Solved Problems: EE160: Analog and Digital Communications
Solved Problems: EE160: Analog and Digital Communications
SOLVED PROBLEMS
1
Digital communication systems
1. With reference to Fig. 1.2 of the textbook, illustrating the basic elements of a digital com-
munication system, answer the following questions:
Solution:
(a) Source coding is the process of efficiently converting the output of either an analog or a
digital source, with as little or no redundancy, into a sequence of binary digits.
(b) The channel encoder introduces, in a controlled (structured) manner, certain amount
of redundancy that can be used at the receiver to overcome the effects of noise and
interference encountered in the transmission of the signal through the channel. This
serves to increase the reliability of the received data and improves the quality of the
received signal. The channel decoder attempts to reconstruct the original information
sequence from knowledge of the code used by the channel encoder, the digital modulation
scheme and the redundancy contained in the received sequence.
(c) The digital modulator serves as the interface to the communications channel. Its pri-
mary purpose is to map the information sequence into signal waveforms. The digital
demodulator processes the corrupted transmitted waveform and reduces each waveform
to a single number that represents an estimate of the transmitted data symbol. If this
number is quantized into more levels that those used in the modulator, the demodulator
is said to produced a soft output. In this case, the channel decoder is known as a soft-
decision decoder. Otherwise, the demodulator produces hard outputs that are processed
by a hard-decision decoder.
(d) The performance of a digital communication system is typically measured by the fre-
quency with which errors occur in the reconstructed information sequence. The proba-
bility of a symbol error is a function of the channel code and modulation characteristics,
the waveforms used, the transmitted signal power, the characteristics of the channel —
e.g., noise power — and the methods of demodulation and channel decoding.
2. What are the dominant sources of noise limiting performance of communication systems in
the VHF and UHF bands?
Solution: The dominant noise limiting performance of communication systems in the VHF
and UHF bands is thermal noise generated in the front end of the receiver.
3. Explain how storing data on a magnetic or optical disk is equivalent to transmitting a signal
over a radio channel.
Solution: The process of storing data on a magnetic tape, magnetic disk or optical disk is
equivalent to transmitting a signal over a wired or wireless channel. The readback process
and the signal processing used to recover the stored information is equivalent to the functions
performed by a communications system to recover the transmitted information sequence.
2
4. Discuss the advantages and disadvantages of digital processing versus analog processing. Do
a web search. An interesting, albeit non-technical, discussion was found at
http://www.usatoday.com/tech/bonus/2004-05-16-bonus-analog x.htm
Solution: A digital communications system does not accumulate errors. Analog signals are
prone to interference and noise. There is no equivalent in an analog system to the correction
of errors. However, a digital system degrades the quality of the original signal thorugh
quantization (analog-to-digital conversion). Also, a digital system requires more bandwidth
than an analog system and, in general, relatively complex synchronization circuitry is required
at the receiver. Analog systems are very sensitive to temperature and component value
variations. It should be noted that no digital technology is used today in the front end of
a transmitter and receiver (RF frequency bands of 1GHz and above), where mixers, channel
filters, amplifiers and antennas are needed. The world today is still a mix of analog and digital
components and will continue to be so for a long time. A key feature of digital technology
is programmability, which has resulted in new concepts, such as software-defined radios and
cognitive radio communications systems.
where xe (t) and xo (t) are the even and odd parts of x(t), defined as
x(t) + x(−t)
xe (t) = ,
2
x(t) − x(−t)
xo (t) = .
2
Solution: It follows directly from the uniqueness of the decomposition of a real signal in an
even and odd part. Nevertheless for a real periodic signal
∞
a0 n n
x(t) = + an cos(2π t) + bn sin(2π t)
2 n=1
T0 T0
The even part of x(t) is
x(t) + x(−t)
xe (t) =
2 ∞
1 n n
= a0 + an (cos(2π t) + cos(−2π t))
2 T0 T0
n=1
n n
+bn (sin(2π t) + sin(−2π t))
T0 T0
∞
a0 n
= + an cos(2π t)
2 T0
n=1
3
The last is true since cos(θ) is even so that cos(θ) + cos(−θ) = 2 cos θ whereas the oddness of
sin(θ) provides sin(θ) + sin(−θ) = sin(θ) − sin(θ) = 0.
Similarly, the odd part of x(t) is
x(t) − x(−t)
xo (t) =
2
∞
n
= bn sin(2π t)
T0
n=1
6. Determine the Fourier series expansion of the sawtooth waveform, shown below
x(t)
t
-3T -2T -T T 2T 3T
-1
Solution: The signal is periodic with period 2T . Since the signal is odd we obtain x0 = 0.
For n = 0
T T
1 n
−j2π 2T t 1 t −j2π n t
xn = x(t)e dt = e 2T dt
2T −T 2T −T T
T
1 n
= 2
te−jπ T t dt
2T −T
1 jT −jπ n t T 2 −jπ n t T
= te T + e T
2T 2 πn π 2 n2 −T
2
1 jT −jπn T 2 −jπn jT 2 jπn T 2 jπn
= e + 2 2e + e − 2 2e
2T 2 πn π n πn π n
j
= (−1)n
πn
7. By computing the Fourier series coefficients for the periodic signal ∞ n=−∞ δ(t − nTs ), show
that
∞ ∞
1 jn 2πt
δ(t − nTs ) = e Ts .
n=−∞
Ts n=−∞
Using this result, show that for any signal x(t) and any period Ts , the following identity holds
∞ ∞
1 n 2πt
x(t − nTs ) = X ejn Ts .
n=−∞
Ts n=−∞ Ts
From this, conclude the following relation, known as Poisson’s sum formula:
∞ ∞
1 n
x(nTs ) = X .
n=−∞
Ts n=−∞ Ts
4
Solution:
∞
∞
∞
1 n
x(t − nTs ) = x(t) δ(t − nTs ) = x(t) ej2π Ts t
n=−∞ n=−∞
Ts n=−∞
∞
1 −1 n
= F X(f ) δ(f − )
Ts n=−∞
T s
∞
1 −1 n n
= F X δ(f − )
Ts n=−∞
Ts Ts
∞
1 n n
= X ej2π Ts t
Ts n=−∞ Ts
where
∆ 1, |t| ≤ 12 ;
Π(t) = ,
0, otherwise.
and shown in the figure below:
p1(t)
t
-1 1
Solution: Using the Fourier transform pair Π(t) ⇐⇒ sinc(f ) and the time scaling property
(from the table of Fourier transform properties), we have that
t
Π ⇐⇒ 2 sinc(2f ).
2
5
9. Determine the Fourier series expansion of the periodic waveform given by
∞
p(t) = p1 (t − 4n),
n=−∞
… …
t
-5 -3 -1 1 3 5
(Hint: Use the Fourier transform P1 (f ) found in the previous problem, and the following
equation to find the Fourier coefficients: pn = T1 F1 ( Tn ).)
Solution: The signal p(t) is periodic with period T = 4. Consequently, the Fourier series
expansion of p(t) is
∞
π
p(t) = pn exp j t n ,
n=−∞
2
where
n
1 1 n n
pn = P1 = sinc 2( + 4)) − sinc 2( − 4)) .
4 4 4j 4 4
10. Classify each of the following signals as an energy signal or a power signal, by calculating the
energy E, or the power P (A, θ, ω and τ are real positive constants).
Solution:
(b) Neither:
T
(Aτ )2
E2 = lim √ √ dt → ∞,
T →∞ −T τ + jt τ − jt
and T
1 (Aτ )2
P2 = lim √ dt = 0.
T →∞ 2T −T τ 2 + t2
6
(c) Energy: ∞
3A2 τ 5
E3 = A2 t4 exp(−2t/τ ) dt = .
0 4
(d) Energy:
τ /2 τ
E4 = 2 (2)2 dt + (1)2 dt = 5τ.
0 τ /2
Solution:
x1(t) x3(t)
1
1
t t
-5/2 1
1/2 -1
x2(t) x4(t)
1
1
t t
4 1 4/3
1/2
-1
x5(t)
-3/2 3/2
12. Classify each of the signals in the previous problem into even or odd signals, and determine
the even and odd parts.
Solution:
The signal xi (t), for 1 ≤ i ≤ 4, is neither even nor odd. The signal x5 (t) is even symmetric.
7
For each signal xi (t), with 1 ≤ i ≤ 4, the figures below are sketches of the even part xi,e (t)
and the odd part xi,o (t). Evidently, x5,e = x5 (t) and x5,o (t) = 0.
x1,e(t)
1/2
t
-5/2 5/2
1/2 1/2
x1,o(t)
1/2
-5/2 5/2
-1/2
1/2
1/2
x2,e(t)
1/2
t
-4 4
1/2 1/2
x2,o(t)
1/2
-4 t
-1/2 4
1/2
1/2
8
x3,e(t)
1/2
-1 1
-1/2
x3,o(t)
1/2
-1 1
-1/2
x4,e(t)
1/2
t
-4/3 -1 1 4/3
-1/2
x4,o(t)
1/2
t
-4/3 -1 1 4/3
-1/2
9
13. Generalized Fourier series
(a) Given the set of orthogonal functions
4 [t − (2n − 1)T /8]
φn (t) = Π , n = 1, 2, 3, 4,
T
sketch and dimension accurately these functions.
(b) Approximate the ramp signal
t t − T /2
x(t) = Π
T T
by a generalized Fourier series using these functions.
(c) Do the same for the set
2 [t − (2n − 1)T /4]
φn (t) = Π , n = 1, 2.
T
(d) Compare the integral-squared error (ISE) N for both parts (b) and (c). What can you
conclude about the dependency of N on N ?
Solution:
(a) These are unit-amplitude rectangular pulses of width T /4, centered at t = T /8, 3T /8, 5T /8,
and 7T /8. Since they are spaced by T /4, they are adjacent to each other and fill the
interval [0, T ].
(b) Using the expression for the generalized Fourier series coefficients,
1
Xn = x(t)φn (t)dt,
cn T
where
T
cn = |φn (t)|2 dt = ,
T 4
we have that
1 3 5 7
, X2 = ,
X1 = X3 = , X4 = .
8 8 8 8
Thus, the ramp signal is approximated by
4
1 3 5 7
x4 (t) = Xn φn (t) = φ1 (t) + φ2 (t) + φ3 (t) + φ4 (t), 0 ≤ t ≤ T.
8 8 8 8
n=1
1
x(t)
x4(t)
0.5
t
T/2 T
10
(c) These are unit-amplitude rectangular pulses of width T /2 and centered at t = T /4 and
3T /4. We find that X1 = 1/4 and X2 = 3/4. The approximation is shown in the figure
below:
1
x(t)
x2(t)
0.5
t
T/2 T
11
where u = −τ . Rename variables to obtain
T
1
R(τ ) = lim x(β)x(τ + β) dβ.
T →∞ 2T −T
15. A filter has amplitude and phase responses as shown in the figure below:
|H(f)|
4
f
-100 -50 0 50 100
H(f)
π/2
f
-100 -50 50 100
-π/2
Find the output to each of the inputs given below. For which cases is the transmission
distortionless? For the other cases, indicate what type of distorsion in imposed.
Solution: Note that the four input signals are of the form xi (t) = a cos(2πf1 t)+ b cos(2πf2 t),
for i = 1, 2, 3, 4. Consequently, their Fourier transforms consist of four impulses:
a b
Xi (f ) = [δ(f + f1 ) + δ(f − f1 )] + [δ(f + f2 ) + δ(f − f2 )] , i = 1, 2, 3, 4.
2 2
With this in mind, we have the following
12
16. Determine the Fourier series expansion of the following signals:
(a) x4 (t) = cos(t) + cos(2.5t)
(b) x8 (t) = | cos(2πf0 t)|
(c) x9 (t) = cos(2πf0 t) + | cos(2πf0 t)|
Solution:
(a) The signal cos(t) is periodic with period T1 = 2π whereas cos(2.5t) is periodic with
period T2 = 0.8π. The ratio T1 /T2 = 5/2 and LCM (2, 5) = 10. It follows then that
cos(t) + cos(2.5t) is periodic with period T = 2(2π) = 5(0.8π) = 4π. The trigonometric
Fourier series of the even signal cos(t) + cos(2.5t) is
∞
n
cos(t) + cos(2.5t) = αn cos(2π t)
n=1
T0
∞
n
= αn cos( t)
2
n=1
By equating the coefficients of cos( n2 t) of both sides we observe that an = 0 for all n
unless n = 2, 5 in which case a2 = a5 = 1. Hence x4,±2 = x4,±5 = 12 and x4,n = 0 for all
other values of n.
1
(b) The signal x8 (t) is real, even symmetric, and periodic with period T0 = 2f0 . Hence,
x8,n = a8,n /2 or
1
4f0
x8,n = 2f0 cos(2πf0 t) cos(2πn2f0 t)dt
− 4f1
0
1 1
4f0 4f0
= f0 cos(2πf0 (1 + 2n)t)dt + f0 cos(2πf0 (1 − 2n)t)dt
− 4f1 − 4f1
0 0
1 4f1 1 4f1
= sin(2πf0 (1 + 2n)t) 10 + sin(2πf0 (1 − 2n)t) 10
2π(1 + 2n) 4f 2π(1 − 2n) 4f0
n
0
(−1) 1 1
= +
π (1 + 2n) (1 − 2n)
(c) The signal x9 (t) = cos(2πf0 t) + | cos(2πf0 t)| is even symmetric and periodic with period
T0 = 1/f0 . It is equal to 2 cos(2πf0 t) in the interval [− 4f10 , 4f10 ] and zero in the interval
[ 4f10 , 4f30 ]. Thus
1
4f0
x9,n = 2f0 cos(2πf0 t) cos(2πnf0 t)dt
− 4f1
0
1 1
4f0 4f0
= f0 cos(2πf0 (1 + n)t)dt + f0 cos(2πf0 (1 − n)t)dt
− 4f1 − 4f1
0 0
1 4f1 1 4f1
= sin(2πf0 (1 + n)t) 10 + sin(2πf0 (1 − n)t) 10
2π(1 + n) 4f0 2π(1 − n) 4f0
1 π 1 π
= sin( (1 + n)) + sin( (1 − n))
π(1 + n) 2 π(1 − n) 2
13
Thus x9,n is zero for odd values of n unless n = ±1 in which case x9,±1 = 12 . When n is
even (n = 2) then
(−1) 1 1
x9,2 = +
π 1 + 2 1 − 2
17. A triangular pulse can be specified by
t + 1, −1 ≤ t ≤ 0;
Λ(t) =
−t + 1, 0 ≤ t ≤ 1.
Solution:
(a) Sketch:
x(t)
… …
t
-4 -3 -2 -1 1 2 3 4
(b) The signal x(t) is periodic with T0 = 3. The Fourier series coefficients are obtained from
the Fourier transform, XT0 (f ), of the truncated signal xT0 (t) as
1
xn = XT0 (f )|f = Tn .
T0 0
In this case,
xT0 (t) = Λ(t) ⇐⇒ XT0 (f ) = sinc2 (f ).
Consequently,
1
n
xn = sinc2 .
3 3
14
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
−10 −8 −6 −4 −2 0 2 4 6 8 10
18. For each case below, sketch the signal and find its Fourier series coefficients.
Solution:
(a) The signals cos(2πt) and cos(3πt) are periodic with periods T1 = 1 and T2 = 23 , respec-
tively. The period T0 of x(t) is the “least common multiple” of T1 and T2 :
2 1 6
T0 = “lcm” 1, = lcm (3, 2) = = 2.
