ANALOG & DIGITAL COMMUNICATION

SOLUTION ASSIGNMENT 1(B)

(1) An angle-modulated signal has the form

u(t) = 50cos[2πfc t + 4sin4000πt],
where fc = 20 MHz.
(a) Calculate the average transmitted power.
(b) Calculate the peak-phase deviation.
(c) Calculate the peak-frequency deviation.
(d) Classify this signal as an FM or a PM signal? .
Solution - (a): An angle modulated signal essentially has a constant amplitude, we have
A2c 502
P = = = 1250 W.
2 2

Solution - (b): The maximum phase deviation is

∆φmax = k max[|m(t)|] = 4(1) = 4,
where it has been considered that φ(t) = km(t).

Solution - (c): Because φ(t) = km(t), thus in order to calculate ∆fmax = kf max[|m(t)|],
we need the value of kf that can be obtained if know the exact expression
for fi (t) in this case. Thus
1 d
fi (t) = fc + φ(t),
2π dt
1 d
= fc + (4sin4000πt),
2π dt
1
= fc + × 4cos(4000πt) × 4000π
2π
= fc + 8000cos(4000πt),
giving us kf = 8000. Therefore, ∆fmax = 8000 × 1 = 8000 Hz.

Solution - (d): If given u(t) is a PM signa, thenl phase deviation constant kp = 4 and
message signal m(t) = sin2000πt. On the other hand, if it is an FM signal,
then it has frequency deviation constant kf = 8000 and message signal in
this case is m(t) = cos2000πt.

(2) Consider a narrowband to wideband FM converter. The output of the

narrowband converter is given by
un (t) = Ac cos (2πfc t + φ(t))
1
 2 SOLUTION ASSIGNMENT 1(B)

with fc = 110 kHz. The peak frequency deviation of the modulated signal
un (t) is 60 Hz, whereas, its bandwidth is 510 Hz. The wideband FM signal
will have carrier frequency of 86 MHz and a βfW = 6. Determine the
(a) Frequency multiplier factor n.
β
(Hint: n = βffW , where βfN is to be calculated for the narrowband
N
signal.)
(b) Two possible local oscillator frequencies.
(c) Center frequency and the bandwidth of Bandpass filter.
Solution - (a): We first evaluate βfN . We note that the peak frequency deviation ∆fmax
is 60 Hz and the bandwidth W of φ(t) is 510 Hz. Therefore,
∆fmax
βfN =
W
60
= = 0.12.
510
This leads us to evaluate frequency multiplier factor n as
βfW 6
n= = = 50.
βfN 0.12
Thus, the carrier frequency at the output of the narrowband FM modulated
signal is
nfc = 50(110, 000) = 5.5 MHz.

Solution - (b): Two possible local oscillator frequencies are calculated based on the fact
that wideband FM signal has a carrier fcW = 86 MHz and nfc = 5.5 MHz.
This leads us to represent possible local oscillator frequencies as
fLO = fcW + nfc = 86 + 5.5 = 91.5 MHz
and
fLO = fcW − nfc = 86 − 5.5 = 80.5 MHz.

Solution - (c): The center frequency of the bandpass filter must be equal to the desired
carrier frequency of the wideband output. Thus, the center frequency of
the bandpass filter is 86 MHz. The bandwidth of the bandpass filter is
evaluated using Carsons rule as
B = 2(βfW + 1)W = 2(6 + 1)(510) = 7140 Hz.

(3) The carrier c(t) = Acos2π105 t is angle modulated (PM or FM) by the
sinusoid signal m(t) = 2cos3000πt. The deviation constants are kp = 1.7
rad/V and kf = 2000 Hz/V.
(a) Calculate βf and βp
(b) Calculate the bandwidth in each case using Carson’s rule.
(c) Analyze the spectrum of the modulated signal in each case. (Consider
only those frequency components that lie within the bandwidth derived
in Part b.)
(d) If the amplitude of m(t) is decreased by a factor of 2, then analyze
your answers to Parts a-c.
 ANALOG & DIGITAL COMMUNICATION 3

Solution - (a):
βp = kp max[|m(t)|] = 1.7 × 2 = 3.4
kf max[|m(t)| 2000 × 2
βf = = ≈ 2.7.
fm 1500

Solution - (b): Using Carson’s rule, the bandwidth in both the cases will be
Bp = 2(βp + 1)fm = 2(3.4 + 1)1500 = 13200 Hz
Bf = 2(βf + 1)fm = 2(2.67 + 1)1500 = 11010 Hz.
Solution - (c): The PM modulated signal can be written
∞
X
u(t) = AJn (3.4)cos(2π(105 + n1500)t).
n=−∞

For the spectrum, we will consider those components whose frequencies lie
in the interval [105 − 6.6 × 103 , 105 + 6.6 × 103 ]. The separation between
two adjacent components will be 1.5 kHz. Thus, we need following Bessel
function values to cover the mentioned bandwidth.
J0 (3.4) = −0.3643, J1 (3.4) = 0.1792, J2 (3.4) = 0.4697, J3 (3.4) = 0.3734, J4 (3.4) = 0.1892.
For FM modulated signal can be written
u(t) = A cos(2πfc t + βf sin3000πt)
∞
X
= AJn (2.7)cos(2π(105 + n1500)t + φn ).
n=−∞

For the spectrum, we will consider those components whose frequencies lie
in the interval [105 − 5.5 × 103 , 105 + 5.5 × 103 ]. The separation between
two adjacent components will be 1.5 kHz. Thus, we need following Bessel
function values to cover the mentioned bandwidth.
J0 (2.7) = −0.1424, J1 (2.7) = 0.4416, J2 (2.7) = 0.4696, J3 (2.7) = 0.2540.

