20. N. Levinson, “The Wiener rms Error Criterion in Filter Design and Prediction,” J. Math. Phys., vol.

25, pp. 261–278, 1947.

21. A. H. Sayed, Fundamentals of Adaptive Filtering, Wiley-IEEE Press, Hoboken, NJ, 2003.
22. J. Y. Stein, Digital Signal Processing: A Computer Science Perspective, Wiley, Hoboken, NJ, 2000.
23. K. K. Paliwal and B. W. Kleijn, “Quantization of LPC Parameters,” in Speech Coding and Synthesis,
W. B. Kleijn and K. K. Paliwal, Eds. Elsevier Science, Amsterdam, 1995.
24. T. E. Tremain, “The Government Standard Linear Predictive Coding Algorithm LPC-10,” Speech
Technol., 40–49, 1982.
25. M. R. Schroeder and B. S. Atal, “Code-Excited Linear Prediction (CELP): High-Quality Speech at
Very Low Bit Rates,” in Proc. IEEE Int. Conf. Acoustics, Speech, Signal Process. (ICASSP),

Modulation Analogique
vol. 10, pp. 937–940, 1985.
26. S. Mallat, “A Theory of Multiresolution Signal Decomposition: The Wavelet Representation,” IEEE
Trans. Pattern Anal. Machine Intel., vol. 11, pp. 674–693, 1989.
27. M. J. Smith and T. P. Barnwell, “Exact Reconstruction for Tree Structured Sub-Band Coders,” IEEE
et Numérique
Trans. Acoustics, Speech, Signal Process., vol. 34, no. 3, pp. 431–441, 1986.
28. B. G. Haskel, A. Puri, and A. N. Netravali, Digital Video: An Introduction to MPEG-2, Chapman &
Hall, New York, 1996.
29.AND
SAMPLING J. L.ANALOG-TO-DIGITAL Exercices:
Mitchell, W. B. Pennebaker, C.E Fogg,Chap
CONVERSION and D. J.6LeGall,
et Chap MPEG7Video Compression Standard,
Chapman & Hall, New York, 1996.
30. ITU-T
ShowRecommendation
that the signal g(t)H.263,
can be Video Coding
recovered for Low
by passing g(t)Bit Rate Communication.
through an ideal low-pass filter of
bandwidth B Hz and a gain of 4.

OBLEMS6.1-4 A signal g(t) = sinc2 (5π t) is sampled (using uniformly spaced impulses) at a rate of (i) 5 Hz;
(ii) 10 Hz; (iii) 20 Hz. For each of the three cases:

6.1-1 (a) Sketch

Figure the sampled
P6.1-1 shows signal.
Fourier spectra of signals g1 (t) and g2 (t). Determine the Nyquist interval
andSketch
(b) the sampling rate of
the spectrum forthe
signals g1 (t),
sampled g2 (t), g12 (t), g2m (t), and g1 (t)g2 (t).
signal.
Hint:
(c) Use the
Explain frequency
whether you canconvolution and the
recover the signal width
g(t) from property
the sampledof the convolution.
signal.
(d) If the sampled signal is passed through an ideal low-pass filter of bandwidth 5 Hz, sketch the
P.6.1-1 spectrum G2(!)
G1(!) of the output signal.

6.1-5 Signals g1 (t) = 104 "(104 t) and g2 (t) = δ(t) are applied at the inputs of ideal low-pass filters
! !
H1 (f ) = "(f /20, 000) and H2 (f ) = "(f /10, 000) (Fig. P6.1-5). The outputs y1 (t) and y2 (t)
of these0 filters
2" are 5
# 10multiplied 0 5
to obtain the signal y(t) = 3" # 10
y1 (t)y 2 (t). Find the Nyquist rate of
y1 (t),(a)
Chap 6
y2 (t), and y(t). Use the convolution (b)
property and the width property of convolution to
determine the bandwidth of y1 (t)y2 (t). See also Prob. 6.1-1.

e P.6.1-5 6.1-2
g ( t )Determine the Nyquist
y (t)
sampling rate and the Nyquist sampling interval for the signals:
1 1
H (f)
(a) sinc (100π
1
t)
y ( t ) = y1 ( t ) y2 ( t )
(b) sinc2 (100π t)
g2 ( t )(c) y2 ( t (50π
sinc (100π t) + sinc ) t)
H (f)
2
(d) sinc (100π t) + 3 sinc2 (60π t)
(e) sinc (50π t) sinc (100π t)

