D2L BEFORE 11:30 AM. You should write legibly
(darker pencils/pens are recommended). You should submit a clear scan/copy with legible
writing.

You must attach and submit this cover page with your answers.

You should read the following and sign below:

I solemnly declare that I will complete this examination independently and meet the
requirements stated in the examination.
I solemnly declare that I will complete this examination with full compliance to the Senate
Policy 60 Academic Integrity.

X ————————
• Question 1 [35 points]
Consider the AM signal φAM (t) = [A + m(t)]cos(2πfc t) with its frequency spectrum
Φ(f ) as shown below. Note that the figure shows only Φ(f ) for f ≥ 0 and the
corresponding negative frequencies exist but not shown in the figure. Assume
that the time average of m(t) is 0 (m(t) has no DC component).

(1) Determine fc , A, and m(t).

(2) Determine the modulation index µ, and the power efficiency η, of the signal φAM (t).
(3) Provide the block diagram for a coherent detector, Specify all the parameters and
elements in your design. Suppose that the amplitude of local oscillator is 1.
′
Express the output m (t) at the demodulator.
(4) Provide the block diagram for non-coherent/envelope detector such that its output
will be proportional to m(t).

Φ(f )
5

0.5 f
0 42 44 45 46 48 kHz

Solution:
(1) The AM signal can be written as

φAM (t) = A cos(2πfc t) + m(t) cos(2πfc t)

From the figure, we can obtain that for carrier A cos(2πfc t), fc = 45 kHz and A = 10.
For message m(t), it has two frequencies at 1 kHz and 3 kHz, with

m(t) = 2 cos(2π × 1000t) + 4 cos(2π × 3000t)

2
(2) Since mp = 6,
mp
µ=
= 6/10 = 0.6
A
and the carrier power and sideband power are

Pc = 102 /2 = 50, Ps = 2 × (12 + 0.52 ) × 2 = 5

therefore,
Ps 5
η= = = 1/11 = 0.09
P c + Ps 50 + 5
(3) The diagram is shown in the figure below.

y(t) = A cos2 (2πfc t) + m(t) cos2 (2πfc t)

After y(t) goes through LPF with bandwidth 3 kHz and DC removal, the output is
1
m′ (t) = m(t)
2

ϕAM (t) y(t) m′ (t)

(3) DC-
LPF
block
0-3 kHz

cos(2π × 45, 000t)

(4)

Figure 1: Solution for Question 1, parts (3) and (4).

(4) ED diagram is shown in the solution figure.

3
• Question 2 [35 marks]
The spectrum of the message signal m(t) is shown below where B is 1 kHz. The
fundamental frequency of the periodic function, w(t), is shown below with
τ= T0
2
= 0.2 × 10−3 second. We have following Fourier Transform pair:
( ) ( )
t τ πf τ
∆ → sinc2
τ 2 2

(1) Find the Fourier series coefficients Dn of w(t), and sketch Dn , specify the minimum
n, n ≥ 0 to make Dn = 0.
(2) Express w(t) as a Fourier series in trigonometric form.
(3) When the input is m(t), based on the results from above steps, find the expression
for y1 (t).
(4) Derive the time-domain output y2 (t) if the BPF has unit gain in the passband with
center frequency being 5 kHz, and passband bandwidth being 2 kHz.
(5) Graphically illustrate the steps in (3) and (4) by plotting spectra of Y1 (f ) and
Y2 (f ) for |f | ≤ 10 kHz.

BPF
2kHz

f kHz
5

m(t) y1 (t) y2 (t)

BPF

w(t)

M (f ) w(t)
1
f t (ms)
−B 0 B −T0
τ 0 τ
T0 = 0.4
1 kHz −
2 2
τ /T0 = 1/2

4
Solution:
(1) From following Fourier Transform pair:
( ) ( )
t τ πf τ
∆ → sinc2
τ 2 2

for w(t), its period

1
T0 = 0.4 × 10−3 and f0 = = 2.5 kHz
T0
The Fourier series can be obtained as:
( ) ( )
1 1 τ π · nf0 · τ 1 nπ
Dn = X(nf0 ) = · sinc2 = sinc2
T0 T0 2 2 4 4

Fourier series is plotted below, where D0 = 14 , and Dn = 0 for n = 4, 8, 12, .... The
minimum nonnegative n is 4 to make Dn = 0.

Figure 2: Solution for Question 2, part (1).

(2) The trigonometric form can be written as:

∞
∑ ∞
∑
j2πnf0 t
w(t) = Dn e = C0 + Cn cos(2πnf0 t + ϕn )
n=−∞ n=1

where C0 = D0 = 14 , Cn = 2Dn and ϕn = 0 for n ≥ 1, then

∞ ( )
1 ∑ 1 nπ
w(t) = + sinc2 cos(2πnf0 t)
4 n=1 2 4

5
(3) When the input is m(t),
∞
( )
1 ∑
y1 (t) = m(t)w(t) = m(t) + Cn cos(2πnf0 t)
4 n=1
∞
( ( ) )
1 ∑ 1 nπ
= m(t) + sinc2 cos(2πnf0 t)
4 n=1 2 4

(4) The center frequency of BPF is 5 kHz, which is 2f0 . The passband bandwidth is 2
kHz, which is the bandwidth of the passband message. Therefore, y2 (t) is the term with
n = 2, given as
( )
1 π 2
y2 (t) = m(t) · sinc2 cos(4πf0 t) = 2 m(t) cos(4πf0 t)
2 2 π

(5) Spectrum of Y1 (f ) and Y2 (f ) are shown as

Y1 (f ) BPF: center at 5 kHz, BW 2 kHz

kHz
-7.5 -5 -2.5 0 2.5 5 7.5

Y2 (f )

kHz
-7.5 -5 -2.5 0 2.5 5 7.5

Figure 3: Solution for Question 2, part (5).

6
Trigonometric Identities:

( ) ( ) √ ( ) √
π 1 π 2 2π 3
sin = sin = sin =
3 2 4 2 3 2
sin(α ± β) = sin α cos β ± cos α sin β
cos(α ± β) = cos α cos β ∓ sin α sin β
1
sin α sin β = − [cos(α + β) − cos(α − β)]
2
1
cos α cos β = [cos(α + β) + cos(α − β)]
2
1
sin α cos β = [sin(α + β) + sin(α − β)]
2
1 1
cos2 (α) = [1 + cos(2α)] sin2 (α) = [1 − cos(2α)]
2 2

Euler’s Identity, FS, Relationship for FT and FS:

∞
∑ 1
ejθ = cos θ + j sin θ xp (t) = Dn ej2πnf0 t Dn = X(nf0 )
n=−∞ T0

Fourier Transform Pairs and Properties:

x(t) X(f )
δ(t) 1
1 δ(f )
δ(t − t0 ) exp(−j2πf t0 )
cos(2πf0 t) 1
2
[δ(f − f0 ) + δ(f + f0 )]
sin(2πf0 t) 1
2j
[δ(f − f0 ) − δ(f + f0 )]
exp(j2πf0 t) δ(f − f0 )
( )
t
rect T sinc(πf T )
( )T ( )
t τ πf τ
∆ sinc2
τ 2 ( )2
f
2Bsinc(2πBt) rect 2B 
( )  |f |
2 f 1− W
|f | ≤ W
Bsinc (πBt) ∆ =
2B  0 |f | > W
g(t − t0 ) G(f )e−j2πf t0
g(t)ej2πf0 t G(f − f0 )
∑∞ ∑∞ ( )
n=−∞ δ(t − nT0 ) n=−∞ δ f −
1 n
T0 T0