3 3 3
Sketch:
15
2.5
1.5
0.5
x(t) 0
−0.5
−1
−1.5
−2
−2.5
−2 −1.5 −1 −0.5 0 0.5 1 1.5 2
t
0.8
0.6
y(t)
0.4
0.2
0
−0.6 −0.4 −0.2 0 0.2 0.4 0.6
t/T0
16
Note that y(t) is periodic with period T1 = T0 /2. A fortunate choice of a truncated
signal yT1 (t), over an interlval of length T1 seconds, is given by
2t
yT1 (t) = cos(2πf0 t) Π ,
T0
1
yn = YT (f )|f = n = 2n
T1 1 T1 T0
1 1 1
= sinc (2n + 1) + sinc (2n − 1) . (2)
2 2 2
The above result can be further simplified by using the definition of the sinc function,
sinc(x) = sin(πx)
πx , noticing that
π +1, n = 0, 2, 4, · · ·
sin (2n + 1) =
2 −1, n = 1, 3, 5, · · ·
= (−1)n ,
and using the odd symmetry of the sine function for negative values of n. This gives
(details omitted):
(−1)n 1 1
yn = + . (3)
π 1 + 2n 1 − 2n
You are invited to verify that both (2) and (3) yield the same result. For example, you
can do this using Matlab with the commands:
n=-9:1:9;
subplot(2,1,1)
stem(n,0.5*(sinc((2*n+1)/2)+sinc((2*n-1)/2)))
subplot(2,1,2)
stem(n,((-1).^n/pi) .* ( (1./(2*n+1)) + (1./(1-2*n)) ) )
17
Equation (2)
0.8
0.6
0.4
yn
0.2
−0.2
−10 −8 −6 −4 −2 0 2 4 6 8 10
n
Equation (3)
0.8
0.6
0.4
yn
0.2
−0.2
−10 −8 −6 −4 −2 0 2 4 6 8 10
n
(c) The sketch of z(t) is shown in the following page. Here the period is T0 . The truncated
signal is
2t
zT0 (t) = cos(2πf0 t) Π ,
T0
with Fourier transform
T0 T0 T0
ZT0 (f ) = sinc (f + f0 ) + sinc (f − f0 ) .
4 2 2
As before, there is a simplification possible (but not necessary!) using the definition of
the sinc function. This gives, z±1 = 12 and
(−1) 1 1
zn = + , n = 2, integer.
π 1 + 2 1 − 2
18
2
1.8
1.6
1.4
1.2
z(t)
0.8
0.6
0.4
0.2
0
−1 −0.8 −0.6 −0.4 −0.2 0 0.2 0.4 0.6 0.8 1
t/T0
0.7
0.6
0.5
0.4
n
0.3
z
0.2
0.1
−0.1
−8 −6 −4 −2 0 2 4 6 8
n
19. Sketch the signal x(t) whose Fourier series coefficients are given by
1, n = 0;
1
n = −2, +2;
2,
xn = + 4 j, n = −4;
1
− 14 j, n = +4;
0, elsewhere.
19
Solution: We are given the Fourier series coefficients. Therefore,
∞
n
j2π T0
t
x(t) = xn e
n=−∞
1 −j2π 2 2 1 j2π 4 4
t j2π t t −j2π t
= 1+ e T0
+e T0
+ e T0
−e T0
2 4j
1
= 1 + cos(4πf0 t) + sin(8πf0 t).
2
2.5
1.5
x(t)
0.5
−0.5
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
t/T0
20. Modify the Matlab script example1s05.m in the web site, to compute the Fourier series
coefficients xn of an even-symmetric train of rectangular pulses of duty cycle equal to 0.12
over the range −50 ≤ n ≤ 50. Attach a printout of the resulting plot.
Solution: Using the Matlab script homework3s05.m (available in the web site) we obtain:
20
0.12
0.1
0.08
0.06
0.04
xn
0.02
−0.02
−0.04
−0.06
−50 −40 −30 −20 −10 0 10 20 30 40 50
n
21. Let xn and yn denote the Fourier series coefficients of x(t) and y(t), respectively. Assuming
the period of x(t) is T0 , express yn in terms of xn in each od the following cases:
Solution:
21
22. Determine whether these signals are energy-type or power-type. In each case, find the energy
or power spectral density abd also the energy or power content of the signal.
Solution:
(a) x(t) = e−αt u(t). The spectrum of the signal is X(f ) = α+j2πf
1
and the energy spectral
density
1
GX (f ) = |X(f )|2 = 2
α + 4π 2 f 2
Thus,
1 −α|τ |
RX (τ ) = F −1 [GX (f )] = e
2α
The energy content of the signal is
1
EX = RX (0) =
2α
(b) x(t) = sinc(t). Clearly X(f ) = Π(f ) so that GX (f ) = |X(f )|2 = Π2 (f ) = Π(f ). The
energy content of the signal is
∞ ∞ 1
2
EX = GX (f )df = Π(f )df = Π(f )df = 1
−∞ −∞ − 12
(c) x(t) = ∞n=−∞ Λ(t − 2n). The signal is periodic and thus it is not of the energy type.
The power content of the signal is
1
1 1 1 0
Px = |x(t)| dt =
2 2
(t + 1) dt + (−t + 1)2 dt
2 −1 2 −1 0
0 1
1 1 3 1 1 3
= 2
t +t +t + t − t + t
2
2 3 2 3 −1 0
1
=
3
The same result is obtain if we let
∞
n
SX (f ) = |xn |2 δ(f − )
n=−∞
2
22
with x0 = 12 , x2l = 0 and x2l+1 = 2
π(2l+1) (see Problem 2.2). Then
∞
PX = |xn |2
n=−∞
∞
1 8 1 1 8 π2 1
= + 2 4
= + 2
=
4 π (2l + 1) 4 π 96 3
l=0
(d)
T T
2 2 T
EX = lim |u−1 (t)| dt = lim
2
dt = lim =∞
T →∞ − T T →∞ 0 T →∞ 2
2
1
Hence, the signal is of the power type and its power content is 2. To find the power
spectral density we find first the autocorrelation RX (τ ).
T
1 2
RX (τ ) = lim u−1 (t)u−1 (t − τ )dt
T →∞ T − T2
T
1 2
= lim dt
T →∞ T τ
1 T 1
= lim ( − τ ) =
T →∞ T 2 2
However, the squared term on the right side is bounded away from zero so that SX (f )
is ∞. The signal is not of the power type either.
23
x(t)
0.5
… … t
-2.5 -1 1 2.5
-0.5
Solution:
(a) T0 = 5/2 and
t 1 2t
xT0 (t) = Λ − Π .
2 2 5
As a result
5 5f
XT0 (f ) = 2 sinc (2f ) − sinc
2
.
4 2
Fourier series coefficients:
1 n 2 2 4 2 5
xn = XT0 = · 2 sinc n − · sinc (n)
T0 T0 5 5 5 4
3 4 4n
= δ(n) + sinc2 .
10 5 5
Fourier transform:
∞
∞
3 4 2 4n 2 3 8 2 4n 2
X(f ) = δ(n) + sinc δ f− n = δ(f )+ sinc δ f− n
n=−∞
10 5 5 5 10 5 5 5
n=1
X(f)
0.3
f
0.4 0.8
-3 -2 -1 1 2 3
6
6 2
(b) H(f ) = 2 Π(2f ). Therefore, Y (f ) = 10 δ(f ) and Py = 10 = 0.36.
24. Matlab problem. This problem needs the Matlab script homework1f04.m, available in the
class web site. The script uses the fast Fourier transform (FFT) to compute the discrete
amplitude spectrum of the periodic signal x(t) = 2sin(100πt) + 0.5cos(200πt) − cos(300πt).
24
(a) Run the script homework1f04.m. To do this, you must save the file to a local directory,
change the working directory in MATLAB to that location, and enter homework1f04 at
the prompt in the command window. You will be requested to enter your student ID
number. The script produces a figure that you are required to either print or sketch.
Also, record in your solution the value of the magic number that will appear in the
command window after execution of the script.
(b) Verify the results of part (a) by computing the Fourier series coefficients of x(t).
Solution:
(a)
Signal
3
−1
−2
−3
−4
0 5 10 15 20 25 30 35 40 45 50
Time (ms)
0.6
0.4
0.2
0
−10 −8 −6 −4 −2 0 2 4 6 8 10
n
25
25. Determine the Fourier transform of each of the following signals:
Solution:
Note that
∞ 1 f + 12
2
Π(f ) Λ(f ) = Π(θ)Λ(f − θ)dθ = Λ(f − θ)dθ = Λ(v)dv,
−∞ − 21 f − 12
A plot of T (f ) is shown in the following figure, and was produced with Matlab script
proakis salehi 2 10 4.m, available in the web site of the class.
26
3
Fourier transform of sinc (t)
0.8
0.7
0.6
0.5
Amplitude
0.4
0.3
0.2
0.1
0
−2.5 −2 −1.5 −1 −0.5 0 0.5 1 1.5 2 2.5
Frequency (Hz)
26. Matlab problems. These two problems needs the following three Matlab scripts: homework2af04.m,
rectpulse.m and homework2bf04.m, available in the class web site.
(a) The scripts homework2af04.m and rectpulse.m plot the amplitude spectrum of the
Fourier transform X(f ) of the signal
t
x(t) = Π .
τ
Run the script homework2af04.m. You will be requested to enter the width τ of the
pulse. Use values of τ equal to 0.1 and 0.2. Print or sketch the corresponding figures.
Based on the scaling property, discuss the results.
(b) The scripts homework2bf04.m uses the inverse fast Fourier transform (IFFT) to compute
numerically the signal associated with a spectrum consisting of pair of impulses:
1 1
X(f ) = δ (f + F c) + δ (f − F c) .
2 2
Run the script homework2a.m and print or sketch the corresponding figures.
27
Solution:
Rectangular pulse
0.8
0.6
0.4
0.2
Amplitude spectrum
0.1
0.08
0.06
0.04
0.02
0
−60 −50 −40 −30 −20 −10 0 10 20 30 40 50 60
Frequency (Hz)
Rectangular pulse
0.8
0.6
0.4
0.2
Amplitude spectrum
0.2
0.15
0.1
0.05
0
−30 −25 −20 −15 −10 −5 0 5 10 15 20 25 30
Frequency (Hz)
28
The plots agree with the theoretical expression:
t
F Π = τ sinc(τ f ).
τ
(b)
0.5
Spectrum Amplitude
0.4
0.3
0.2
0.1
−0.1
−60 −40 −20 0 20 40 60
Normalized frequency
1
Signal Amplitude
0.5
−0.5
−1
10 20 30 40 50 60
Time (samples)
and
−1 1 f
F Π = sinc(αt),
α α
As a result,
1
sinc(αt) sinc(βt) = sinc(αt).
β
29
28. Find the output y(t) of an LTI system with impulse response h(t) = e−αt u(t) when driven by
the input x(t) = e−βt u(t). Treat the special case α = β separately. Determine if y(t) is an
energy signal or a power signal by finding the energy E or the power P .
1
If α = β then X(f ) = H(f ) = α+j2πf . In this case
1
y(t) = F −1 [Y (f )] = F −1 [( )2 ] = te−αt u−1 (t)
α + j2πf
29. Can the response of an LTI system to the input x(t) = sinc(t) be y(t) = sinc2 (t)? Justify
your answer.
Solution: The answer is no. Let the response of the LTI system be h(t) with Fourier
transform H(f ). Then, from the convolution theorem we obtain
Y (f ) = H(f )X(f ) =⇒ Λ(f ) = Π(f )H(f )
This is impossible since Π(f ) = 0 for |f | > 1
2 whereas Λ(f ) = 0 for 1
2 < |f | ≤ 1.
Solution:
30
(1) XT0 (f ) = sinc2 (f ), and T0 = 2. Therefore,
n
1 2 n 1
xn = sinc = sinc2 .
T0 T0 2 2
1
… …
t
-3 -2 -1 1 2 3
It follows that X2 (f ) = δ(f ). The signal can also be consider as periodic with period
T0 = 1 and therefore xn = δ(n). In other words, x0 = 1 and xn = 0, ∀n = 0.
30
25
20
15
dBm
10
−5
−10
−15 −10 −5 0 5 10 15
Frequency (Hz)
31
32. MATLAB problem
Based on the script homework3f04.m of the previous problem, write a Matlab script to find
numerically the energy E1 contained in the first “lobe” of the energy spectral density, that is,
1
E1 = G(f )df,
−1
Solution:
E1 = 0.902823 Joules. This was produced by the following script:
% Name: homework3_2.m
% For the EE160 students of San Jose State University in Fall 2004
N = 4096;
f = -1:1/N:1;
G = sinc(f).^2;
E = sum(G)/N;
fprintf(’The energy in the main lobe of G(g) is %8.6f Joules\n’, E);
33. Sketch carefully the following signals and their Fourier transform
(a) x1 (t) = Π 3t .
12
(b) x2 (t) = Λ 2 (t − 3) .
Solution:
2
2
(a) X1 (f ) = 3 sinc 3 f .
X1(f)
x1(t)
2/3
1
t f
-1/3 1/3 -9/2 -2 -3/2 3/2 3 9/2
x2(t) |X2(f)|
2
1
t f
2 3 4 -3/2 -1 -1/2 1/2 1 3/2
32
34. MATLAB problem.
Download and execute the Matlab script homework4f04.m from the web site of the class. The
script illustrates two signals in the time domain and their corresponding Fourier transforms.
This serves to verify that the time variation is proportional to the bandwidth. Sketch or print
the plots.
Solution:
x1(t) |X1(f)|
2
1
1 0.8
0.6
0
0.4
−1
0.2
−2 0
−0.2 0 0.2 −10 −5 0 5 10
x2(t) |X2(f)|
2
0.6
0.5
1
0.4
0 0.3
0.2
−1
0.1
−2 0
−0.2 0 0.2 −10 −5 0 5 10
Time (s) Frequency (Hz)
2 2
1
1
t t 1 2 t
-2 -1 0
-2 0 2 -2 -1 0 1 2
-1
33
Solution:
(a) Write x1 (t) = 2 Π( 4t ) − 2 Λ( 2t ). Then
t t
X1 (f ) = F 2 Π −F 2Λ = 8 sinc(4f ) − 4 sinc2 (2f )
4 2
Solution:
F[x(t) y(t)] = F[x(t)] · F[y(t)] = X(f ) · Y (f )
Thus
sinc(t) sinc(t) = F −1 [F[sinc(t) sinc(t)]]
= F −1 [F[sinc(t)] · F[sinc(t)]]
= F −1 [Π(f ) · Π(f )] = F −1 [Π(f )]
= sinc(t)
Solution:
(a)
∞ ∞
−αt
e sinc(t)dt = e−αt u−1 (t)sinc(t)dt
0 −∞
∞ 1
1 2 1
= Π(f )df = df
−∞ α + j2πf − 21 α + j2πf
1 1/2 1 α + jπ 1 π
= ln(α + j2πf )−1/2 = ln( ) = tan−1
j2π j2π α − jπ π α
34
(b)
∞ ∞
−αt
e 2
sinc (t)dt = e−αt u−1 (t)sinc2 (t)dt
0
−∞
∞
1
= Λ(f )df df
−∞ α + j2πf
0 1
f +1 −f + 1
= df + df
−1 α + jπf 0 α + jπf
But x
a+bx dx = x
b − a
b2
ln(a + bx) so that
0
∞
f α
−αt
e 2
sinc (t)dt = ( + 2 ln(α + j2πf ))
0 j2π 4π −1
1 1
f α 1
−( + 2 ln(α + j2πf )) + ln(α + j2πf )
j2π 4π 0 j2π −1
1 2π α α
= tan−1 ( ) + 2 ln( √ )
π α 2π α2 + 4π 2
(c)
∞ ∞
−αt
e cos(βt)dt = e−αt u−1 (t) cos(βt)dt
0 −∞
1 ∞ 1 β β
= (δ(f − ) + δ(f + ))dt
2 −∞ α + j2πf 2π 2π
1 1 1 α
= [ + ]= 2
2 α + jβ α − jβ α + β2
Solution:
A f
x(t) = A sinc(1000πt) ⇐⇒ X(f ) =
Π( )
1000 1000
Thus the bandwidth W of x(t) is 1000/2 = 500. Since we sample at fs = 2000 there is a gap
between the image spectra equal to
2000 − 500 − W = 1000
The reconstruction filter should have a bandwidth W such that 500 < W < 1500. A filter
that satisfy these conditions is
f 1 f
H(f ) = Ts Π = Π
2W 2000 2W
and the more general reconstruction filters have the form
1
2000 |f | < 500
H(f ) = arbitrary 500 < |f | < 1500
0 |f | > 1500
35
39. The lowpass signal x(t) with a bandwidth of W is sampled at intervals of Ts seconds, and the
signal
∞
xp (t) = x(nTs )p(t − nTs )
n=−∞
is generated, where p(t) is an arbitrary pulse (not necessarily limited to the interval [0, Ts ]).