6.1-3 AAzero-order
6.1-6 signal g(t)
holdband-limited to B Hz
circuit (Fig. P6.1-6) is sampled
is often by a periodic
used to reconstruct pulse
a signal g(t)train (t) made up of a
from pitsTssamples.
rectangular pulse of width 1/8B second (centered at the origin) repeating at the Nyquist rate
e P.6.1-6 Input (2B pulses per second). Show that−the sampled signal g(t) is given by
Output
∫ Delay
Ts
Σ
∞
1 + 2
! " nπ #
g(t) = g(t) + sin g(t) cos 4nπBt
4 nπ 4
n=1

(a) Find the unit impulse response of this circuit.

(b) Find the transfer function H (f ) and sketch |H (f )|.
(c) Show that when a sampled signal g(t) is applied at the input of this circuit, the output is a
staircase approximation of g(t). The sampling interval is Ts .
2
6.1-7 (a) A first-order hold circuit can also be used to reconstruct a signal g(t) from its samples. The
impulse response of this circuit is
 6.1-8 Prove that a signal cannot be simultaneously time-limited and band-limited.
bandwidth required to transmit this signal.
(c)
Hint:
Hint: Show
ForShow that
single
that the
thecontrary
error detection
contrary assumption
capability,
assumption leads
an to contradiction.
additional
leads Assume
bit (parity
to contradiction. bit) isa signal
Assume added simultaneously
to the
a signal code of each
simultaneously
time-limited
character.and band-limited
Modify your so that
answers G(fin =
parts
) 0 for
(a) |f
and | > B.
(b) inIn this
view
time-limited and band-limited so that G(f ) = 0 for |f | > B. In this case, G(f ) = G(f ) !(f case,
of G(f
this = G(f
information.
) ) !(f /2B′/2B
) ′)
′ ′ ′
forBB′>>B.B.This
for Thismeans
meansthat g(t)
that is is
g(t) equal
equal to to
g(t)
g(t)∗ 2B∗ 2B ′ sinc
sinc (2πB (2πB t). ′Show
t). Show thatthat
the latter cannot
the latter cannot
6.2-2 beA time-limited.
compact disc
be time-limited. (CD)
bandwidth equals 15 kHz.
records audio Chap 6
signals digitally by using PCM. Assume that the audio signal