Solution - (d): If the amplitude of m(t) is decreased by 2, then m(t) = cos3000πt and
βp = kp max[|m(t)|] = 1.7 × 1 = 1.7
kf max[|m(t)| 2000 × 1
βf = = ≈ 1.3.
fm 1500
Using Carson’s rule, now the bandwidth in both the cases will be
Bp = 2(βp + 1)fm = 2(1.7 + 1)1500 = 8100 Hz
Bf = 2(βf + 1)fm = 2(1.3 + 1)1500 = 6900 Hz.
The PM modulated signal can be written
∞
X
u(t) = AJn (1.7)cos(2π(105 + n1500)t).
n=−∞

For the spectrum, we will consider those components whose frequencies

lie in the interval [105 − 4 × 103 , 105 + 4 × 103 ]. The separation between
two adjacent components will be 1.5 kHz. Thus, we need following Bessel
function values to cover the mentioned bandwidth.
J0 (1.7) = −0.3980, J1 (1.7) = 0.5778, J2 (1.7) = 0.2817.
 4 SOLUTION ASSIGNMENT 1(B)

For FM modulated signal can be written

u(t) = A cos(2πfc t + βf sin3000πt)

∞
X
= AJn (1.3)cos(2π(105 + n1500)t + φn ).
n=−∞

For the spectrum, we will consider those components whose frequencies lie
in the interval [105 − 3.4 × 103 , 105 + 3.4 × 103 ]. The separation between
two adjacent components will be 1.5 kHz. Thus, we need following Bessel
function values to cover the mentioned bandwidth.

J0 (1.3) = 0.6201, J1 (2.7) = 0.5220, J2 (2.7) = 0.1830.

(4) Let m1 (t) and m2 (t) represent two message signals and let u1 (t) and u2 (t)
represent the corresponding modulated versions.
(a) Examine that when the combined message signal m1 (t) + m2 (t) DSB-
SC modulates a carrier Ac cos2πfc t, the result is the sum of the two
DSB amplitude-modulated signals u1 (t) + u2 (t).
(b) Examine that if m1 (t)+m2 (t) frequency modulates a carrier, the mod-
ulated signal is not equal tou1 (t) + u2 (t) .
Solution - (a): For a DSB-SC AM substituting m1 (t) + m2 (t) as message signal results in

u(t) = Ac (m1 (t) + m2 (t)) cos(2πfc t),

= Ac m1 (t)cos(2πfc t) + Ac m2 (t)cos(2πfc t)
= u1 (t) + u2 (t).

Thus linearity is proved.

Solution - (b): In a frequency modulated signal substituting m1 (t) + m2 (t) as message
signal results in
 Z t 
u(t) = Ac cos 2πfc t + 2πkf (m1 (τ ) + m2 (τ ))dτ
−∞
 Z t 
6= Ac cos 2πfc t + 2πkf m1 (τ )dτ
−∞
 Z t 
+ Ac cos 2πfc t + 2πkf m2 (τ )dτ
−∞
= u1 (t) + u2 (t).

Hence, an FM system is nonlinear.

(5) Consider a message signal is the sum of two sinusoids and can be represented
as

m(t) = A cos (2πf1 t) + B cos (2πf2 t) .

Follow the derivations of spectral characteristics of an angle-modulated

signal for a single sinusoid to derive the spectral representation for the
given m(t).
Solution: If

m(t) = A cos (2πf1 t) + B cos (2πf2 t) ,

 ANALOG & DIGITAL COMMUNICATION 5

then for FM, phase φ(t) is given by

Z t
φ(t) = 2πkf m(α)dα,
−∞
= β1 sin (2πf1 t) + β2 sin (2πf2 t) ,
Ak Bk
where β1 = f1f , and β2 = f2f .
The modulated signal u(t) is given by
u(t) = Ac cos (2πfc t + β1 sin (2πf1 t) + β2 sin (2πf2 t)) ,
which can be expressed as
n o
u(t) = Ac Re ejβ1 sin(2πf1 t) ejβ2 sin(2πf2 t) ej2πfc t .
Using the Fourier series
∞
X
ejβ1 sin(2πf1 t) = Jn (β1 )ej2πf1 nt ,
n=−∞
and
∞
X
ejβ2 sin(2πf2 t) = Jm (β2 )ej2πf2 mt .
m=−∞
The modulated output can be written now as
(" ∞ ∞
# )
X X
u(t) = Ac Re Jn (β1 )ej2πf1 nt Jm (β2 )ej2πf2 mt ej2πfc t .
n=−∞ m=−∞
∞
X X∞
= Ac Jn (β1 )Jm (β2 ) cos [2π(fc + nf1 + mf2 )t] .
n=−∞ m=−∞

The signal xc (t) not only contains frequency components at fc + nf1 and
fc + mf2 , but also contains frequency components at fc + nf1 + mf2 for all
combinations of n and m. Therefore, the spectrum of the modulated signal
due to a message signal consisting of the sum of two sinusoids contains
additional components over the spectrum formed by the superposition of
the two spectra resulting from the individual message components. This
example therefore illustrates the nonlinear nature of angle modulation.