Solution:
(a)
∞
xp (t) = x(nTs )p(t − nTs )
n=−∞
∞
= p(t) x(nTs )δ(t − nTs )
n=−∞
∞
= p(t) x(t) δ(t − nTs )
n=−∞
Thus
∞
Xp (f ) = P (f ) · F x(t) δ(t − nTs )
n=−∞
∞
= P (f )X(f ) F δ(t − nTs )
n=−∞
∞
1 n
= P (f )X(f ) δ(f − )
Ts n=−∞
Ts
∞
1 n
= P (f ) X(f − )
Ts n=−∞
Ts
(b) In order to avoid aliasing T1s > 2W . Furthermore the spectrum P (f ) should be invertible
for |f | < W .
f
(c) X(f ) can be recovered using the reconstruction filter Π( 2W Π
) with W < WΠ < T1s − W .
In this case
f
X(f ) = Xp (f )Ts P −1 (f )Π( )
2WΠ
40. Consider a signal s(t) whose Fourier transform is given below:
S(f)
f
-1 1
36
Sketch carefully the Fourier transform Sδ (f ) of the sampled signal
∞
sδ (t) = s(t) δ(t − nT )
n=−∞
∞
= s(nT )δ(t − nT )
n=−∞
for (a) T = 2/3 and (b) T = 1/2. For each case, only if possible, specifiy a filter characteristic
that allows a complete reconstruction of s(t) from sδ (t).
Solution:
(a) T = 2/3: Sδ(f)
3/2
… …
f
-3 -2 -1 1 2 3
2
… …
f
-3 -2 -1 1 2 3
H(f)
Reconstruction filter:
(only for T=1/2) 1/2
f
-1 1
Solution:
37
(a) Ts = 1/2, i.e., the inverse of the Nyquist rate which in this case is 2 Hz.
(b) At fs = 3 Hz, the sampled signal spectrum consists of nonoverlapping copies of the
signal:
… …
f
-4 -3 -2 -1 0 +1 +2 +3 +4
(c) In this case, fs = 3/2 Hz which is less than the Nyquist rate. The copies of the signal
spectrum overlap and the samples can be those of a signal with lower frequency.
… …
f
-1.5 -1 0 +1 +1.5
42. The signal x(t) = cos(2πt) is ideally sampled with a train of impulses. Sketch the spectrum
Xδ (f ) of the sampled signal, and find the reconstructed signal x̂(t), for the following values
of sampling period Ts and ideal lowpass reconstruction filter bandwidth W :
(a) Ts = 1/4, W = 2.
(b) Ts = 1, W = 5/2.
(c) Ts = 2/3, W = 2.
Solution: The spectra of the signal X(f ), and that of the sampled signal Xδ (f ), for each
case are shown in the figure below:
38
X(f)
1/2 1/2
f
-1 1
Xδ(f)
(a)
2 2 2 2 2 2 2 2 2 2
… …
f
-9 -7 -5 -3 -1 1 3 5 7 9
(b) Xδ(f)
1 1 1 1 1 1 1 1 1 1
… …
f
-5 -3 -1 1 3 5
(c) Xδ(f)
… …
f
-1 1
The reconstructed signals for each value of sampling period Ts and ideal lowpass reconstruc-
tion filter bandwidth W are:
43. The signal x(t) = sinc2 (t) is ideally sampled with a train of impulses. Sketch the spectrum
Xδ (f ) of the sampled signal, for the sampling periods below. For those values of Ts for
which reconstruction is possible, specify the range of the cutoff frequency W of the ideal
reconstruction filter.
(a) Ts = 2/3.
(b) Ts = 1.
(c) Ts = 1/4.
Solution: The spectra of the signal X(f ), and that of the sampled signal Xδ (f ), for each
case are shown in the figure below:
39
X(f)
f
-1 1
Xδ(f)
(a)
1.5
… …
f
-3 -2 -1 1 2 3
(b) Xδ(f)
1
… …
f
-3 -2 -1 1 2 3
(c) Xδ(f)
… 4
…
f
-4 -3 -1 1 3 4
The value of Ts for which reconstruction is possible, is (c) Ts = 1/4. The range of the cutoff
frequency W of the ideal reconstruction filter is 1 < W < 3.
3/2
1/2
f
-2 -1 0 1 2
This signal is sampled at fs samples/second with impulses and reconstructed using an ideal
lowpass filter (LPF) of bandwidth W = 2 and amplitude 1/fs . Let x̃(t) denote the output of
40
the LPF.
Solution:
4.5
3.5
2.5
x(t)
1.5
0.5
−0.5
−4 −3 −2 −1 0 1 2 3 4
t (sec)
Xδ(f)
3/2
… …
1/2
f
-4 -3 -2 -1 0 1 2 3 4
41
4.5
3.5
2.5
x~(t)
1.5
0.5
−0.5
−4 −3 −2 −1 0 1 2 3 4
t (sec)
(c) As shown by the sampled spectrum below, there is no overlap between the shifted copies
of X(f ). Therefore, x̃(t) = x(t).
Xδ(f)
3/2
… …
1/2
f
-6 -5 -4 -3 -2
2 -1 0 1 2 3 4 5 6
45. A compact disc (CD) records audio signals digitally using PCM. Assume the audio signal
bandwidth to be 15 KHz.
42
Solution:
Bandpass signals
46. Consider a signal s(t) whose Fourier transform is given below:
S(f)
1/2
f
-4 -3 -2 -1 1 2 3 4
for (a) T = 1/4 and (b) T = 1/2. For each case, only if possible, specifiy a filter characteristic
that allows a complete reconstruction of s(t) from sδ (t).
Solution:
43
(a) T=1/4:
Sδ(f)
Original (scaled)
2
… -1 -2 -1 +1 +2 +1 …
f
-4 -3 -2 -1 1 2 3 4
(b) T=1/2:
Sδ(f)
Reconstrucion filter:
H(f)
1/4
f
-4 -3 -2 -1 1 2 3 4
47. Determine the range of permissible cutoff frequencies for an ideal low pass filter used to
reconstruct the signal
x(t) = 10 cos(600πt) cos 2 (1600πt),
which is sampled at 4000 samples per second. Sketch X(f ) and Xδ (f ). Find the minimum
allowable sampling frequency.
Solution:
The cutoff frequency of the reconstruction filter can be in the range between W = 1900 Hz
and fs − W = 2100 Hz.
44
X(f)
f
-1600 0 1600 1900
Xδ(f)
… …
f
-5600 -4000 -1600 0 1600 4000 5600 5900
-5900
-2100 -1900
1900 2100
48. Given the bandpass signal spectrum shown in the figure below, sketch the spectra for the
following sampling rates fs and indicate which ones are suitable for the reconstruction of the
signal fom its samples: (a) 2B (b) 2.5B (c) 3B (d) 4B (e) 5B (f) 6B.
X(f)
f
-3 -2 -1 1 2 3
Solution:
For bandpass sampling and recovery, all but (b) and (e) will work theoretically, although an
ideal filter with bandwidth exactly equal to the unsampled signal bandwidth is necessary. For
lowpass sampling and recovery, only (f) will work.
In the figure next page, the case (a) fs = 2B, with B = 1 for convenience, is illustrated. The
terms used are n = 0, ±1, ±2, ±3 in the expression of the sampled spectrum:
∞
∞
Xδ (f ) = fs X(f − n fs ) = 2 X(f − 2n).
n=−∞ n=−∞
Spectra for higher values of n do not overlap with the spectrum of the original signal and are
therefore not shown.
45
Case fs=2B: (B=1) X(f)
A n=0
f
-4 -3 -2 -1 0 1 2 3 4
X(f-2), X(f+2):
-2 +2 2A -2 +2
… … n=1,-1
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9
X(f-4), X(f+4):
-4 -4 2A +4 +4
… … n=2,-2
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9
X(f-6), X(f+6):
-6 -6 2A +6 +6
n=3,-3
… … f
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9
2A
-9 -8 -7 -6 -5 -4 -3 -2 -1 0 1 2 3 4 5 6 7 8 9
46
49. (Downconversion by bandpass sampling) Consider the bandpass signal x(t) whose spectrum
is shown below.
X(f)
f
-8 -7 0 7 8
Solution:
t
(a) The signal is x(t) = 4 sinc2 2 cos(15πt), which is plotted below via the Matlab script:
t=-4:0.01:4; plot(t,4*sinc(t./2).^2.*cos(15*pi.*t))
−1
−2
−3
−4
−5
−4 −3 −2 −1 0 1 2 3 4
47
(b)
Xδ(f)
f
-9 -8 -7 -5 -4 -3 -1 0 1 3 4 5 7 8 9
(c) There is no overlap between copies of X(f ) and therefore x̃(t) = x(t).
(d) The spectrum at the output of the LPF is:
~
X(f)
-1 0 1
t
Therefore, the output signal is given by x(t) = 4 sinc2 2 cos(πt), which is plotted below
via the Matlab script:
t=-4:0.01:4; plot(t,4*sinc(t./2).^2.*cos(pi.*t))
−1
−2
−4 −3 −2 −1 0 1 2 3 4
48
50. Assume that the Fourier transform of a signal x(t) is real and has the shape shown in the
figure below:
X(f)
f
-W 0 W
Determine and plot the spectrum of each of the following signals, where x̂(t) denotes the
Hilbert transform of x(t),
3
(a) x1 (t) = 4 x(t) + 14 j x̂(t)
3
(b) x2 (t) = x(t) + 34 j x̂(t) ej2πf0 t , f0 W
43
(c) x3 (t) = x(t) + 14 j x̂(t) ej2πW t
43
(d) x3 (t) = 4 x(t) − 14 j x̂(t) ejπW t
Solution:
(c) The answer here is the same as in part (a), with the difference of a shift to the right in
the frequency domain by W ,
2 X(f − W ), f < W ;
1
X3 (f ) =
X(f ), f ≥ W.
(d) Here,
3 1 W W
X4 (f ) = − sgn(f − ) X(f − ) (4)
4 4 2 2
49
X1(f) X2(f)
2A
1.5A
f f
-W 0 W 0 f0 F0-W
X3(f) X4(f)
2A 2A
A A
f f
0 W 2W -W/2 W/2 3W/2
Solution:
(a) The spectrum of the analytical signal is
Xp (f ) = X(f ) + j[−j sgn(f )]X(f ) = [1 + sgn(f )] X(f ),
where X(f ) is the Fourier transform of x(t), given by
1 f + f0 f − f0
X(f ) = Π +Π .
2 2W 2W
Consequently,
f − f0
Xp (f ) = Π, f0 > 2W,
2W
a rectangular pulse of width 2W centered at f = f0 .
(b) The complex envelope xp (t) is
xp (t) = x̃(t)ej2πf0 t .
Therefore, x̃(t) = xp (t)e−j2πf0 t , and
∆ f
X̃(f ) = F{x̃(t)} = [Xp (f )]f →f +f0 = Π ,
2W
a rectangular pulse of width 2W centered at f = 0.
50
(c) The complex envelope is given by
Solution: For t < −τ /2, the output is zero. For |t| ≤ τ /2, the result is
α/2
y(t) = ! cos[2π(f0 + ∆f )t − θ] − e−α(t+τ /2) cos[2π(f0 + ∆f )t + θ] ,
α2 + (2π∆f )2
(α/2)e−αt
y(t) = ! eατ /2 cos[2π(f0 + ∆f )t − θ] − e−ατ /2 cos[2π(f0 + ∆f )t + θ] .
α2 + (2π∆f )2
53. The bandpass signal x(t) = sinc(t) cos(2πf0 t) is passed through a bandpass filter with impulse
response h(t) = sinc2 (t) sin(2πf0 t), Using the lowpass equivalents of both input and impulse
response, find the lowpass equivalent of the output and from it find the output y(t).
Solution:
1 1
x(t) = sinc(t) cos(2πf0 t) ⇐⇒ X(f ) = Π(f + f0 )) + Π(f − f0 ))
2 2
1 1
h(t) = sinc2 (t) sin(2πf0 t) ⇐⇒ H(f ) = − Λ(f + f0 )) + Λ(f − f0 ))
2j 2j
The lowpass equivalents are
Taking the inverse Fourier transform of Yl (f ) we can find the lowpass equivalent response of
51
the system. Thus,
yl (t) = F −1 [Yl (f )]
0 1
1 j2πf t 1 2
= (f + 1)e df + (−f + 1)ej2πf t df
2j − 1 2j 0
2
1 1 j2πf t 1 j2πf t 0 1 1 j2πf t 0
= fe + 2 2e 1 + 2j j2πt e 1
2j j2πt 4π t −2 −2
1 1
1 1 1 2 1 1 j2πf t 2
− f ej2πf t + 2 2 ej2πf t + e
2j j2πt 4π t 2j j2πt
0 0
1 1
= j − sin πt + 2 2 (cos πt − 1)
4πt 4π t
The output of the system y(t) can now be found from y(t) = Re[yl (t)ej2πf0 t ]. Thus
1 1
y(t) = Re (j[− sin πt + 2 2 (cos πt − 1)])(cos 2πf0 t + j sin 2πf0 t)
4πt 4π t
1 1
= [ 2 2 (1 − cos πt) + sin πt] sin 2πf0 t
4π t 4πt
Solution:
f −3/2
(a) The Fourier transform of the signal is X(f ) = 12 12 Π f +3/2
2 + 1
2 Π 2 . The
following sketch shows that the signal is wideband, as B = 2 and f0 = 3/2.
X(f)
1/4
f
-1.5 1.5
2 2
(b) From the quadrature modulator expression x(t) = xc (t) cos(2πf0 t) − xs (t) sin(2πf0 t),
it follows that xs (t) = 0 and therefore x (t) = xc (t) = sinc(2t). The corresponding
spectrum is sketched below:
52
Xl(f)
1/2
f
-1 1
(c) Use the expression x̂(t) = xc (t) sin(2πf0 t) + xs (t) cos(2πf0 t), from which it follows that
x̂(t) = sinc(2t) sin(3πt).
55. As shown in class, the in-phase and quadrature components, xc (t) and xs (t), respectively, of
the complex baseband (or lowpass) equivalent x (t) of a bandpass signal x(t) can be obtained
as
xc (t) cos(2πf0 t) sin(2πf0 t) x(t)
= ,
xs (t) − sin(2πf0 t) cos(2πf0 t) x̂(t)
where x̂(t) is the Hilbert transform of x(t).
(a) Sketch a block diagram of a system — using H to label the block that performs the
Hilbert transform — that has as input x(t) and as outputs xc (t) and xs (t).
(b) (Amplitude modulation) Let x(t) = a(t) cos(2πf0 t). Assume that the bandwidth W of
the signal a(t) is such that W f0 . Show that a(t) can be recovered with the following
system
H(f)
2
x(t) a(t)
f
-W W
Solution:
(a)
53
Σ xc(t)
x(t) cos(2πf0t)
H
Σ xs(t)
-1
(b) Use the modulation property of the Fourier transform. Let y(t) denote the mixer output.
The spectra are shown shown in the figure below.
X(f)
A0/2
f
-f0 f0
B=2W B=2W
Y(f)
2 H(f)
A0/2
A0/4
f
-W W
-2f0 2f0
B=2W B=2W
A(f)
A0
f
-W W
56. A lowpass signal x(t) has a Fourier transform shown in the figure (a) below.
54
X(f)
1
f
-W -W/2 W/2 W
(a)
sin(2πf0t)
x1(t) x3(t)
H
2cos(2πf0 t)
-
x2(t) x4(t)
H
(b)
The signal is applied to the system shown in figure (b). The blocks marked H represent
Hilbert transform blocks and it is assumed that W f0 . Determine the signals xi (t) and
plot Xi (f ), for 1 ≤ le7.