6.2-1
6.2-1 TheAmerican
The AmericanStandard
StandardCode CodeforforInformation
Information Interchange
Interchange (ASCII)
(ASCII) has has
128128characters,
characters,whichwhich
(a)binary-coded.
are If the NyquistIfsamples
a certain are uniformly
computer quantized
generates 100,000 = 65, 536per
into L characters levels and then
second, binary-coded,
determine the
are binary-coded. If a certain computer generates 100,000 characters per second, determine the
determine the number of binary digits required to encode a sample.
following:
following:
(b) The
(a) If the audio of
number signal has average
bits (binary digits)power of 0.1
required perwatt and peak voltage of 1 volt. Find the resulting
character.
(a) signal-to-quantization-noise
The number of bits (binary digits) requiredof
ratio (SQNR) perthecharacter.
uniform quantizer output in part (a).
(b) The number of bits per second required to transmit the computer output, and the minimum
(b) Determine
(c) The number theofnumber
bits perofsecond
binary required pertosecond
transmit the computer
required tooutput,encodeand thethe minimum
bandwidth required to transmit thisdigits
signal. (bit/s) audio signal.
bandwidth required to transmit this signal.
(d) For
(c) Forsingle
practical
errorreasons
detectiondiscussed in the
capability, an text, signals
additional bitare sampled
(parity bit) is at added
a rate well
to theabove
code the Nyquist
of each
(c) rate.
For single
PracticalerrorCDs
detection
use capability,
44,100 samples an additional
per second. bit If(parity
L = bit)
65, is added
536, to the code
determine the of each
number
character. Modify your answers in parts (a) and (b) in view of this information.
character.
of bits perModify
second your answers
required in partsthe
to encode (a)signal,
and (b)and in view of this information.
the minimum bandwidth required to
transmit
6.2-2 A compact discthe encoded
(CD) records signal.
audio signals digitally by using PCM. Assume that the audio signal
6.2-2 bandwidth
A compactequals disc (CD) records audio signals digitally by using PCM. Assume that the audio signal
15 kHz.
6.2-3 A bandwidth
televisionequals
signal15(video
kHz. and audio) has a bandwidth of 4.5 MHz. This signal is sampled,
(a) If the Nyquist samples are uniformly quantized into L = 65, 536 levels and then binary-coded,
quantized,
(a)determine and binary
If the Nyquist coded
samples touniformly
obtain a PCM signal. L = a65,
the number ofare
binary quantized
digits required tointo
encode 536 levels and then binary-coded,
sample.
determine the
(a) IfDetermine the number
sampling ofrate
binary digits
if the required
signal is toand to encode
be sampled a atsample.
aofrate 20% above the Nyquist
(b) the audio signal has average power of 0.1 watt peak voltage 1 volt. Find the resulting
(b)signal-to-quantization-noise
rate.
If the audio signal has average ratiopower
(SQNR) of 0.1 wattuniform
of the and peak voltageoutput
quantizer of 1 volt. Find(a).
in part the resulting
(c) signal-to-quantization-noise
(b) Determine
If the samples are quantized
the number of binary ratio
into (SQNR)
1024per
digits levels, of
second the uniform
determine quantizer
the number
(bit/s) required output
of binary
to encode in part
pulses
the audio (a).
required
signal.
(c)For
(d) to encode
practicaleach
Determine the sample.
number
reasons of binary
discussed in the digits
text,per second
signals (bit/s) required
are sampled at a rateto wellencode
abovethe theaudio signal.
Nyquist
(c)
(d)rate.
Determine
For Practical thereasons
practical binary
CDs pulse
usediscussed
44,100 ratesamples
(bits
in theper
text,
persecond)
signals
second. ofare
theL
If binary-coded
sampled
= 65, 536,at a rate signal, and
well above
determine thethe minimum
the Nyquist
number
ofbandwidth
rate. required
bits Practical
per second CDs to
usetransmit
required 44,100
to encodethis signal.
samples per second.
the signal, and the If L = 65, 536,
minimum determine
bandwidth the number
required to
transmit the encoded signal.
of bits per second required to encode the signal, and the minimum bandwidth required to
3
6.2-4 Five transmit
