Mobile

Communications
Homework 5


Problem 1 (CLO-1):
Consider a linearly modulated signal of the form s(t) = Re{sl(t)ej2πfct }, where
!
𝑠! 𝑡 = !!!! 𝑏! 𝑝(𝑡 − 𝑛𝑇! )

and where all pairs of symbols bn and bm are independent for n ≠m and each bn ε {-
3,-1,1,3}, taking the values with equal probability. Furthermore, assume the pulse
p(t) is rectangular with unit height and width Ts . Sketch the power spectral density
(PSD), including labeling its peak height.

Problem 2 (CLO-3):
Let the carrier frequency be 100 MHz and let the symbol period be 1 microsecond.
Consider a transmitted BPSK signal that uses the 25% excess bandwidth Root
Raised Cosine pulses (you can use the definition in Wikipedia, where beta=0.25).
Using MATLAB or your favorite programming language, plot the RF modulated BPSK
signal,

!

𝑥 𝑡 = 𝑥! 𝑝!!" (𝑡 − 𝑛𝑇! ) cos (2𝜋𝑓! 𝑡)

!!!

assuming the symbol sequence 1, -1, -1, 1, -1, -1, -1, 1.

Problem 3 (CLO-3):
Using the same parameters as in Problem 1, plot the RF modulated QPSK waveform

𝑥 𝑡 = !!!! 𝑥! 𝑝!!" (𝑡 − 𝑛𝑇! ) cos 2𝜋𝑓! 𝑡 − !!!! 𝑦! 𝑝!!" (𝑡 − 𝑛𝑇! ) sin 2𝜋𝑓! 𝑡

Use the same x! sequence as in Problem 1, and let the quadrature symbols of the
signal, y! , be -1, -1, 1, 1, 1, -1, 1, -1.

Problem 4:

Suppose the constellation of 8 signals for a communication system is given below

 Problem 3.1:

Suppose that the constellation for the 8 signals for a digital communications link is given
below:

e2
2
1
e1
-3 -1 1 3
-1
-2


a) Find the energy of each symbol and the average symbol energy of this
constellation (a) Give a Gray Code assignment for this constellation.
b) Give a complete union bound for the probability of symbol error.
First Bit
c) If the basis functions are the usual ones for QPSK, give the expression of the Second Bit
symbol waveform as a function of time for the symbol that is indicated by the
Top = 0
arrow.
0 _ _ 0 _ _ 0 _ _ 0 _ _ 01_ 01_ 00_ 00_

Problem 5 (CoveringF CLO-1): B 0 Left = 1 0 Right = 0
i o
1 _ _Transform
Say if the pulse Fourier 1 _ _ 1 P(f)
_ _ shown
1 _ _ below implies 1 1that
_ 1the 1 _pulse
1 0 is
_ a1Nyquist
0_
r t 0 0
Problemyour
pulse or nor? Explain 2.2: answer.
Saysif theBottom = 1Transform P(f) shown below implies that the pulse p(t) is a
t Fourier
pulse
t Pulseoor not a Nyquist0Pulse. Explain.
Nyquist Third Bit _
m P ( f )
Inner = 0
B Ts

i =
011 010 000 001
t
1
0__
__ f
1 1 1 1 1 1 1 0 1
1 0 0 1 0 1
2Ts
_ _1
Ts 2Ts T
Outer = 1 _
s

b) Consider the pulse below,

To test if plotted versus
a P(F) corresponds time
to a Nyquist in must
Pulse, microseconds.
to see thatCould this pulse be a
check _
Nyquist pulse for a binary
P f
m transmission with a 10MHz data rate?1 Explain
a constant. A sketch of the interval between to
1 why or why not
below,
m Ts 2Ts 2Ts
shows that the sum (the blue solid curve) cannot be a constant. Therefore, this P(f)
cannot correspond to a Nyquist Pulse.

P( f )
Ts

t, s

0 .3 0 .2 0 .1 0 0 .1 0 .2 0 .3
f
1 1 1 1
Ts 2Ts 2Ts Ts

3.Problem
Suppose
2.3:that a non-line-of-sight wireless channel at 1.9 GHz has an rms delay
Let twoofsignals
spread 100 ns.be described
Further, suppose the mobile terminal is moving with a speed of
80km/hr. If8 a BPSK signal is transmitted with a bit rate of 20 Mbps, then give a
cos t 0 t TS
complete
s1 (t ) description
TS of the small-scale fading that this signal will undergo.
0 otherwise
6 0 t TS
cos t /4
s 2 (t ) TS
otherwise
Problem 6 (Covers CLO-2):

Two signal points Sa and Sb are shown below

a) Suppose the noise spectral height is N0/2=25/16.
Evaluate the BER if these two signals are used in a
wireless communication link.
b)

Construct an expression of signal labeled Sa in terms of t and Ts.

c) Evaluate the value of K such that e2(t) has a unit norm.