0 t) = −x(t) cos(2πf0 t) and x(t) cos(2πf
(Hint: Use the fact that x(t) sin(2πf 0 t) = x(t) sin(2πf0 t),
when the bandwidth W of x(t) is much smaller than f0 .)
x2 (t) = x̂(t)
X2 (f ) = −jsgn(f )X(f )
0 t) = −x(t) cos(2πf0 t)
x3 (t) = xˆ1 (t) = x(t) sin(2πf
1 1
X3 (f ) = − X(f + f0 ) − X(f − f0 )
2 2
55
x5 (t) = x̂(t) sin(2πf0 t) + x(t) cos(2πf0 t)
1 1
X5 (f ) = X4 (f ) − X3 (f ) = X(f + f0 )(sgn(f + f0 ) − 1) − X(f − f0 )(sgn(f − f0 ) + 1)
2 2
2jX1 (f ) −jX2 (f )
1) 2)
A A
AA AA
LL −f0 f0 AA
L A
2X3 (f ) 4) 2X4 (f )
3)
AA
A
AA −f0 A f0 AA −f0 f0
A AA A
X5 (f )
5)
−f0 f0
LL
L
6) X6 (f )
−2f0 2f0
AA A
A AA
7) X7 (f )
A
AA
56
Analog amplitude-modulation (AM) systems
57. The message signal m(t) = 2 cos(400t) + 4 sin(500t + π3 ) modulates the carrier signal c(t) =
A cos(8000πt), using DSB amplitude modulation. Find the time domain and frequency do-
main representation of the modulated signal and plot the spectrum (Fourier transform) of
the modulated signal. What is the power content of the modulated signal?
The figure figure below shows the magnitude and the phase of the spectrum U (f ).
|U (f )| A . . . . . . . . . . . . . . . . . . . . . .
6 6 6 6
. . . . . . . . . . . . . . . . . . . . . . .A/2
6 6 6 6
∠U (f )
π
s . . . . . . . . . . . . . . . . . . . . . . .6. . . . . . s
−π
s . . . . . . . . . .6. . . . . . . . . . . . . . . . . . . s
57
To find the power content of the modulated signal we write u2 (t) as
200 200
u2 (t) = A2 cos2 (2π(4 × 103 + )t) + A2 cos2 (2π(4 × 103 − )t)
π π
250 π 250 π
+4A2 sin2 (2π(4 × 103 + )t + ) + 4A2 sin2 (2π(4 × 103 − )t − )
π 3 π 3
+terms of cosine and sine functions in the first power
Hence,
T
2 A2 A2 4A2 4A2
P = lim u2 (t)dt = + + + = 5A2
T →∞ − T 2 2 2 2
2
58. In a DSB AM system, the carrier is c(t) = A cos(2πfc t) and the message signal is given by
m(t) = sinc(t) + sinc2 (t). Find the frequency domain representation and the bandwidth of
the modulated signal.
Solution:
u(t) = m(t)c(t) = A(sinc(t) + sinc2 (t)) cos(2πfc t)
59. A DSB-modulated signal u(t) = A m(t) cos(2πfc t) is mixed (multiplied) with a local carrier
xL (t) = cos(2πfc t+θ) and the output is passed through a lowpass filter (LPF) with bandwidth
equal to the bandwidth of the message signal m(t). Denote the power of the signal at the
output of the LPF by Pout and the power of the modulated signal by PU . Plot the ratio PPout
U
,
as a function of θ, for 0 ≤ θ ≤ π.
58
A plot of Pout
PU for 0 ≤ θ ≤ π is given in the next figure.
0.5
0.45
0.4
0.35
0.3
0.25
0.2
0.15
0.1
0.05
0
0 0.5 1 1.5 2 2.5 3 3.5
Theta (rad)
(a) Determine the message signal m(t) and the carrier c(t). (Hint: Look at the spectrum of
u(t).)
(b) Determine the modulation index.
(c) Determine the ratio of the power in the sidebands to the power in the carrier.
Solution:
(a)
Psidebands 50 1
= =
Pcarrier 200 4
61. An SSB AM signal is generated by modulating an 800 kHz carrier by the message signal
m(t) = cos(2000πt) + 2 sin(2000πt). Assume that the amplitude of the carrier is Ac = 100.
59
(b) Find the time-domain expression for the lower sideband SSB (LSSB) AM signal.
(c) Determine the spectrum of the LSSB AM signal.
Solution:
(a) The Hilbert transform of cos(2π1000t) is sin(2π1000t), whereas the Hilbert transform of
sin(2π1000t) is − cos(2π1000t). Thus
62. The system shown in the figure below can be used to generate an AM signal.
Nonlinear
x(t) y(t)
m(t) memoryless Filter u(t)
system AM signal
c(t)
The carrier is c(t) = cos(2πf0 t) and the modulating signal m(t) has zero mean and its
maximum absolute value is Am = max |m(t)|. The nonlinear device has a quadratic input-
output characteristic given by
y(t) = a x(t) + b x2 (t).
60
(b) Specify the filter characteristics such that an AM signal is obtain at its output.
(c) What is the modulation index?
Solution:
(a)
(b) The filter should reject the low frequency components, the terms of double frequency
and pass only the signal with spectrum centered at f0 . Thus the filter should be a BPF
with center frequency f0 and bandwidth W such that f0 − WM > f0 − W 2 > 2WM where
WM is the bandwidth of the message signal m(t).
(c) The AM output signal can be written as
2b
u(t) = a(1 + m(t)) cos(2πf0 t)
a
Since Am = max[|m(t)|] we conclude that the modulation index is
2bAm
α=
a
(a) (Suppressed carrier AM) Plot the power Pu of the modulated signal as a function of the
∆
phase difference (between transmitter and receiver) ∆φ = φc − φr .
(b) (Conventional AM) With a modulation index a = 0.5 (or 50%), sketch carefully the
modulated signal u(t).
Solution:
(a) The demodulated signal power is given by Py = Pu cos2 (∆φ), which has maximum value
Py = Pu for ∆φ = 0, π and minimum value Py = 0 for ∆φ = π/2, 3π/2. This is shown
below, which is a plot of Py /Pu as a function of ∆φ.
61
Demodulated signal power
0.8
0.6
Py/Pu
0.4
0.2
0 1 2 3 4 5 6
Phase error in radians
and plotted in the figure below, together with the carrier cos(10πt) (top graph). Both
plots are normalized with respect to the carrier amplitude Ac .
Carrier signal
1
0.5
−0.5
−1
−0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4
Modulated signal
1.5
0.5
−0.5
−1
−1.5
−0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4
62
Probability and random signals
64. A (random) binary source produces S = 0 and S = 1 with probabilities 0.3 and 0.7, respec-
tively. The output of the source S is transmitted over a noisy (binary symmetric) channel
with a probability of error (converting a 0 into a 1, or a 1 into a 0) of = 0.2.
1−ε
S=0 R=0
ε
ε
S=1 R=1
1−ε
Solution:
(a)
P (R = 1) = P (R = 1|S = 1)P (S = 1) + P (R = 1|S = 0)P (R = 0)
= 0.8 · 0.7 + 0.2 · 0.3 = 0.62
Solution:
(a)
x < −1 ⇒ FX (x) = 0
x
x
1 2 1 1
−1 ≤ x ≤ 0 ⇒ FX (x) = (v + 1)dv = ( v + v) = x2 + x +
−1 2 −1 2 2
0 x
1 1
0 ≤ x ≤ 1 ⇒ FX (x) = (v + 1)dv + (−v + 1)dv = − x2 + x +
−1 0 2 2
1 ≤ x ⇒ FX (x) = 1
63
(b)
1 1 7 1
p(X > ) = 1 − FX ( ) = 1 − =
2 2 8 8
(c)
1 p(X > 0, X < 12 ) FX ( 12 ) − FX (0) 3
p(X > 0X < ) = = =
2 1
p(X < 2 ) 1 − p(X > 2 ) 1 7
(d) We find first the CDF
1 1 p(X ≤ x, X > 12 )
FX (xX > ) = p(X ≤ xX > ) =
2 2 p(X > 12 )
If x ≤ 12 then p(X ≤ xX > 12 ) = 0since the events E1 = {X ≤ 12 } and E1 = {X > 12 }
are disjoint. If x > 12 then p(X ≤ xX > 12 ) = FX (x) − FX ( 12 ) so that
1 FX (x) − FX ( 12 )
FX (xX > ) =
2 1 − FX ( 12 )
66. A random process is given by X(t) = A + Bt, where A and B are independent random vari-
ables uniformly distributed in the interval [−1, 1]. Find:
Solution:
mX (t) = E[A + Bt] = E[A] + E[B]t = 0
where the last equality follows from the fact that A, B are uniformly distributed over [−1 1]
so that E[A] = E[B] = 0.
Thus
1 1
RX (t1 , t2 ) = + t1 t2
3 3
and the provess is not stationary.
64
67. A simple binary communication system model, with additive Gaussian noise, is the following:
S is a binary random variable, representing the message bit sent, taking values −1 and +1
with equal probability. Additive noise is represented by a Gaussian random variable N of
zero mean and variance σ 2 . The received value is a random variable R = S + N . This is
illustrated in the figure below:
S R
Solution: Note that S is a discrete random variable. Therefore, it can be specified by its
probability mass function (PMF), P [S = +1] = P [S = −1] = 1/2, or by a PDF that consists
of two impulses, each of weight 1/2, centered at ±1.
Given a value of S = s, the conditional PDF of R is
1 1
pR|S (r) = pN (r − s) = √ exp − 2 (r − s) .
2
2πσ 2 2σ
68. The joint PDF of two random variables X and Y can be expressed as
c 1
x 2
y 2
pX,Y (x, y) = exp − +
π 2 3 2
Solution:
(a) The given joint PDF can be factored as the product of two functions f (x) and g(y). As
a consequence, X and Y are independent.
65
(b) pX,Y (x, y) has the form of the joint PDF of two zero-mean independent Gaussian random
variables (i.e., ρ = 0) with variances σX 2 = 9 and σ 2 = 4:
Y
1 1 x 2
y 2
pX,Y (x, y) = exp − + ,
2π · 3 · 2 2 3 2
Let the input be a Gaussian random variable X, of zero mean and unit variance, N (0; 1).
You are asked to compute the probability mass function (PMF) of the output Y = g(X), i.e.,
P [Y = y] for y ∈ {±1, ±3}. Express your result in terms of the Gaussian Q-function.
∆ ∆
2 = E{(X − m )2 }. Find the mean
70. Let X be an r.v. of mean mX = E{X} and variance σX X
mY and variance σY2 , in terms of mX and σX
2 , of the r.v. Y = X + a, where a is a constant.
71. Let G be a Gaussian r.v. of mean m = 0 and variance σ 2 = 1. Find the mean mN and
2 of the r.v. N = aG, where a is a constant. This transformation is used in
variance σN
computer simulations to generate samples of an AWGN process. If it is desired to generate
samples with mN = 0 and variance σN2 = N /2, what is the value of a?
0
66
72. Let X be a uniform r.v. over the unit interval [0, 1]. Show that P [X ≤ a] = a, where
0 < a ≤ 1. This fact is used in computer simulations to generate random bits.
Solution:
120 120
100 100
80 80
60 60
40 40
20 20
20 40
0
30
−20
20
−40
−60 10
−80
50 100 150 200 250 20 40 60 80 100 120
Matched filter and correlator outputs for a rectangular pulse and a sine pulse.
67
To see why the outputs have the same value at t = T , write the convolution integral of the
matched filter:
v(T ) = h(t) ∗ s(t)|t=T
T T
= h(τ )s(T − τ )dt = s(T − τ )s(T − τ )dt
0 T 0
= s(u)2 du,
0
with u = T − τ . Therefore, the output of the matched filter sampled at t = T equals the
output of a correlator over the interval [0, T ], with s(t) as the reference signal.
74. The received signal in a binary communication system that employs antipodal signals is
r(t) = s(t) + n(t),
where s(t) is shown in the figure below and n(t) is AWGN with power spectral density N0 /2
W/Hz.
s(t)
t
0 1 2 3
(a) Sketch carefully the impulse response of the filter matched to s(t)
(b) Sketch carefully the output of the matched filter when the input is s(t)
(c) Dtermine the variance of the noise at the output of the matched filter at t = 3
(d) Determine the probability of error as a function of A and N0
Solution:
(a) The impulse response of the filter matched to s(t) is
h(t) = s(T − t) = s(3 − t) = s(t)
T 3
where we have used the fact that s(t) is even with respect to the t = 2 = 2 axis.
(b) The output of the matched filter is
t
y(t) = s(t) s(t) = s(τ )s(t − τ )dτ
0
0 t<0
2t 0≤t<1
A
A (2 − t) 1 ≤ t < 2
2
2A2 (t − 2) 2 ≤ t < 3
=
2A2 (4 − t) 3 ≤ t < 4
A2 (t − 4) 4 ≤ t < 5
A2 (6 − t) 5 ≤ t < 6
0 6≤t
68
A scetch of y(t) is depicted in the next figure
2A2 . . . . . . . . . . . . . . . .
A
A
A2 . . . . . . A
@ A @
@ AA @
@ @
1 2 3 4 5 6
(c) At the output of the matched filter and for t = T = 3 the noise is
T
nT = n(τ )h(T − τ )dτ
0
T T
= n(τ )s(T − (T − τ ))dτ = n(τ )s(τ )dτ
0 0
we obtain "
4A2
P (e) = Q
N0
(a) Download and execute the Matlab script integrate and dump.m from the web site of the
class. Upon completion, a plot shows waveforms associated with the transmission of ten
random bits over an AWGN channel, with a rectangular pulse shape and an integrate-
and-dump receiver. Execute the script with your student ID, an amplitude a = 1 and
with N0 = 1. Sketch or print the plot.
69
(b) Download and execute the Matlab scripts intdmp simulation.m and Q.m from the web
site of the class. The first script simulates the transmission of random bits over and
ideal AWGN channel and computes the bit error rate (BER) as a function of the signal-
to-noise ratio (SNR), Es /N0 in dB. The script will plot the simulated BER versus
! SNR
as well as the theoretical expression for the bit error probability P [e] = Q( 2Es /N0 ).
Execute the script with your student ID and sketch or print the resulting plot.
Solution:
(a)
Original bit sequence Received pulse sequence
1 2
1
0.5 0
−1
0 −2
20 40 60 80 100 20 40 60 80 100
Transmitted pulse sequence Integrate and dump receiver output
1 1
0 0
−1 −1
20 40 60 80 100 20 40 60 80 100
Received pulse sequence Estimated bit sequence
2 1
1
0 0.5
−1
−2 0
20 40 60 80 100 20 40 60 80 100
samples samples
70
Error performance of binary transimission over an AWGN chan nel
−1
10
Simulated
Theory
−2
10
−3
10
Bit error rate
−4
10
−5
10
−6
10
0 1 2 3 4 5 6 7 8 9 10
Eb/N0 (dB)
(b)
76. In computer communications at 10 Mbps using the Ethernet standard (IEEE 802.3), Manchester
or bi-phase pulse formatting is used. Polar mapping is employed such that, in the interval
[0, T ], a “0” is sent as s0 (t) = a g(t), and a “1” is sent as s1 (t) = −a g(t), where the pulse
shape gT (t) is sketched in the figure below:
gT(t)
1/T
t
T/2 T
- 1/T
(a) Find the impulse response h(t) of the matched filter (MF) for gT (t).
(b) Assume that a “0” is sent and no noise is present. The input of the MF is r(t) = s0 (t).
Find the corresponding response y(t) = r(t) ∗ h(t), sampled at t = T . i.e., Y = y(T ).
What is the value of Y if a “1” is sent?
(c) A key advantage of the Manchester format is its capability to detect collisions. To see
this, consider two bit sequences from two users. User 1 transmits the sequence “001101”
71
and user 2 the sequence “011011”. Sketch the corresponding transmitted sequences
s1 (t) and s2 (t), as well as the received sequence s1 (t) + s2 (t). Comment on your results.
Specifically, how is a collision detected?
Solution:
(a)
gT(t)
h(t)
1/T
1/T
t
T/2 T t
T/2 T
- 1/T
- 1/T
(c)
s1(t)
a 1/T
t/T
1 2 3 4 5 6
-a 1/T
s2(t)
a 1/T
t/T
1 2 3 4 5 6
-a 1/T
[s2(t)+s2(t)]/2
a 1/T
t/T
1 2 3 4 5 6
-a 1/T
Collisions are detected in the received sequence s1 (t)+ s2 (t) by the absence of transitions
at times nT + T /2, for n = 1, 3, 4.