telemetrythe encoded
signals, eachsignal.
of bandwidth 240 Hz, are to be transmitted simultaneously by binary
6.2-3 APCM. television signal must
The signals (video beandsampledaudio) has a20%
at least bandwidth
above the of Nyquist
4.5 MHz. This
rate. signal and
Framing is sampled,
synchroniz-
6.2-3 quantized,
A television
ing requires andansignal
binary coded0.5%
(video
additional toand
obtain
audio)
extra abits.
PCM hasAsignal.
a bandwidth
PCM encoderof 4.5 MHz.
is used to convert This these
signalsignals
is sampled,
before
quantized,
they
(a) Determine andthebinary
are time-multiplexed
samplingcoded intoto aif
rate obtain
single a data
the signalPCM signal.
isstream. Determine
to be sampled at a the
rateminimum
20% above possible data rate
the Nyquist
(bitsrate.
Chap 6
per second) that must be transmitted, and the minimum bandwidth required to transmit the
(a) Determine the sampling rate if the signal is to be sampled at a rate 20% above the Nyquist
multiplex signal.
(b) Ifrate.
the samples are quantized into 1024 levels, determine the number of binary pulses required
6.2-5 It (b)istoIf encode
desired
each sample.
the samples
to set areup quantized
a central into station 1024 forlevels, determine
simultaneous the numberofofthe
monitoring binary pulses required
electrocardiograms
(c)
SAMPLING AND(ECGs) Determine
to encode
ANALOG-TO-DIGITAL the
eachbinary pulse
sample.
of 10 hospital patients. rate
CONVERSION (bits per second) of the binary-coded
The data from the 10 patients are brought to a processing signal, and the minimum
center over
bandwidth
(c) Determine
wires required
and are sampled,
the binary to transmit
quantized, this signal.
pulse ratebinary-coded,
(bits per second) and of time-division-multiplexed.
the binary-coded signal, and Thethemultiplexed
minimum
324 SAMPLING AND ANALOG-TO-DIGITAL CONVERSION
databandwidth required to to
are now transmitted transmit this signal.station (Fig. P6.2-5). The ECG signal bandwidth
the monitoring
6.2-4 Five telemetry signals, each of bandwidth 240 Hz, are to be transmitted simultaneously by binary
data are now transmitted to the monitoring station (Fig. P6.2-5). The ECG signal bandwidth
is 100 Hz. The maximum
PCM. The signalsismust 100 Hz.beThe
acceptable
sampled
maximumat
error
least 20%
acceptable
in
error above
sample
in samplethe
amplitudes
Nyquist
amplitudes
isthe0.25%
rate.ofFraming
is 0.25%
of the peak signal
and synchroniz-
peak signal
6.2-4 Five telemetry
amplitude. The signals,
sampling
amplitude. each
The rateof must
samplingbandwidthbe
rate must at 240
leastHz,
least
be at are
twice
twice to beNyquist
the
the Nyquist transmitted
rate. rate.
Determine simultaneously
Determine
the minimum by binary
thebefore
minimum
ing requires an additional 0.5% extra bits. A PCM encoder is used to convert these signals
PCM.
cable The signals must
cable be
bandwidth sampled at
needed to transmitleast
these20%
data. above the Nyquist rate. Framing and synchroniz-
they arebandwidth needed into
time-multiplexed to transmit
a singlethese data data.
stream. Determine the minimum possible data rate
ingP.6.2-5
Figure requires an additional 0.5% extra bits. A PCM encoder is used to convert these signals before
(bits per second) that must be transmitted, and the minimum bandwidth required to transmit the
P.6.2-5 they are time-multiplexed into a single data stream. Determine the minimum possible data rate
multiplex signal.
(bits per second) that must be transmitted, and the minimum bandwidth required to transmit the
6.2-5 Itmultiplex
is desiredsignal.
to set up a central station for simultaneous monitoring of the electrocardiograms
(ECGs) of 10 hospital patients. The data from the 10 patients are brought to a processing center over
6.2-5 wires
It is and
desired to set upquantized,
are sampled, a central binary-coded,
station for simultaneous monitoring of theThe
and time-division-multiplexed. electrocardiograms
multiplexed
(ECGs) of 10 hospital patients. The data from the 10 patients are brought to a processing center over
wires and are sampled, quantized, binary-coded, and time-division-multiplexed. The multiplexed

4
6.2-6 A message signal m(t) is transmitted by binary PCM without compression. If the SQNR is required
to be at least 47 dB, determine the minimum value of L = 2n required, assuming that m(t) is
sinusoidal. Determine the actual SQNR obtained with this minimum L.
 Problems 389
because the signal amplitude takes on the same values every quarter cycle, the power can also be
found byDATA
PRINCIPLES OF DIGITAL averaging the signal energy over a quarter cycle.
TRANSMISSION
(a) Assuming half-width rectangular pulse, sketch the duobinary signal y(t) for the random
binary sequence
RENCES
P.6.2-7
Chap 6 ···
1110001101001010
1. A. Lender, “Duobinary
(b) Determine R0 , RTechnique for High Speed Data Transmission,” IEEE Trans. Commun.
1 , and R2 for this code. Assume (or you may show if you like) that Rn = 0
6.2-6 A Electron.,
message vol. CE-82,
m(t) pp. 214–218,
for all n > 2. Find and sketch theby
signal is transmitted May 1963.
binary
PSD forPCM without
this line code compression. If the SQNR
(assuming half-width is required
pulse). Show
2. A. Lender, “Correlative
to bethat
at least 47 dB, Level Coding
determine for
the Binary-Data
minimum Transmission,”
value of L = 2 n IEEE
required,Spectrum,
assuming vol.
that3, no. 2,is
m(t)
its bandwidth is Rb /2 Hz, half that of bipolar.
pp. 104–115,
sinusoidal. Feb. 1966.
Determine the actual SQNR obtained with this minimum L.
3. P. Bylanski and D. G. W. Ingram,
7.3-1 Data at a rate of 6 kbit/s is to Digital Transmission
be transmitted over a Systems,
leased line Peter Peregrinus4Ltd.,
of bandwidth kHz Hertshire,
by using
6.2-7
6.2-8 England,
Repeat
For a PCM
Nyquist 1976.
Prob. 6.2-6
signal,
criterion for m(t)
determine
pulses. shown in
L if the
Determine theFig. P6.2-7. value
compression
maximum parameter
of the µ = 100factor
roll-off and the minimum
r that SNR
can be used.
4. H. Nyquist,
required“Certain
is 45 dB. Topics
Determinein Telegraph
the outputTransmission
SQNR withTheory,”
this value AIEE
of L.Trans.,
Remember vol. 47,
thatp.L 817,
mustApril
be a
1928.
power
Hint: of
The 2, that
power is,
of a
L = 2 n for signal
periodic a binaryis PCM.
its energy averaged over one cycle. In this case, however,
7.3-2 In a certain telemetry system, there are eight analog measurements, each of bandwidth 2 kHz.
5. E. D.Samples
Sunde, Communication
because the signalsignals
of these amplitude Systems
are takesEngineering
on the sameTechnology,
time-division-multiplexed,values every Wiley,
quarterNew
quantized, York,
cycle,
and the 1969.
power can
binary-coded. Thealso
errorbe
6.6.2-9 ALucky
R. W.in signal
found by band-limited
and H. R.
averaging to
Rudin,
the 1 MHz is sampled
“Generalized
signal energy at a rate
Automatic
over a 50%
quarter
sample amplitudes cannot be greater than 1% of the peak amplitude. higher
Equalization
cycle. than the
for Nyquist
Communicationrate and quantized
Channels,”
into 256
IEEE Int.levels
Commun. by using
Conf., a µ-law
vol. 22, 1966. with µ = 255.
quantizer
(a) Determine L, the number of quantization levels.
P.6.2-7 7. W. F.(a)Trench,
Determine“An theAlgorithm for the Inversion ofratio.
signal-to-quantization-noise Finite Toeplitz Matrices,” J. SIAM, vol. 12,
(b) Sept. 1964. bandwidth BT if Nyquist criterion pulses with roll-off factor r = 0.2
Find the transmission
pp. 515–522,
(b) are
8. A. Lender, The SQNR
used.
Chapter 7,(the
The received
insampling
Digital ratesignal
must quality)
be at least
Communications: found
25%in
Microwave part
above (a) Nyquist
the was unsatisfactory.
Applications, Feher, Ed.,It Prentice-
rate.
K. must be
increased at least by
Hall, Englewood Cliffs, NJ, 1981. 10 dB. Would you be able to obtain the desired SQNR without increasing
7.3-3 A leased telephone line of bandwidth 3 kHz is used to transmit binary data. Calculate the data
rate (in bits per second) that can be transmitted if we use:

BLEMS (a) Polar signal with rectangular half-width pulses.

6.2-8 (b)
For Polar
a PCM signal
signal,with rectangular
determine L iffull-width pulses. parameter µ = 100 and the minimum SNR
the compression
(c)
7.2-1 required
Polarissignal
Consider a45 using
full-width
dB. Nyquist
rectangular
Determine criterion
the pulse pulses
output shape of
SQNR r =this
with 0.25.
value of L. Remember that L must be a
power of 2, that is,with n
L = rectangular
2 for a binary PCM. Problems 389
(d) Bipolar signal half-width5 pulses.
p(t) = !(t/T )
(e) Bipolar signal with rectangular full-width pulses. b
6.2-9 A(a)
signal band-limited
Assuming to 1 MHz
half-width is sampled
rectangular at a rate
pulse, 50%the
sketch higher than thesignal
duobinary Nyquist rate
y(t) forand
thequantized
random
into
7.3-4 The 256 levels by
binary sequence using a µ-law quantizer with µ = 255.
(a) Fourier
Find PSDs transform P(f ) of
for the polar, the basic
on-off, pulse p(t)
and bipolar used in a certain binary communication
signaling.
system is shownthe
(a) Determine in signal-to-quantization-noise
Fig. P7.3-4. ratio.
(b) The
(a) From SQNR
to thethe shape
case (the
where received
of P(f explain
p(t)), is
Chap 7
(b) Sketch roughly the PSDs and find 1110001101001010
signal
their bandwidths. For
quality) found
at what rectangular
a half-width pulse rate thisin
· · each
part
·
(a)
case, compare the bandwidth
was unsatisfactory.
pulse would satisfy
pulse. Nyquist’s It must be
criterion.
(b) increased
Determine R 0
at least, R 1 ,
by 10and R 2 for this code. Assume (or you may show if you like) that Rn = 0
(b) Find p(t)
for all n> and 2.verify
Find and thatdB. Would
this
sketch pulse you
doesbefor
the PSD
able
(orthis
to obtain
does not)
line
the desired
satisfy
code
SQNR without
the Nyquist’s
(assuming
increasing
criterion.
half-width pulse). Show
7.2-2 (a) A random binary data sequence 110100101· · · is transmitted by using a Manchester (split-
(c) Ifthat
theitspulse does satisfy
bandwidth is Rb /2theHz,Nyquist criterion,
half that what is the transmission rate (in bits per
of bipolar.
phase) line code with the pulse p(t) shown in Fig. 7.7a. Sketch the waveform y(t).
second) and what is the roll-off factor?
7.3-1 (b)
DataDerive Sy (fof),6the
at a rate PSDis of
kbit/s to abeManchester
transmitted(split-phase)
over a leasedsignal part (a) assuming
in bandwidth
line of 4 kHz by 1 and 0
using
equally likely. Roughly sketch this PSD and find its bandwidth.
Nyquist criterion pulses. Determine the maximum value of the roll-off factor r that can be used.
7.3-4
7.2-3
7.3-2 IfInthe pulse shape
a certain is system, there are eight analog measurements, each of bandwidth 2 kHz.
telemetry
Samples of these signals are time-division-multiplexed,
! "quantized, and binary-coded. The error
in sample amplitudes cannot be greater than 1% of t peak amplitude.
the
p(t) = !
0.5T
(a) Determine L, the number of quantization levels. b
(b) differential
use Find the transmission bandwidth
code (see Fig. T if Nyquist
7.18) toBderive criterion
the PSD pulses signal.
for a binary with roll-off
Determine r =PSD
factorthe 0.2
are used. The sampling rate must be at least 25% above the Nyquist rate.
Sy (f ).

7.3-3 AThe
7.3-5
7.2-4 Apulse
leased
p(t)telephone
duobinarywhose line line
spectrum of bandwidth
coding P(f ) is shown
proposed 3 kHz
by is used
in Fig.
Lender P7.3-5
is also to transmit binary
satisfieslike
ternary Nyquist’sdata. Calculate
criterion.
bipolar, but If f1the
it requires data
= only
0.8
rate the
MHz
half (in
andbits
f2 =per1.2
bandwidth second)
MHz, that canInbepractice,
determine
of bipolar. transmitted
the maximum if rate
duobinary we use: at whichisbinary
coding data realized
indirectly can be transmitted
by using a
390 PRINCIPLES OF DIGITAL DATA TRANSMISSION
by(a)this
special pulse
pulse
Polar using
shape
signal Nyquist’s
with criterion.
as rectangular
discussed Sec.What
inhalf-width ispulses.
7.3 (see the
Fig.roll-off
7.18).factor?
In this code, a 0 is transmitted by no
Figure P.7.3-5
pulse, and a 1 is transmitted by a pulse p(t) or −p(t) using the following rule. A 1 is encoded
1
(b) Polar signal
7.3-6 Binary with rectangular full-width pulses. using Nyquist criterion pulses with P(f )
by the data
sameatpulse
a rate of that
as 1P(Mbit/s
f ) usedisfor to be transmitted
the previous 1 by if there are an even number of 0s between
0.5
(c)
shown Polar
in signal
Fig. using
P7.3-5, Nyquist
The criterion
frequencies
them. It is encoded by a pulse of opposite f pulses
and f
1 polarity2of r
of =
the 0.25.
spectrum
if there are anare adjustable.
odd number ofThe 0s channel
between
available
them. for
A number
(d) Bipolar the transmission
signal0 iswith
considered of this
0 to be
rectangular data has a bandwidth
an even number.
half-width of 700 kHz.
f1 = 0.8Like1 bipolar,
pulses. Determine f and f and
f2 = 1.2 this code also1 has a 2single
the
errorroll-off factor.
detection capability, because correct 6
reception f, MHz
(e) Bipolar signal with rectangular full-width pulses.implies that between successive pulses of
the same polarity, an even number of 0s must occur, and between successive pulses of opposite
polarity, an odd number of 0s must occur.
 pr (Tb ) = 0.1 pr (−Tb ) = 0.3
pr (Tb ) = 0.1 pr (−Tb ) = 0.3
pr (2T ) = −0.02 pr (−2T ) = −0.07
pr (2Tbb) = −0.02 pr (−2Tbb) = −0.07

(a) Determine the tap settings of a three-tap, zero-forcing equalizer.