72
77. Download the Matlab script rf pulse mfoutput.m from the class website. This script com-
putes the output of a correlator for an RF pulse,
'
2
g(t) = cos(2πf0 t), t ∈ [0, T ],
T
where f0 = n0 /T , and n0 is a positive integer. You are required to run the script and record
the resulting plots for two values of n0 , n0 = 2 and n0 = 20. What conclusion can you draw
from these plots?
Solution:
0.5
−0.5
−1
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Correlator output
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
73
RF pulse: sqrt(2/T)*cos(2πf0t), f0=10
0.5
−0.5
−1
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
Correlator output
0.8
0.6
0.4
0.2
0
0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2
The MF output approaches a ramp as the value of the center frequency f0 increases.
78. Compare NRZ and RZ signaling techniques in terms of probability of a bit error P [e]. The
bit rate R = 1/T and pulse amplitude a are fixed.
(a) Plot or sketch the curve of energy-to-noise ratio E/N0 (dB) versus probability of P [e].
(Hint: The energy of an RZ pulse is one half that of an NRZ pulse.)
(b) How should the rate R be modified in the case of RZ signaling so that its P [e] is the
same as NRZ signaling?
Solution:
(a)
74
0
10
−1
10
−2
10
RZ
−3
10
−4
10
P[e]
−5
10 NRZ
−6
10
−7
10
−8
10
−9
10
0 2 4 6 8 10 12 14 16
E/N0 (dB)
(b) To achieve the same P [e] as NRZ, for the same peak amplitude constraints, the rate
of RZ should be 50% smaller than that of NRZ. To see this, consider as an example
recangular pulse signaling with polar mapping. Then EN RZ = A2 T and ERZ = A2 T /2.
Therefore, the period T in RZ should be doubled.
79. Compare polar and unipolar bit mapping techniques in terms of P [e]. The peak pulse energy
is fixed. Plot or sketch curve of energy-to-noise ratio E/N0 (dB) versus probability of a√bit
error for both techniques.
√ (Hint: The distance between the signal points d12 is d12 = 2 E
for polar and d12 = E for unipolar.)
Solution: If the peak pulse energy E is fixed. Then for polar mapping
'
2E
P [e] = Q ,
N0
75
0
10
−1
10
−2
10
Unipolar
−3
10
−4
10
P[e]
−5
10 Polar
−6
10
−7
10
−8
10
−9
10
0 2 4 6 8 10 12 14 16 18
E/N (dB)
0
80. The bit stream {bn } = 0, 1, 1, 0, 0, 0, 1, 0 is to be sent through a channel (lowpass LTI system
with large bandwidth). Assume that rectangular pulses of amplitude A are used and the bit
rate is 1/T bps. In polar mapping, use the rule:
bn an
0 −A
1 +A
Sketch the transmitted signal for each of the following line coding schemes:
Solution: (Line coding. Bit stream {bn } = 0, 1, 1, 0, 0, 0, 1, 0.) The transmitted signals are
shown in the figure below (amplitude A = 2), and can be reproduced using the Matlab script
homework6s05.m available in the class web site.
76
1
Bits
0
0 1 2 3 4 5 6 7 8
2
U−NRZ
0
0 1 2 3 4 5 6 7 8
2
U−RZ
0
0 1 2 3 4 5 6 7 8
2
P−NRZ
−2
0 1 2 3 4 5 6 7 8
2
P−RZ
−2
0 1 2 3 4 5 6 7 8
2
AMI−NRZ
−2
0 1 2 3 4 5 6 7 8
2
AMI−RZ
−2
0 1 2 3 4 5 6 7 8
2
Manchester
−2
0 1 2 3 4 5 6 7 8
77
81. Compare the seven schemes in problem 1, in terms of average (DC) power and average (DC)
amplitude level.
Solution:
A2 T A
U-NRZ
2 2
A2 T A
U-RZ
4 4
P-NRZ A2 T 0
A2 T
P-RZ 0
2
A2 T
AMI-NRZ 0
2
A2 T
AMI-RZ 0
4
Manhester A2 T 0
82. Manchester coding has the desirable feature that it is possible to detect the presence of errors
in the received signal. Explain how this is achieved. Sketch a block diagram of an error
detection circuit.
Solution:
Manchester coding has the desirable feature that it is possible to detect the presence of errors
in the received signal. This is done by checking that there is always a transition in the middle
of a bit period. A simple block diagram on to achieve this is shown below:
XOR
input flag
+
The input is assumed to be sampled at a proper timing phase t = kT /2 + τ , so that the XOR
gate outputs a “1” at least every other sample (spaced by T /2). The second sampler checks
that there is a transition every bit. If this is the case, then the flag is always equal to “1”;
otherwise, the flag is “0” and an error is detected.
78
A Simulink model based on this idea can be found in the web site of the class under the
name ETHERNET ERROR CHECK.mdl. After a transition period, the system always outputs a
“1” whenever the input is Manchester coded, and a “0” whenever there is no transition in (at
least) the following bit.
(a) Determine the SNR at the output of the integrator sampled at t = T . (Hint: Show that
the signal power at the output of the integrator, sampled at t = T , is Es = Eb T . The
noise power at the output of the integrator, sampled at t = T , is Pn = N20 T . The output
SNR is then the ratio Es /Pn .)
(b) Without any computation, what is the answer to part (a)? (See class notes. √ The
integrator is the correlator of a rectangular pulse of width T and amplitude 1/ T .)
Solution:
(a) The output of the integrator is
t t
y(t) = r(τ )dτ = [si (τ ) + n(τ )]dτ
0 0
t t
= si (τ )dτ + n(τ )dτ
0 0
At time t = T we have
'
T T T
Eb
y(T ) = si (τ )dτ + n(τ )dτ = ± T+ n(τ )dτ
0 0 T 0
The signal energy at the output of the integrator at t = T is
' 2
Eb
Es = ± T = Eb T.
T
The noise power can be computed as follows:
T T
Pn = E n(τ )n(v)dτ dv
0 0
T T
= E[n(τ )n(v)]dτ dv
0 0
T T
N0
= δ(τ − v)dτ dv (5)
2 0 0
T
N0 T
= δ(τ − v)dτ dv (6)
2 0 0
N0 T N0
= 1 dv = T
2 0 2
79
Hence, the output SNR is
Es 2Eb
SNR = =
Pn N0
(b) Without any computation, what is the answer to part (a)?
s0 = 0, 0 ≤ t ≤ T ;
s1 = A, 0 ≤ t ≤ T,
for transmission of the information. This is called on-off signaling. The demodulator cross-
correlates the received signal r(t) with s1 (t) and sampled the output of the correlator at
t = T.
(a) Determine the optimum detector for an AWGN channel and the optimum threshol,
assuming the the signals are equally probable
(b) Find the probability of error as a function of the SNR. How does on-off signaling compare
with antipodal signaling?
Solution:
Assuming that s(t) has unit energy, then the sampled outputs of the correlators are
r = sm + n, m = 0, 1
√
where s0 = 0, s1 = A T and the noise term n is a zero-mean Gaussian random variable
with variance
T T
1 1
σn = E √
2
n(t)dt √ n(τ )dτ
T 0 T 0
1 T T
= E [n(t)n(τ )] dtdτ
T 0 0
N0 T T N0
= δ(t − τ )dtdτ =
2T 0 0 2
80
The probability density function for the sampled output is
2
1 −r
f (r|s0 ) = √ e N0
πN0
√ 2
1 − (r−A T)
f (r|s1 ) = √ e N0
πN0
Since the signals are equally probable, the optimal detector decides in favor of s0 if
PM(r, s0 ) = f (r|s0 ) > f (r|s1 ) = PM(r, s1 )
otherwise it decides in favor of s1 . The decision rule may be expressed as
√ √ √ s0
PM(r, s0 ) (r−A T )2 −r 2 (2r−A T )A T
N0
− N0 >
=e =e < 1
PM(r, s1 )
s1
or equivalently
s1
1 √
r >
< 2A T
s0
√
The optimum threshold is 12 A T .
(b) The average probability of error is
1 1
P (e) = P (e|s0 ) + P (e|s1 )
2 2
∞ 1 √
1 1 2A T
= f (r|s0 )dr + f (r|s1 )dr
2 1 A√T 2 −∞
2
1 √ √ 2
1 ∞ 1 −Nr2 1 2A T 1 − (r−A T)
= √ e 0 dr + √ e N 0 dr
2 1 A√T πN0 2 −∞ πN0
2
∞ 2
− 1 2 A√ T
1 1 x 1 2 N0 1 x2
= √ √ e− 2 dx + √ e− 2 dx
2 1 2 A T 2π 2 −∞ 2π
2 N0
' √
1 2 √
= Q A T = Q SNR
2 N0
where
1 2
2A T
SNR =
N0
Thus, the on-off signaling requires a factor of two more energy to achieve the same
probability of error as the antipodal signaling.
Multi-dimensional signals
81
86. In class, the first three steps of the Gramm-Schmidt procedure for problem 7.6 in the textbook
were completed.
∆
(a) Perform step 4 of the procedure and find the 3-dimensional basis, Φ = {φ1 (t), φ2 (t), φ3 (t)},
that contains the 4 signals sm (t), m = 1, 2, 3, 4. Represent the signals as points s̄m ,
m = 1, 2, 3, 4, in this three-dimensional signal space.
(b) Verify that the distance to the origin (or norm) of the points s̄m in the signal space
with basis Φ of part (a) is the same as that of the signal points obtained with the basis
∆
Φ = {φ1 (t), φ2 (t), φ3 (t)}, where φn (t), n = 1, 2, 3, are illustrated in Fig. 1 below.
1 1 1
t t t
1 2 3 1 2 3 1 2 3
Solution:
Compute the projections, c41 and c42 , of s4 (t) onto the signals φ1 (t) and φ2 (t), respec-
tively,
3
4
c41 = s4 (t)φ1 (t)dt = √ ,
0 3
3 '
2
c42 = s4 (t)φ2 (t)dt = .
0 3
82
The coordinates of the vector representation s̄m = (sm1 sm2 sm3 ) of the signal sm (t),
with respect to the basis Φ = {φ1 (t), φ2 (t), φ4 (t)}, 1 ≤ m ≤ 4, are computed as
3
smn = sm (t)φn (t), 1 ≤ m ≤ 4, 1 ≤ n ≤ 3.
0
This results in
√
s̄1 = 2 3 0 0
(
s̄2 = √2 2
3
2 3 0
(
s̄3 = − √43 2 23 0
(
s̄4 = √4 2 √2
33 2
The energies (squared distances to the origin) of the signals are computed from their
vector representation as
3
Em = |smn |2 , 1 ≤ m ≤ 4.
n=1
It follows that
4+8 16 + 8 16 + 2
E1 = 12, E2 = = 4, E3 = = 8, E4 = + 2 = 8.
3 3 3
(b) In this case, by inspection, the vector representation of the signals is easy to obtain:
s̄1 = 2 2 2
s̄2 = 2 0 0
s̄3 = 0 −2 −2
s̄4 = 2 2 0
Note that s3 (t) = s2 (t) − s1 (t) and therefore the dimensionality of the signal space
spanned by Φ is 3. It is easy to check that the signal energies obtained from this vector
representation are the same as those obtained in part (a) above.
( (
87. The signals φ1 (t) = T2 cos(2πfc t) and φ2 (t) = T2 sin(2πfc t), for (k − 1)T < t ≤ kT , where
k is an integer and T is the symbol duration, are used as basis functions for bandpass digital
modulation.
Show that φ1 (t) and φ2 (t) are orthogonal over the interval [(k − 1)T, kT ].
83
88. The figure below depicts two pulse shapes used in a binary orthogonal signaling scheme
know as binary PPM (or 2-PPM). Transmission takes place over an AWGN channel with
SN (f ) = N20 W/Hz.
s1(t) s2(t)
a a
t t
T/2 T T/2 T
(a) Find the value of the amplitude a in terms of the energy per bit Eb and the bit duration
T.
(b) Determine the orthonormal basis signals ψ1 (t) and ψ2 (t) for 2-PPM and sketch the
location of the signal points s̄1 and s̄2 , corresponding to signals s1 (t) and s2 (t), in the
two-dimensional signal space.
(c) Give an expression
! for the probability of a bit error. (Hint: Use the general expression
P [error] = Q 2
d12 /2N0 .)
Solution:
(
(a) a = 2E b
T .
(b)
ψ1(t) ψ2(t)
2 2
T T
t t
T/2 T T/2 T
ψ2(t)
ψ1(t)
E
84
(
Eb
(c) P [error] = Q N0 .
89. The two pulses shown in the figure below are employed in a binary orthogonal signaling
scheme to be used in a digital communication system.
s1(t) s2(t)
a a
t t
T/2 T
-a -a
(a) Sketch carefully the impulse responses of a pair of matched filters for this system.
(b) Find the means and variances of the outputs of the matched filters, sampled at t = T ,
when s1 (t) is sent through an AWGN channel with noise energy N0 /2.
(c) A correlator receiver is employed. However, due to a timing error, the output is incor-
rectly sampled at t = 0.75T . What is the increase in the probability of a bit error, Pb ,
compared to that of an ideal system?
Solution:
h2(t) h1(t)
a a
t t
T/2 T
-a -a
Alternatively,
! defining the orthonormal functions ψ1 (t) and ψ2 (t), we can write s1 (t) =
a T /2 ψ1 (t). In this case, the outputs of the filters matched to the orthonormal!func-
tions, given that s1 (t) is sent, are such that Y1 is a Gaussian r.v. of mean m1 = a T /2
and variance σ12 = N20 , and Y2 is a Gaussian r.v. of mean m2 = 0 and variance σ22 = N20 .
85
(c) Correlator outputs:
y1(t) y2(t)
a2T/2 a2T/2
t t
T/2 T T/2 T
(a) (b)
(a) First output when s1 (t) sent. (b) Second output when s2 (t) sent.
When sampled at t = 0.75T , the
mean and variance of Y1 , given that s1 (t) is sent,
2 2
remain at m1 = a 2T and σ12 = N20 a 2T . However, when s2 (t) is sent, the mean and
2 2
2
2
variance of Y2 change to m2 = (0.5) a 2T = a 4T and σ12 = (0.5) N20 a 2T = N20 a 4T .
From this observation, it follows that
' "
E a2T
P [error|s̄1 sent] = Q = Q , and
N0 2N0
' "
E 2
a T
P [error|s̄2 sent] = Q = Q .
2 N0 4N0
The increase in the probability of a bit error, due to the timing error, is
" "
1 2
a T 2
a T
∆Pb = Q − Q .
2 4N0 2N0
90. A binary PAM communication system is used to transmit data over an AWGN channel. The
prior probabilities for the bits are Pr{am = 1} = 1/3 and Pr{am = 1} = 2/3.
86
(a) Determine the optimum maximum-likelihood decision rule for the detector
(b) Find the average probability of a bit error
Solution:
(a) The optimum threshold is given by
N0 1−p N0
η = √ ln = √ ln 2
4 Eb p 4 Eb
N0
√
(b) The average probability of error is (η = 4 Eb
ln 2)
∞ √ 2
1
P (e) = p(am = −1) e−(r+ Eb ) /N0 dr
√
η πN0
η √ 2
1
+p(am = 1) √ e−(r− Eb ) /N0 dr
−∞ πN0
√
√
2 η + Eb 1 Eb − η
= Q ! + Q !
3 N0 /2 3 N0 /2
! '
' !
2 2N0 /Eb ln 2 2Eb 1 2Eb 2N0 /Eb ln 2
= Q + + Q −
3 4 N0 3 N0 4
91. Determine the average energy of a set of M PAM signals of the form
sm (t) = sm φ(t), m = 1, 2, · · · , M, 0 ≤ t ≤ T,
where !
sm = Eg Am , m = 1, 2, · · · , M.
The signals are equally probable with amplitudes that are symmetric about zero and are
uniformly spaced with distance d between adjacent ampltiudes.
1 2
M M
d2
Eave = sm = Eg (2m − 1 − M )2
M m=1 4M m=1
d2 M
= Eg [4m2 + (M + 1)2 − 4m(M + 1)]
4M
m=1
M
d2
M
= Eg 4 m + M (M + 1) − 4(M + 1)
2 2
m
4M
m=1 m=1
d2 M (M + 1)(2M + 1) M (M + 1)
= Eg 4 + M (M + 1) − 4(M + 1)
2
4M 6 2
M −1d
2 2
= Eg .