(a)
(b)
(b)
Chap 7
Determine the tap settings of a three-tap, zero-forcing equalizer.
Using the equalizer in part (a), find the residual nonzero ISI.
Using the equalizer in part (a), find the residual nonzero ISI.
7.7-1 In a PAM scheme with M = 16:
7.7-1 In a PAM scheme with M = 16:
(a) Determine the minimum transmission bandwidth required to transmit data at a rate of 12,000
(a) Determine the minimum transmission bandwidth required to transmit data at a rate of 12,000
bits/sec with zero
bits/sec with zero ISI.
ISI.
(b)
(b) Determine the transmission
Determine the transmission bandwidth
bandwidth ifif Nyquist
Nyquistcriterion
criterionpulses
pulseswith
witha aroll-off
roll-offfactor
factor
rr = 0.2 are used to transmit data.
= 0.2 are used to transmit data.

7.7-2
7.7-2 An
An audio signal of
audio signal of bandwidth
bandwidth 44 kHz
kHz isis sampled
sampled atataarate
rate25%
25%above
abovethe
theNyquist
Nyquistrate
rateand
and
quantized. The quantization error is not to exceed 0.1% of the signal peak amplitude.
quantized. The quantization error is not to exceed 0.1% of the signal peak amplitude. The The
resulting quantized samples
resulting quantized samples are
are now
nowcoded
codedand andtransmitted
transmittedbyby4-ary
4-arypulses.
pulses.
(a)
(a) Determine the minimum
Determine the minimum number
numberof of4-ary
4-arypulses
pulsesrequired
requiredtotoencode
encodeeach
eachsample.
sample.
(b)
(b) Determine
PRINCIPLES OF DIGITAL the minimum
minimum transmission
DATA TRANSMISSION
Determine the transmission bandwidth
bandwidthrequired
requiredtototransmit
transmitthis
thisdata
datawith
withzero
zero
ISI.
ISI.
(c) If 4-ary pulses satisfying Nyquist’s criterion with 25% roll-off are used to transmit this data,
determine the transmission bandwidth.

7.7-3 Binary data is transmitted over a certain channel at a rate Rb bit/s. To reduce the transmission
bandwidth, it is decided to use 16-ary PAM signaling to transmit this data.
(a) By what factor is the bandwidth reduced?
(b) By what factor is the transmitted power7increased, assuming minimum separation between
pulse amplitudes to be the same in both cases?
Hint: Take the pulse amplitudes to be ±A/2, ±3A/2, ±5A/2, ±7A/2, · · · , ±15A/2 so that
the minimum separation between various amplitude levels is A (same as that in the binary
case pulses ±A/2). Assume that all 16 levels are equally likely. Recall also that multiplying
a pulse by a constant k increases its energy k 2 -fold.

7.7-4 An audio signal of bandwidth 10 kHz is sampled at a rate of 24 kHz, quantized into 256 levels
and coded by means of M-ary PAM pulses satisfying Nyquist’s criterion with a roll-off factor
r = 0.2. A 30 kHz bandwidth is available to transmit the data. Determine the best value of M .

7.7-5 Consider a case of binary transmission via polar signaling that uses half-width rectangular pulses
of amplitudes A/2 and −A/2. The data rate is Rb bit/s.
(a) What is the minimum transmission bandwidth and the transmitted power.
(b) This data is to be transmitted by M-ary rectangular half-width pulses of amplitudes

±A/2, ±3A/2, ±5A/2, . . . , ±[(M − 1)/2]A

Note that to maintain about the same noise immunity, the minimum pulse amplitude sepa-
ration is A. If each of the M-ary pulses is equally likely to occur, show that the transmitted
power is

(M 2 − 1)A2
P=
24 log2 M
Also determine the transmission bandwidth.

7.8-1 Figure P7.8-1 shows a binary data transmission scheme. The baseband signal generator uses
full-width pulses and polar signaling. The data rate is 1 Mbit/s.
(a) If the modulator generates a PSK signal, what is the bandwidth of the modulated output?
(b) If the modulator generates FSK with the difference f −f = 100 kHz (see Fig. 7.32c),