3 4
87
Moreover, note that if a = d/2 and Eg = 1, as we did in class, then
M2 − 1 2
Eave = a ,
3
and therefore '
3 Eave
a= .
M2 − 1
92. A speech signal is sampled at a rate of 8 KHz, logarithmically compressed and encoded into
a PCM format using 8 bits per sample. The PCM data is transmitted through an AWGN
baseband channelvia M -level PAM signaling. Determine the required transmission bandwidth
when (a) M = 4, (b) M = 8 and (c) M = 16. (Assume rectangular pulses and the zero-to-null
definition of bandwidth.)
94. The 16-QAM signal constellation shown in the figure below is an international standard for
(analog) telephone-line modems known as V.29. Determine the decision boundaries for the
detector, assuming equiprobable signals and high SNR so that errors only occur between
adjacent points.
88
5
-5 -3 -1 1 3 5
-1
-3
-5
95. A digital communication system transmits data using QAM signaling over a voice-band tele-
phone channel at a rate 2400 symbols/s (baud). The additive noise is assumed to be white
and Gaussian. You are asked to determine the energy-per-bit-to-noise ratio Eb /N0 required
to achieve an error probability of 10−5 for a bit rate equal to:
(a) 4800 bits/s
(b) 9600 bits/s
(c) 19200 bits/s
89
(d) 31200 bits/s
(e) What conclusions can you draw from parts (a) to (d)?
(Hint: Use the following expression for the error probability of an M -ary QAM system:
"
3Eb
PM ≈ 4 Q ,
(M − 1)N0
where M = 2 and is the number of bits per symbol. Also, if needed, use the following
approximation of the Gaussian Q-function:
1 − x2
Q(x) ≈ e 2 .)
2
Solution:
We assume that ideal Nyquist signaling is employed so that the bandwidth of the bandpass
signal is equal to the baud rate (symbols/second), i.e., 1/T Hz.
90
(e) The following table gives the SNR per bit and the corresponding number of bits per
symbol for the constellations used in parts (a)-(d):
2 4 8 13
SNR per bit (dB) 10.24 14.22 23.52 36.48
As observed in the next plot, there is an increase in average transmitted power of ap-
proximately 2.5 dB per additional bit per symbol:
−5
SNR per bit required to achieve P = 10
e
40
35
30
SNR per bit
25
20
15
10
2 4 6 8 10 12 14
Number of bits per symbol
96. Plots of the approximated probability of a bit error Pb are compared with the simulated BER,
for M -PSK and M -QAM, in Figs. 1 and 2, respectively, with Matlab scripts digmodMPSK.m
and digmodMQAM.m. The approximated Pb values were produced by the Matlab scripts
digmodMPSK approx.m and digmodMQAM approx.m posted in the web site.
91
Error performance of M−PSK. SJSU. EE160 − Spring 2004
0
10
BPSK
QPSK
8PSK
−1
10
−2
10
−3
10
BER
−4
10
−5
10
−6
10
−7
10
0 2 4 6 8 10 12 14 16 18 20
E/N0 (dB)
Simulation (blue solid line) versus approximated Pb (red dotted line) for M -PSK.
92
Error performance of M−QAM. SJSU − Spring 2004
0
10
QPSK
16−QAM
64−QAM
−1
10
−2
10
BER
−3
10
−4
10
0 5 10 15 20 25
E/N0 (dB)
Simulation (blue solid line) versus approximated Pb (red dotted line) for M -QAM.
93
97. HOMEWORK 9, Spring 2005. Problems 2, 4c and 5
98. Binary PSK (BPSK) is used for data transmission over an AWGN channel with power spectral
density N0 /2 = 10−10 W/Hz. The transmitted signal energy is Eb = A2 T /2, where T is the
bit duration and A is the signal amplitude. Determine the value of A needed to achieve an
error probability of 10−6 , if the data rate is:
(a) 10 Kbit/s
(b) 100 Kbit/s
(c) 1 Mbit/s
If the data rate is 10 Kbps, then the bit interval is T = 10−4 and therefore, the signal
amplitude is !
A = 44.9352 × 10−10 × 104 = 6.7034 × 10−3
Similarly we find that when the rate is 105 bps and 106 bps, the required amplitude of the
signal is A = 2.12 × 10−2 and A = 6.703 × 10−2 respectively.
Solution:
94
(a) Using the trigonometric identity cos(A + B) = cos(A) cos(B) − sin(A) sin(B), we have
that
'
2E
si (t) = cos(2πfc t + θi )
T
'
2E
= [cos(θi ) cos(2πfc t) − sin(θi ) sin(2πfc t)]
T
√ √
= E cos(θi ) φ1 (t) − E sin(θi ) φ2 (t)
E s4
s3 s1
φ1
- E E
- E s2
Z4
φ2
E s4
s3 s1
Z3 φ1 Z1
- E E
- E s2
Z2
100. Using a computer environment (e.g., Matlab), compare the exact probability of a symbol
error, P [], with the union bound. What is the minimum value of E/N0 (dB) after which the
union bound is tight?
95
Solution: With a Matlab script, we may compare the exact expression
'
' 2
E E
P [] = 2 Q − Q ,
N0 N0
with the union bound ' '
E 2E
P [] = 2 Q +Q .
N0 N0
The plots are shown in the graph below:
0
10
Union
bound
Exact
−1
10
P[ε]
−2
10
−3
10
−2 0 2 4 6 8 10
E/N0 (dB)
Note that the bound is tight after a minimum value of approximately (E/N0 )min = 9 dB.
The Matlab script used to compute the points in the graph is the following:
esno=[-3 -2 -1 0 1 2 3 4 5 6 7 8 9 10];
esnor=10.^(esno/10);
pexct=2*erfc(sqrt(esnor)/sqrt(2))/2 - (erfc(sqrt(esnor)/sqrt(2))/2).^2;
pebnd=2*erfc(sqrt(esnor)/sqrt(2))/2 + erfc(sqrt(2*esnor)/sqrt(2))/2;
semilogy(esno,pexct);
hold on;
semilogy(esno,pebnd);
96
101. HOMEWORK 7, Spring 2004. Problems 2 and 3
N0
102. Consider a QPSK communication system over an AWGN channel with SN (f ) = 2 W/Hz.
Bit values 0 and 1 are assumed to be equally likely.
(a) Using the fact that Q(x) < Q(y) for x > y, show that the probability of a bit error,
Pb with Gray-mapped QPSK modulation is approximately the same as that with BPSK
modulation,
! as a function of the energy per bit-to-noise ratio Eb /N0 . (Hint: P [error] ≈
Q E/N0 , and E = 2 Eb .)
(b) Using the signals ψ1 (t) and ψ2 (t) from problem 2, sketch carefully the signals si (t)
corresponding to the points s̄i , for i = 1, 2, 3, 4, in the QPSK constellation shown in the
figure below.
ψ2(t)
s2 s1
E
ψ1(t)
s3 s4
Solution:
(a) Since Q(x) < Q(y) for x > y, P [error] is dominated by the probability of transmitting a
signal point and making a decision favorable to one !of two possible
√ nearest points. The
distance between two nearest signal points is d12 = 2 E/2 = 2E. From this it follows
that '
E
P [error] ≈ 2 Q .
N0
Since each QPSK signal carries two bits, E = 2Eb . In addition, with Gray mapping,
Pb = P [error]/2. As a result, '
2Eb
Pb ≈ Q .
N0
(b)
97
s1(t) s2(t)
E E
T T
t t
T/2 T T/2 T
E
-
T
s3(t) s4(t)
E E
T T
t t
T/2 T T/2 T
E E
- -
T T
(a) Run the script, sketch or plot the resulting curves and include them in your solution.
(b) Compare the simulated error performance of QPSK and 16-QAM against the approxi-
mated expressions given in the book. You will need to write a script to plot the approx-
imated expressions and the simulation results in the same graph. Refer to the notes on
digital modulation in the web site for the expressions.
Solution: The figures below show plots of simulated (symbols) and theoretical (symbols and
line) BER for M -PSK and M -QAM, respetively. For BER values below 10−2 , the theoretical
expressions match the simulation results very closely.
98
Error performance of M−PSK. SJSU − Fall 2004.
0
10
BPSK
QPSK
−1 8PSK
10
−2
10
BER
−3
10
−4
10
−5
10
−6
10
0 5 10 15 20
Es/N0 (dB)
−2
10
BER
−3
10
−4
10
−5
10
−6
10
0 5 10 15 20 25 30
Es/N0 (dB)
99
104. A digital communication system is designed using bandpass QPSK modulation with Gray
mapping. Due to a DC offset and phase error in the modulator, the constellation is translated
and rotated as shown in the figure below.
ψ2(t)
s1
ψ1(t)
a 2a 4a
s4
s2
s3
(a) Compute the average energy of the signals and sketch carefully the decision regions.
(b) Find the probability of a bit error Pb of this system.
(c) If a demodulator for conventional QPSK is employed, what is the resulting value of Pb ?
Solution:
1
4
E = Ei
4
i=1
1
= (−2a)2 + (a)2 + (−4a)2 + (−2a)2 + (−a)2 + (−4a)2 + (a)2 + (−a)2
4
1
= 44a2 = 11 a2 .
4
!
From which it follows that a = E/11.
100
ψ2(t)
Z1
s1 Z4
ψ1(t)
a 2a 4a
s4
s2
Z2
s3
Z3
and '
26Eb
Pb ≈ Q .
22N0
Compared to conventional QPSK, there is a loss of 10 log 10 44 26 = 2.28 dB in SNR.
This is due to the DC offset. That is, the average of the signal points is not the origin,
resulting in an increase of the average probability of a bit error.
(c) In conventional QPSK, the decision regions Zi , i = 1, 2, 3, 4, are the quadrants of the
Y1 /Y2 -plane. Consequently, the probability of error is given by
1
4
P [error] = P [error|s̄i ],
4
i=1
where
'
'
8E 2E
P [error|s̄1 ] = P [Ȳ ∈
/ Z1 |s̄1 ] = 1−Q 1−Q ,
11N0 11N0
101
'
'
8E 32E
/ Z2 |s̄2 ] = 1 − Q
P [error|s̄2 ] = P [Ȳ ∈ 1−Q ,
11N0 11N0
' '
2E 32E
P [error|s̄3 ] = P [Ȳ ∈
/ Z3 |s̄3 ] ≈ Q +Q ,
11N0 11N0
'
2E
P [error|s̄4 ] = P [Ȳ ∈
/ Z4 |s̄4 ] ≈ 2Q ,
11N0
where the fact that Q(x) > Q(y), for x < y, has been used. It follows that
' -
1 2E
P [error] ≈ 2 + 3Q ,
4 11N0
and therefore, for QPSK modulation (i.e., 2 bits per symbol), the desired result is
' -
1 4Eb
Pb ≈ 2 + 3Q .
8 11N0
Finally, note that at high SNR values, bit error events are dominated by error events
associated with the transmission of signal points s̄1 and s̄2 , which are located in the
wrong quadrants. Therefore,
1
Pb ≈ , for high values of Eb /N0 .
4
In this case, we say that there is an “error floor”. The value of Pb is never less than
0.25, regardless of how high the value of Eb /N0 is!!!
105. The transmitter of a BFSK communication system sends an RF rectangular pulse sm (t), for
m = 1, 2, in the interval 0 ≤ t ≤ T , and in correspondence to the value of a source bit
M ∈ {0, 1}, as follows:
(a) Find the probability of a bit error, P [error], in terms of the amplitude a and N0 .
(b) It is desired to transmit bits at a rate of 1 Mbps with P [error] = 10−5 . The receiver in-
troduces AWGN with N0 = 10−10 . Determine the minimum value amin of the amplitude
of the RF rectangular pulses.
102
(c) It is desired to increase the rate to 2 Mbps. The other conditions are the same as in part
(b). What is the minimum value amin of the amplitude of the RF rectangular pulses?
Express in dB the additional power required compared to part (b).
Solution:
( (
2E 2E
(a) Note that si (t) = T cos(2πfi t), for i = 1, 2. In other words, a = T . Alternatively,
a2 T
the energy of s1 (t) (or s2 (t)) is equal to E = . It follows that
2
' "
E 2
a T
P [error] = Q = Q .
N0 2N0
From theMatlab −5
−1 −5
function invQ.m given in the web site, we have that Q(x) = 10 , for
x=Q 10 = 4.2649. Therefore,
'
a2 (104 )
= 4.2649
2
and from this it follows that a2min = 3.64 × 10−3 V2 , or amin = 60.3 mV.
√
(c) The bit period is reduced by 1/2. Therefore amin = 2 (60.3) mV = 85.3 mV. Compared
to part (b), the additional signal power required to achieve a probability of a bit error
equal to 10−5 is 3 dB.
106. Find z = Eb /N0 required to give Pb = 10−5 for the following coherent digital modulation
schemes: (a) on-off keying (ASK), (b) BPSK, (c) BFSK and (d) BPSK with a phase error of
5 deg.
Solution:
√
(a) For ASK, PE = Q( √ z) = 10−5 gives z = 18.19 or z = 12.6 dB.
(b) For BPSK, PE = Q( 2z) = 10−5 gives z = 9.1 or z = 9.59 dB.
(c) Binary FSK is the same as ASK.
(d) The degradation of BPSK with a phase error of 5 degrees is
so the required SNR to give a bit error probability of 10−5 is 9.59 + 0.033 = 9.623 dB.
103
PAM over bandlimited channels
107. Binary data at a rate of 4 × 106 bits/s are transmitted using M -PAM over a bandlimited
channel, using the raised cosine roll-off characteristic shown in the figure below:
P(f)
T/2
f (MHz)
(a) Determine the roll-off factor α and the value of M of this system.
(b) Signals are transmitted over a bandlimited AWGN channel. Determine the minimum
required bit energy-to-noise ratio, Eb /N0 , to achieve a probability of a bit error less than
10−3 . Express your result in decibels (dB).
Solution:
(a) Rb = 4 Mbits/s. Note that 1/2T = 0.5 MHz, the point of symmetry of the raised cosine.
As a result, 1/T = 1 MHz. Since Rb = 1/Tb = log2 M (1/T ), it follows that log2 M = 4
and therefore M = 16.
To find the roll-off factor, note that 2α(1/2T ) = 0.75 − 0.25 = 0.5 MHz (the region of
the cosine characteristic) and therefore, with 1/2T = 0.5 MHz, we obtain α = 0.5.
(b) The probability of a bit error for M -PAM is
'
M −1 6 log2 M Eb
P [] = Q .
M M 2 − 1 N0
104
108. Binary data are transmitted using 8-PAM over a bandlimited channel, using the raised cosine
roll-off characteristic shown in the figure below:
P(f)
T/2
f (MHz)
(a) Determine the roll-off factor α and the bit rate, in bits/s, of this system.
(b) Signals are transmitted over a bandlimited AWGN channel. Determine the minimum
probability of a bit error if the bit energy-to-noise ratio is limited to a maximum of 20
dB.
Solution:
(a) M = 8, log2 M = 3. Note that 1/2T = 2.5 MHz, the point of symmetry of the raised
cosine. As a result, 1/T = 5 Mhz and the bit rate is therefore Rb = log2 M (1/T ) = 15
Mbits/s.
To find the roll-off factor, note that 2α(1/2T ) = 3.75 − 1.25 = 2.5 MHz (the region of
the cosine characteristic) and therefore, with 1/2T = 2.5 MHz, we obtain α = 0.5.
(b) The probability of a bit error for M -PAM is
'
M −1 6 log2 M Eb
P [] = Q .
M M 2 − 1 N0
105
110. Plot the raised-cosine pulse for different values of rolloff factor, α. For this purpose, download
the script plot raised cosine pulse.m from the web site, execute it and sketch or print the
resulting waveforms.
Solution:
Raised−cosine pulses for various values of the rolloff fact or,α
α=1
1 α=0.5
α=0.25
0.8
0.6
pulse amplitude
0.4
0.2
−0.2
−4 −3 −2 −1 0 1 2 3 4
t/T
111. The eye diagram is useful in analyzing pulse shapes used in communication over bandlimited
channels. You will experiment with this tool using MATLAB. Download the script ayayay.m
from the web site.
(a) Run the script for the following pairs of values E/N0 (dB) and α: [35, 1], [35, 0.5] and
[35, 0.25]. Print or sketch the resulting eye diagrams and comment on the effects of α.
(b) Run the script for the following pairs of values of E/N0 (dB) and α: [35, 1], [25, 1] and
[15, 1]. Print or sketch the resulting eye diagrams and comment on the effects of E/N0 .
Solution:
106
Eye diagram of raised−cosine pulse. E/N =35 dB, and α =1
0
1.5
0.5
Amplitude
0
−0.5
−1
−1.5
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
1.5
0.5
0
Amplitude
−0.5
−1
−1.5
−2
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
1.5
0.5
Amplitude
−0.5
−1
−1.5
−2
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
(a) It is observed from the eye diagrams that the timing error margin increases with the
value of α. On the other hand, the occupied bandwidth grows linearly with the rolloff
factor. Thus there is a tradeoff between timing error margin and occupied bandwidth.
107
Eye diagram of raised−cosine pulse. E/N =35 dB, and α =1
0
1.5
0.5
Amplitude
0
−0.5
−1
−1.5
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
0.5
Amplitude
−0.5
−1
−1.5
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
1.5
0.5
Amplitude
−0.5
−1
−1.5
−2
−0.5 −0.4 −0.3 −0.2 −0.1 0 0.1 0.2 0.3 0.4 0.5
Normalized time, t/T
(b) The eye opening and timing error marging are reduced with increasing noise levels (de-
creasing SNR values).
108
PCM, quantization and delta modulation
112. A message signal m(t) is transmitted by binary PCM without compression. Let the signal-
to-quantization noise (SNRq) required be at least 47 dB. Determine the minimum number
of quantization levels L required, assuming that m(t) is sinusoidal. With this value of L,
determine the SNRq.
Solution: 1.8 + 6v ≥ 47. Therefore, v = 8 bits and L = 256. This gives SNRq = 49.8 dB.
113. In a delta modulation system, a sinusoidal signal m(t) = sin(6000πt) is sampled at 30,000
samples/sec. Determine the minimum step size ∆ to avoid slope overload.
d m(t)
Solution: = 6000π. As a result, ∆ ≥ 6000πTs = π/5.
dt max
114. Find the maximum amplitude of a 1 KHz sinusoidal signal input to a delta modulator that
will prevent slope overload, when the sampling rate is 10,000 samples/sec and the step size
is ∆ = 0.1.
1
Solution: = 10000. ∆ = 0.1. This implies that
Ts
∆
≥ 2000π · A
Ts
1000 ≥ 2000π · A
and consequently,
1
Amax = .
2π
115. Let m(t) = Am cos(2πfm t) be the input to a delta modulator with parameters ∆ (step size)
and Ts (sampling period). Show that the minimum sampling frequency fs,min , needed to
avoid slope overload distorsion, is given by
2π fm Am
fs,min = .
∆
109
from which it follows that
∆ 1 2πfm Am
fs = ≥ .
Ts ∆
117. Data are encoded with a binary (7, 4) Hamming code and sent over a noisy channel. A
received vector is r = (1101111). Determine the syndrome s, the corresponding error vector
e and the estimated code vector v.
Observe that the syndrome s is equal to the third column of H. The corresponding
error
vector is e = 0 0 1 0 0 0 0 , and the estimated code vector v = r⊕e = 1 1 1 1 1 1 1 .
Solution: Any binary matrix with 15 columns consisting of all non-zero combinations of
m = 15 − 11 = 4 bits can be taken as a parity check matrix of a binary (15, 11) Hamming
code. An example is shown below:
0 0 0 0 0 0 0 1 1 1 1 1 1 1 1
0 0 0 1 1 1 1 0 0 0 0 1 1 1 1
H=
0
1 1 0 0 1 1 0 0 1 1 0 0 1 1
1 0 1 0 1 0 1 0 1 0 1 0 1 0 1
119. Consider the application of a (7,4) Hamming code in a digital communication system.
110
Solution:
The syndrome s̄ is equal to the transpose of the fifth column of the parity-check matrix
H. Consequently,
the most
likely event is a single error in the fifth position. That is,
ē = 0 0 0 0 1 0 0 , and the estimated codeword is:
ĉ = r̄ ⊕ ē = 0 1 0 1 1 0 1 ⊕ 0 0 0 0 1 0 0 = 0 1 0 1 0 0 1 .
(b) We have BPSK modulation with 10 log10 NE0 = 8 dB or NE0 = 6.3. The probability of
error without coding is
'
2E
=Q = Q(3.55) = 2.33 × 10−4 .
N0
With coding,
Pb ≈ 32 = 1.63 × 10−7 .
111
(c) Digital logic design of encoder and decoder of the binary (7,4) Hamming code.
1. Encoder Let m̄ represent the message bits. Then
c̄ = c1 c2 c3 c4 c5 c6 c7
1 0 0 0 0 1 1
0 1 0 0 1 1 0
= m̄G = m1 m2 m3 m4 0 0
,
1 0 1 0 1
0 0 0 1 1 1 1
c1 = m1
c2 = m2
c3 = m3
c4 = m4
c5 = m2 ⊕ m3 ⊕ m4
c6 = m1 ⊕ m2 ⊕ m4
c7 = m1 ⊕ m3 ⊕ m4
s1 = r2 ⊕ r3 ⊕ r4 ⊕ r5
s2 = r1 ⊕ r2 ⊕ r4 ⊕ r6
s3 = r1 ⊕ r3 ⊕ r4 ⊕ r7
e1 = s¯1 s2 s3
e2 = s1 s2 s¯3
e3 = s1 s¯2 s3
e4 = s1 s2 s3
e5 = s1 s¯2 s¯3
e6 = s¯1 s2 s¯3
e7 = s¯1 s¯2 s3
112
The circuits for encoding and decoding the (7,4) Hamming code are shown below.
m1 c1
m2 c2
m3 c3
m4 c4
c5
c6
c7
c1
c2
c3
c4
c5 ~
c1
c6
c7 ~
c2
s1 s2 s3 ~
e1 c3
e2 ~
c4
e3 ~
c5
e4 ~
c6
e5 ~
c7
e6
e7
113
120. MATLAB experiment
Download the Matlab script homework10f04.m from the website. Record the output and
verify it. That is, verify that, for the given message and error vectors, the codeword, the
received word and the syndrome are correct. (Note that in Matlab the position of the identity
matrices in the generator and parity-check matrices of the code is different from that given
in class.)
Solution:
To get started, select MATLAB Help or Demos from the Help menu.
>> homework10
Enter your student ID (tower card) number: 123456789
G =
1 1 0 1 0 0 0
0 1 1 0 1 0 0
1 1 1 0 0 1 0
1 0 1 0 0 0 1
H =
1 0 0 1 0 1 1
0 1 0 1 1 1 0
0 0 1 0 1 1 1
trt =
0 0 0 0 0 0 0
0 0 1 0 0 0 0
0 1 0 0 0 0 0
0 0 0 0 1 0 0
1 0 0 0 0 0 0
0 0 0 0 0 0 1
0 0 0 1 0 0 0
0 0 0 0 0 1 0
msg =
1 0 1 1
code =
1 0 0 1 0 1 1
err =
0 1 0 0 0 0 0
recd =
1 1 0 1 0 1 1
114
12/15/04 12:25 AM MATLAB Command Window 2 of 2
errvect =
0 1 0 0 0 0 0
correctedcode =
1 0 0 1 0 1 1
121. Find the capacity of the cascade connection of n binary-symmetric channels with the same
crossover probability . What is the capacity when the number of channels goes to infinity?
Solution: The overall channel is a binary symmetric channel with crossover probability p.
To find p note that an error occurs if an odd number of channels produce an error. Thus,
n
p= k (1 − )n−k
k
k=odd
which is equal to
1
p= 1 − (1 − 2)2
2
and therefore,
C = 1 − h(p)
If n → ∞, then (1 − 2)n → 0 and p → 12 . In this case
1
C = lim C(n) = 1 − h( ) = 0
n→∞ 2
122. In the transmission and reception of signals to and from moving vehicles, the transmitted
signal frequency is shifted in direct proportion to the speed of the vehicle. The so-called
Doppler frequency shift imposed on a signal that is received in a vehicle traveling at a velocity
v relative to a (fixed) transmitter is given by the formula
v
fD = ± ,
λ
where λ is the wavelength and the sign depends on the direction (moving toward or moving
away) that the vehicle is traveling relative to the transmitter. Suppose that a vehicle is
traveling at a speed of 100 km/hr relative to a base station in a mobile cellular communication
system. The signal is narrowband and transmitted at a carrier frequency of 1 GHz.
115
Solution:
The plus sign holds when the vehicle travels towards the transmitter whereas the minus
sign holds when the vehicle moves away from the transmitter.
(b) The maximum Doppler frequency shift is obtain when f = 1 GHz + 1 MHz and the
vehicle moves towards the transmitter. In this case
3 × 108
λmin = m = 0.2997 m
109 + 106
and therefore
100 × 103
fD max = = 92.6853 Hz
0.2997 × 3600
Thus, the Doppler frequency spread is Bd = 2fD max = 185.3706 Hz.
123. A multipath fading channel has a multipath spread of Tm = 1 s and a Doppler spread
Bd = 0.01 Hz. The total channel bandwidth at bandpass available for signal transmission is
W = 5 Hz. To reduce the effect of intersymbol interference, the signal designer selects a pulse
duration of T = 10 s.
Solution:
(b) Since the channel bandwidth W Bcb , the channel is frequency selective.
(c) Since the signal duration T Tct , the channel is slowly fading.
116
124. Demonstrate that a DS spread-spectrum signal (without coding) provides no improvement in
performance against AWGN.
Solution: The probability of error for DS spread spectrum with binary PSK may be expressed
as "
2W/Rb
P2 = Q
PJ /PS
where W/R is the processing gain and PJ /PS is the jamming margin. If the jammer is a
broadband, WGN jammer, then
PJ = W J0
PS = Eb /Tb = Eb Rb
Therefore, '
2Eb
P2 = Q
J0
125. A DS spread-spectrum system is used to resolve the multipath signal component in a two-
path radio signal propagation scenario. If the path length of the secondary path is 300 m
longer than that of the direct path, determine the minimum chip rate necessary to resolve
the multipath components.
Solution: The radio signal propagates at the speed of light, c = 3×108 m/ sec . The difference
in propagation delay for a distance of 300 meters is
300
Td = = 1µ sec
3 × 108
The minimum bandwidth of a DS spread spectrum signal required to resolve the propagation
paths is W = 1 M Hz. Hence, the minimum chip rate is 106 chips per second.
117
SAMPLE EXAMS
118
EE160. Spring 2005. San José State University
5j
-1/2 f
0 1/2
B=1/5
-5j
X (f ) = j Xs (f ) = −2j · 5 Π(5f ),
which is sketched below.
Xl(f)
f
0
-10j
1/5
119
Spectrum of the complex baseband signal.
f
-1/3 0 1/3
(b) Sketch the ideal sampled spectrum Xδ (f ), for the sampling rate fs = 3fs,min .
The sampled spectrum Xδ (f ) for fs = 2 is shown below.
Xδ(f)
… …
f
-6 -4 -2 0 2 4 6
2/3
Sampled spectrum Xδ (f ).
(c) Sketch the output x̃(t) of an ideal (rectangular) bandpass filter with f0 = 4, B = 0.7, and
gain 1/4.
The spectrum X̃(f ) at the output of the filter is:
~
X (f)
1/2
f
-4 0 4
2/3 2/3
120
The amplitude envelope is a sinc function with zero crossings at multiples of 3/2. Also, the
sinusoidal waveform was period T0 = 1/4. A sketch is shown below:
~
x(t)
1/3
-3 -3/2 3/2 3
-1/3 T0=1/4
Problem 3 (15 points) Using the properties of the Fourier transform, find the value of the integral
∞
I= sinc2 (t) cos(6πt) dt.
−∞
Note (based on an idea of C. Igwebuike): The same result is obtained using Parserval’s theorem.
This follows from the evaluation property of the impulse function:
∞
X(f )δ(f − f0 )df = X(f0 ).
−∞
X(f)
2/3
2/3 sinc2(2f/3)
f
-4.5 -3 -1.5 0 1.5 3 4.5
3/8
121
Spectrum of a signal.
(a) Is this signal energy-type or power-type? Justify your answer. (Hint: You do not need to
compute the power or the energy.)
The signal spectrum,
∞
2 2 3n 3n
X(f ) = sinc2 δ f− ,
n=−∞
3 3 8 8
and
∞
∞
8 3 8n
x(t) = xT0 (t − nT0 ) = Λ t− ,
n=−∞
3 n=−∞
2 3
8/3
t
-16/3 -8/3 0 8/3 16/3
4/3
122
EE160. Fall 2003. San José State University
s(t)
1
… … t
-2 −τ/2 τ/2 2
τ τ
Solution:
(a) The Fourier transform of the signal sT (t), over one period T = 2 seconds, is
ST (f ) = F (Π(t)) = τ sinc(τ f ).
123
S(f)
τ/2
fτ/2
-3 -2 -1 1 2 3
1/T
S(f)
2
f
-8 8
2 2
(a) Find an expresion of signal s(t). (Hint: The spectrum is a linear combination of a
rectangular pulse and a triangular pulse.)
(b) Sketch carefully the spectrum of the sampled signal using a sampling rate equal to
Rs = 1.5Rnyq , where Rnyq denotes the Nyquist rate.
(c) Specify a lowpass filter characteristic that allows reconstruction of s(t) from its samples.
Solution:
(b) This is a bandpass signal. Nyquist rate is Rs = 2fu /[fu /2] = 4.5. With this rate, the
spectrum is shown in the figure below, where a = 4.5
124
Sδ(f)
2a
… … f
-8 -6 -4 -2 2 4 6 8
2a
f
-27 -8 8 27
3. (20 points) Sketch the spectrum of the complex baseband equivalent of the bandpass signal
whose spectrum is depicted in the following figure.
S(f)
1
f
-5 5
2 2
f
-1 1
125
EE160. Fall 2003. San José State University
1. (35 points) A binary baseband communication system transmits one bit every T seconds,
with T = 4. The pulse used is shown in Fig. 3 below.
x(t)
1/2
1 2 3 4 t
-1/2
(a) Find the impulse response h(t) of the matched filter for x(t).
(b) Determine the output of the matched filter at t = 4 when the input is x(t).
(c) For transmission over an AWGN channel, polar mapping is employed, i.e., a bit bk is
mapped onto a voltage level ak as follows:
bk → ak
0 -1
1 +1
After polar mapping the levels are input to a transmit filter with impulse response x(t)
every 4 seconds. The AWGN process has power spectral density equal to 1/2. Determine
the probability of a bit error of the optimal receiver. Express your answer in terms of
the Gaussian Q-function.
Solution:
1/2
1 2 3 4 t
-1/2
126
(c) Polar mapping:
' "
2Ex 2( 12 )
P [] = Q = Q = Q (1)
N0 (1)
2. (30 points) Binary data at a rate of 4 × 106 bits/s are transmitted using M -PAM over a
bandlimited channel using a raised cosine roll-off characteristic as shown in Fig. 4 below.
P(f)
T/2
f (MHz)
(a) Determine the roll-off factor α and the value of M of this system.
(b) Signals are transmitted over a bandlimited AWGN channel. Determine the minimum
required bit energy-to-noise ratio, Eb /N0 , to achieve a probability of a bit error less than
10−3 . Express your result in decibels (dB).
Solution:
(a) Rb = 4 Mbits/s. Note that 1/2T = 0.5 MHz, the point of symmetry of the raised cosine.
As a result, 1/T = 1 MHz. Since Rb = 1/Tb = log2 M (1/T ), it follows that log2 M = 4
and therefore M = 16.
To find the roll-off factor, note that 2α(1/2T ) = 0.75 − 0.25 = 0.5 MHz (the region of
the cosine characteristic) and therefore, with 1/2T = 0.5 MHz, we obtain α = 0.5.
(b) The probability of a bit error for M -PAM is
'
M −1 6 log2 M Eb
P [] = Q .
M M 2 − 1 N0
127
3. (35 points) A communication system is designed for binary data transmission over an ideal
AWGN channel, with RZ unipolar pulses of average energy equal to Es = A2 T /4. This
system is then applied to an optical link that has the feature that the variance of the AWGN
process varies with the signal level.
Specifically, it is found that the sampled output of the matched filter (with kM F = 1) Y is a
Gaussian random variable with conditional PDF given by
y2
√ 1
2πσ2 Es exp − 2σ2 Es , ak = 0 (bk = 0);
pY (y|ak ) =
√ 1
exp − (y−E s)
2
, ak = +1 (bk = 1).
4πσ2 Es 4σ2 Es
In other words, the variance of Y equals σ 2 Es is the transmitted level is 0 and 2σ 2 Es if the
transmitted level is +1.
(a) Find the probability of a bit error for the optical link in terms of the energy-to-noise
power ratio, Es /σ 2 . Express your result in terms of the Gaussian Q-function.
(b) Give an expression to determine the threshold of the decision device used in an optimal
receiver for the optical link.
Solution:
(a) The conditional PDF’s of the matched filter output are sketched below:
pY(y|ak)
0 Es/2 +Es
Probability of error:
Es /2 − 0 Es /2 − Es
P [] = p Q √ 2 + (1 − p) 1 − Q √
σ Es 2σ 2 Es
' '
Es Es
= pQ 2
+ (1 − p) Q
4σ 8σ 2
128
λ2 (λ − Es )2 1 1−p
− = − − ln 2 + ln
2σ 2 Es 4σ 2 Es 2 p
1−p
−→ λ2 + 2λEs − Es2 + σ 2 Es 2 ln − ln 2 = 0.
p
129
EE160. Spring 2005. San José State University
Solution Midterm Exam # 2-B
Problem 1 (25 points) A line coding scheme uses Manchester encoding with rectangular pulses.
x(t)
1 1 0 1 0 1
A
t
T 2T 3T 4T 5T 6T
-A
(b) Sketch carefully, showing all relevant labels, the power spectral density of this scheme.
0.7
2
0.6 E=A T
0.5
0.4
S(fT)/E
0.3
0.2
0.1
0
0 0.5 1 1.5 2 2.5 3 3.5 4
fT
(c) The energy per bit is Eb = 1 × 10−9 Joules and the bit rate is Rb = 1 × 106 bits/sec. What
is the amplitude of the pulses?
The bit = 106 . The energy
√b = 1/T !
! rate is R
2
√ per bit is Eb = A T from which it follows that
−9 6 −3
A = Eb /T = Eb Rb = (10 )(10 ) = 10 = 31.62 mV.
(d) The noise at the receiver is AWGN with σn2 = 1.25 × 10−10 W/Hz. Determine the probability
of a bit error.
' '
2Eb 2 · 10−9
√
Pe = Q =Q = Q 8 = 2.34 × 10−3 .
N0 2 · 1.25 × 10−10
130
Problem 2 (40 points) A binary communication system uses NRZ pulses and polar mapping:
2 ψ(t), bit = “0”
s(t) =
−2 ψ(t), bit = “1”,
(b) Sketch carefully a block diagram of the demodulator. In particular, you must give a detailed
specification of the impulse response of the matched filter, the sampling time, and the decision
rule.
Y=y(3)
r(t) ψ(3-t) M
t=3
Decision rule:
0, Y > 0;
M̂ =
1, Y ≤ 0,
(c) Noise at the receiver is AWGN with σn2 = 3. Determine the probability of a bit error, Pe .
' ' '
2Eb 2·4 4
Pe = Q =Q =Q = 1.24 × 10−1 .
N0 2·3 3
(d) By mistake, the transmitter sends RZ pulses with polar mapping. Determine the resulting
probability of a bit error, Pe,mistake .
If RZ pulses of amplitude A are sent and the receiver has a filter (or correlator) matched to
NRZ pulses, then the energy is scaled by one half. Consequently,
' ' '
Eb 4 2
Pe,mistake = Q =Q =Q = 2.07 × 10−1 .
N0 2·3 3
131
Problem 3 (35 points) The pulse waveform shown below is used for binary communication with
unipolar mapping (also known as on-off keying).
s(t)
3a
2a
a
t(µsec)
0 1 2 3
(a) Determine the value of a in terms of the average energy per bit Eb .
!
The energy of the pulse is Es = (3a)2 (1)+(2a)2 (1)+a2 (1) = 14a2 , and therefore a = Es /14.
! other hand, the average energy per bit with unipolar mapping is Eb = Es /2. Therefore,
On the
a = Eb /7.
h(t)=s(3-t)
3a
2a
a
t(µsec)
0 1 2 3
(c) It is desired that the bit rate Rb (bits/sec) be at least 1 × 106 . Using M -PAM modulation,
you are asked to determine the minimum value of M .
The pulse duration is T = 4 × 10−6 sec. The bit rata rate of M -PAM, with M = 2m , is
Rb = m/T . The requirement is
m
≥ 106 → m ≥ 3 × 10−6 106 → m ≥ 3.
T
Therefore, the minimum value is M = 23 = 8.
(d) With the value of M found in part (c), determine the new value of a in terms of the average
energy per symbol Eave .
For 8-PAM,
! the average
! energy is: Eave = (M 2 − 1)Es /3 = 21Es . As a result, the new value
is a = Es /7 = Eave /147.
132
EE160. Spring 2005. San José State University
Solution Final Exam
Problem 1 (35 points) The raised-cosine spectrum X(f ) of a digital communication system is
shown in Fig. 5 below. The excess bandwidth is 25%.
X(f)
b/2
f
-f3 -f2 -f1 f1 f2 f3
(a) The system transmits information at a rate of 10 Mbps and utilizes 4-PAM modulation.
Determine the values of b, f1 , f2 and f3 .
We have m = 2 and Rb = 10×106 bps. Since Rb = m/T , it follows that 1/T = Rb /2 = 5×106
baud. Therefore,
b = T = 2 × 10−7 = 200 ns
1
f2 = = 2.5 × 106 = 2.5 MHz
2T
(1 + α)
f3 = = (1.25) 2.5 × 106 = 3.125 × 106 = 3.125 MHz
2T
(1 − α)
f1 = = (0.75) 2.5 × 106 = 1.875 × 106 = 1.875 MHz
2T
(Type B: b = 400 ns, f2 = 1.25 MHz, f3 = 1.6625 MHz, and f1 = 837.5 KHz.)
(b) RZ rectangular pulses are utilized. Determine the amplitude of the pulses if N0 = 1 × 10−8
and the probability of a bit error Pb is desired to be lower than 10−2 .
With RZ pulses, E = a2 T /2. For 4-PAM modulation, the requirement that Pb < 10−2
translates into
' '
3 4Eb 4Eb
Pb ≈ Q < 10−2 −→ Q < 1.33 × 10−2 .
4 5N0 5N0
With N0 = 1 × 10−8 ,
133
'
−8 a2 T 2(6.6125 × 10−8 )
Eb = 6.6125 × 10 = −→ a= = 0.813 V.
2 2 × 10−7
(Type B: Eb /N0 = 6.8 dB, and a = 1.095 V.)
(c) The reliability is improved by requiring that Pb < 10−3 ? What amount of additional power
will be needed? Express your answer in dB.
(
With the aid of Table 1, the requirement Pb < 10−3 translates into 5N
4Eb
0
> 3.1, from which
it follows that
Eb 5
= (3.1)2 = 12.0125 = 10.8 dB.
N0 impr 4
s(t)
0.075
t(µsec)
0 0.35
(a) The noise at the receiver is AWGN with σn2 = 5 × 10−10 . Determine the probability of a bit
error.
134
(b) Sketch the output of the correlator in the range 0 to 1 µsec. (Hint: No integrals.)
y(t)
1.97x10-9
t(µsec)
0 0.35 1
(Type B:
y(t)
1x10-9
t(µsec)
0 0.25 0.5 1
(c) Sketch carefully, with as much detail as possible, a block diagram of this system, including
both transmitter and receiver. (The more detail is shown the higher the grade of this part.)
0 - E
M
1 + E
Bits-to
Signal ψ(t)
Mapping
n(t) Σ
0
t=T=1 µs
ψ(t)
0, Y <= 0
M =
1, Y > 0
!
The waveform ψ(t) is a rectangular pulse of duration 0.35 µs and amplitude 1/(0.35 × 10−6 ) .
135
(d) If the transmitter sends NRZ pulses at the same rate and with the same energy as the pulse
of Fig. 6, and the receiver is the same as that of part (c), determine the resulting probability
of a bit error.
−6
An NRZ pulse with√ the same energy and duration is a rectangular pulse of width 1 × 10
−3
s and amplitude 1.96875 × 10 V. The correlator output is a waveform similar to that of
the previous figure with maximum amplitude (energy) a = (1.96875 × 10−3 )(0.35 × 10−6 ) =
6.89 × 10−10 . Therefore,
'
2(6.89 × 10−10 )
Pb = Q = Q(1.174) ≈ 4 × 10−2 .
1 × 10−9
Problem 3 (35 points) A bandpass QPSK modulation system uses the following orthonormal
signals,
ψ1 (t) = 110 cos 104 πt , ψ2 (t) = 110 sin 104 πt .
Noise is AWGN with N0 = 1 and the target probability of a bit error is Pb = 10−5 .
For QPSK modulation, m = 2 and Pb = 10−5 , we have that (Homework 9!) Eb /N0 ≥ 9.12.
Therefore, with N0 = 1, E = 2Eb ≥ 18.24. Also, the amplitude of the signal is
'
2E 1 (110)2
= 110 −→ ≤ = 331.7 baud.
T T 2 × 18.24
As a result,
2
Rb,max = = 663.38 bps.
T
4
(Type B: Rb,max = T = 4002 bps.)
(b) The channel bandwidth is 450 Hz and a raised-cosine spectrum is employed to eliminate
ISI. Determine the excess bandwidth and sketch carefully the spectrum, paying particular
attention to the frequency values.
We have B = 1
T (1 + α) = 450. Therefore, α = 450
331.7 − 1 = 0.3567, or 35.7%
X(f)
A/2
f (KHz)
4.775 4.8712 4.8974 5 5.107 5.1658 5.225
136
In the figure above, A = 2.12 × 10−3 .
(Type B: B = 1220, α = 0.22 or 21.94%, and Rb /B = 3.28 bps/Hz. Also, A = 7.07 × 10−4 .)
Ψ 2(t)
10 E/2 00
Ψ 1(t)
- E/2 E/2
11 01
- E/2
(Type B:
Ψ 2(t)
Ψ 1(t)
-3 E/10 - E/10 E/10 3 E/10
137
(d) A new RF technology produces a 5 dB improvement in SNR at the receiver. Discuss how this
can be used to enhance performance.
(1) To achieve a reduced probability of a bit error or bit error rate (BER), for the same
transmitted power
(2) A lower transmitted power with the same target BER
(3) An increased transmission rate for the same BER. For example, the bit rate could be
doubled to Rb,new = 2 × 663.38 = 1266.8 bps by using 16-QAM modulation, which
requires 14.6 dB (5 dB more than QPSK modulation). See also the solution of Homework
9. (Type B: Can use 32-QAM and increase the rate to Rb,new = 54 × 4002 = 5002.5 bps.)
138
EE160. Fall 2003. San José State University
1A. The orthonormal pulses shown in Fig. 7 are used in a transmission system using quadrature
modulation.
φ1(t) φ2(t)
2 2
t t
1/2 1 1/2 1
- 2 - 2
(a) Determine the pulses associated with the points in the φ1 φ2 -plane shown in Fig. 8 (a).
(b) Determine the signal point s associated with the pulse s(t) shown in Fig. 8 (b).
φ2(t)
2α s(t)
s1 α
1/3
α 2α t
φ1(t)
−2α −α 1/2 1
-1/3
−α -2/3
−2α
s2
(a) (b)
139
Solution:
(a)
s1(t) s2(t)
2α 2 2α
t
2α
1/2 1
t
− 2α
1/2 1
(b) The coordinates of the point are equal to the outputs of the filters matched to φ1 (t) and
φ2 (t), respectively, and given by
1/2 √
1 1 √ 2
s1 = s(t)φ1 (t) dt = − ( 2) = − ,
0 2 3 6
1 √
1 2 √ 2
s2 = s(t)φ2 (t) dt = − (− 2) = ,
1/2 2 3 3
φ2(t)
s
2/3
φ1(t)
- 2/6
140
1B. The orthonormal pulses shown in Fig. 11 are used in a transmission system using quadrature
modulation.
φ1(t) φ2(t)
2 2
t t
1/2 1 1/2 1
- 2 - 2
(a) Determine the pulses associated with the points in the φ1 φ2 -plane shown in Fig. 12 (a).
(b) Determine the signal point s associated with the pulse s(t) shown in Fig. 12 (b).
φ2(t)
s(t)
2α s1
1
α
1/2
α 2α t
φ1(t)
−2α −α 1/2 1
−α -1/2
-1
−2α
s2
(a) (b)
141
Solution:
s1(t) s2(t)=-s1(t)
2 2α 2 2α
2α 2α
t t
1/2 1 1/2 1
− 2α − 2α
−2 2α −2 2α
(b) The coordinates of the point are equal to the outputs of the filters matched to φ1 (t) and
φ2 (t), respectively, and given by
3/4 √
1 1 √ 1 1 √ 2
s1 = s(t)φ1 (t) dt = − ( 2) + (− 2) = − ,
1/4 4 2 4 2 4
√
1/4 1
1 √ 1 √ 2
s2 = s(t)φ2 (t) dt + s(t)φ2 (t) dt = (1)( 2) + (−1)(− 2) = .
0 3/4 4 4 2
φ2(t)
s
2/2
φ1(t)
- 2/4
142
2. A binary modulation system uses quadrature modulation with the signal points shown in Fig. 15.
Transmission of equally likely bits takes place over an AWGN channel with power spectral density
N0 /2.
φ2(t)
Z1
2α
s1
α Z2
α 2α φ1(t)
−2α −α
−α
−2α s2
(a) Find the value of the constant α as a function of the average signal energy E.
(b) Sketch carefully the decision regions.
(c) Find the probability of a bit error P [], in terms of E/N0 and the Q-function, and compare
it with binary transmission using square pulses and polar mapping.
Solution:
(
E
(a) α = 5.
143
3. This problem deals with a Hamming code with m = 2 redundant bits. This (3, 1) code is also
know as a repetition code of length n = 3.
(b) Construct a look-up table with syndrome s as input and error vector e as output.
(c) If the
vector
r = 1 0 0 is received, determine the estimated code vector v. (Type B:
r = 1 0 1 .)
Solution:
sT e
00 0 0 0
11 1 0 0
10 0 1 0
01 0 0 1
144
Table 1: The Gaussian Q-function
0.0 0.1 0.2 0.3 0.4 0.5 0.6 0.7 0.8 0.9
0.0 5.00e-01 4.60e-01 4.21e-01 3.82e-01 3.45e-01 3.09e-01 2.74e-01 2.42e-01 2.12e-01 1.84e-01
1.0 1.59e-01 1.36e-01 1.15e-01 9.68e-02 8.08e-02 6.68e-02 5.48e-02 4.46e-02 3.59e-02 2.87e-02
2.0 2.27e-02 1.79e-02 1.39e-02 1.07e-02 8.20e-03 6.21e-03 4.66e-03 3.47e-03 2.56e-03 1.87e-03
3.0 1.35e-03 9.68e-04 6.87e-04 4.83e-04 3.37e-04 2.33e-04 1.59e-04 1.08e-04 7.23e-05 4.81e-05
4.0 3.17e-05 2.07e-05 1.33e-05 8.54e-06 5.41e-06 3.40e-06 2.11e-06 1.30e-06 7.93e-07 4.79e-07
5.0 2.87e-07 1.70e-07 9.96e-08 5.79e-08 3.33e-08 1.90e-08 1.07e-08 5.99e-09 3.32e-09 1.82e-09
Q(x) x
10−1 1.28
10−2 2.33
10−3 3.10
10−4 3.73
10−5 4.27
10−6 4.76
10−7 5.20
10−8 5.61
10−9 6.00
10−10 6.63
10−11 6.71
10−12 7.03
10−13 7.35
10−14 7.65
145