Dlgltal communication Systems

(a) a cohoront earner I'.:. not req 1 ruo

~ Companding In PCM systems lead to improved generated at the receI10, l"l ~

J(7 signal to quantiza tion noise ratio . This (b) tor a given energy per bit, the PrOi.Jat;
improvement is for error is less , O!
{a) lower frequency components only (c) the 180° pha se shift s of the carr,(;r .,
unimportant d '=
(b) higher frequency components only
{c) lower amplitudes only (d) more protection is provided against 1mPV,serr,,
{d) higher amplitudes only [1 989 · 2l , 'c
• IV]ar1r.s1
(1987 : 2 Marks]
modulated by a
~A
3.6 A 4 GHz carrier is DSB-SC ·
I w1 th maximum lc,i
. frequ
signal having uniformly distributed amplitud e pass message sIgna ~~
.
In the interval (- Vto + V), is to be encoded using of 2 MHz. The resu Itant signal is to be idea11
. . 1
PCM with uniform quantization . The signal to sampled . The minimum frequency of the sampling
quantizing noise ratio is determined by the impulse train should be
(a) 4 MHz (b) 8 MHz
(a) dynamic range of the signal
(c) 8 GHz (d) 8.004 GHz
(b) sampling rate
(c) number of quantizing levels
[1990 : 2 Marks]
(d) power spectrum of signal 3.7 In a BPSK signal detector, the local oscillator has
[1988: 2 Marks] a fixed phase error of 20°. This phase error
deteriorates the SNR at the output by a factor of
3.3 The message bit sequence to a DPSK modulator
(a) cos 20° (b) cos 2 20°
is 1, 1, 0, 0, 1, 1. The carrier phase during the
(c) cos 70° (d) cos 2 70°
reception of the first two message bits is 1t, 1t. The
· [1990 : 2 Marks]
carrier phase for the remaining four message bits
is 3.8 A signal has frequency components from 300 Hz
(a) 1t, 1t, O, 1t (b) 0, 0, 1t, 1t · to 1.8 kHz. The minimum possible rate at which
(c) 0, 1t, 1t, 1t (d) 1t, 1t, 0, 0 the signal has to be sampled is
[1988 : 2 Marks] [1991 : 2 Marks]

3.4 In a digital communication system, transmissions 3.9 For the signal constellation shown in the figure,
of success ive bits through a noisy channel are the type of modulation is
assumed to be independent events with error cos21t(n!T)t

probability p. The probability of at most one error

in the transmission of an 8-bit sequence is --t- -- sin2n:(n/T)t
T = symbol duration
71
(a) 7(1- p)/8 ( - p)
• S4
B+p/8
(b) (1 - p)B + 8p(1 - p)7
~ [1991 : 2 Marks]

(C) (1 - p) 8 + (1 - p)7 ~ ~ signal has frea.u, ncy components from 300.Hz

(d) (1 - p) 8 + p(1 - p)7 to 1.8 kHz m possible rate at whi~h
[1988: 2 Marks] the signal led is _ _ (fill in
3.5 fn binary data transmission DPSK is preferred to the blank)
PSK because
 ,,,unication Systems
colf1
.tJ I 01
001 is differentially encod
f/ 0arT1
~it str and Ex
OR' h
sc eme for DPSK
ed 3,18 A 1.0kHz s ~ --- S6S
J 1800 ignal ,s flat to ~
ira ,peltlY ,Assurning the reference bit as , , samples/se P ,ampled <11 L'lo r,jtr of
,i~i~ g p~ases of 'O' and 'n' for 1's an:0~ 8
to a · c
n ideal rectang I
of 11 00 Hz th
and th
e samples arlj &pp!•"'1
u ar LPF w,th cut-off freo:.senc-
-
rf s#ign1 in the encoded sequence, the
, en the out . , I
I 8
8
cIiv \ase sequence becomes (a) only 800 H put of the filter coma ns
; ~i«ed p (b} 01t 1t O0 (b) 800 H z component
Zand 900 H
ro""o
i O (d} 1t1t01t1t
.
(c) 800 H
Zand 1000H
z components
,1 o""" [1992 : 2 Marks] (d) 800 H z components

~·
z, 900 Hz and 100 Hz components
~
rriodulation of FSK signal can b
~ de e d .Th [1 995 : 1 Mark]
....hBre . g ~ esignal to u • . noise .
of ,iy· ted usin . PCM q antization ratio in an n-b1t
.;.ec 1ation receiver system
~cJffB filters and envelope detectors
I ndP855 (a) depends upon the sampli ng frequency
1D) b3 tched filter . employed
(cl ~a ·rninator detection (b) . 1 .
uJ\, d1SCfl [ 1992 :_2 Marks] '. s nd ependent of the value of ·n'
(
(c) increasing with increasing value of 'n'
coding in a data communication system (d) decreases with the increasing value of ·n'
cy ~rceen [1995 : 1 Mark]
ly ,J. dOne in order to .
g ~ enhance the information transmission
I~b ndpass filters and envelope rate detectors °
3·2 For a given data rate, the bandwidth B of a BPSK
. p
signal and the bandwidth B0 of the OOK signal
~) \serve the transmitted power
(c) G0 . are related as
(d) discriminator detection [ 1992 : 2 Marks]
(a) BP= B0/4 (b) BP= B0/2
11 ~ tchthewaveform (with properly marked axes) at (c) BP= B0 (d) BP= 2B0
1s ~ Pl o.Jtput of a matched filter matched for a signal [1995: 1 Mark]
Dr
,f ~0,of duration T, given by 3 .21 A rectangular pulse of duration Tis applied to a filter
A for O ~ t < ~ T matched to this input. The output of the filter is a
~t)= 3 (a) rectangular pulse of duration T
] 2
[ O tor T 5 t< T (b) rectangular pulse of duration 2T
3 (c) triangular pulse
[1993 : 2 Marks]
(d) sine function [1996 : 1 Mark]
dth in the flat top sampling, c:)he number of bits in a binary PCM system is
leads to ~ncreased from n ton+ 1. As a result, the signal
(a) attenuation of high frequencies in reproduction to quantization noise ratio will improve by a factor
(b) attenuation of low frequencies in reproduction
(a) (n + 1)/n
(c) greater aliasing errors in reproduction
(b) 2(n + 1)/n
(d) no harmful effects in reproguction .
. [1994: 1 Mark] (c) 22(n + 1)/n
(d) which is independent of n
[1996: 2 Marks]
~bandwidth required for the transmission of a PCM
Signal increases by a factor of when the number . ode that has zero de component for pulse

~
of quan11 -
.on levels is increased
·zat1 from 4 to 64.
[1994 : 1 Mark] @ e I,nee . .
transmission of random binary data is
(a}, non-return to zero (NAZ)
t
is~enumber of bits per sample in a PCM syS em (b) return to zero (RZ) .
· improvement ·in
s·1ncre ased from a n to n + 1 the (c) alternate mark inversion (AMI )
19na1t . '
(a) d~ quantization nose ratio will be of the above [1997 : 1 Mark]
3 (d) none
~ ~) 6dB
2nd8 (d) n dB
fj
995 : 1 Mark]
 566 I GATI P,evlou• v... ,. Solv•rl hp•u t Ill
GA doto1ml111&11c • gnol h• !ho power
g,von 1n l1uuro lhc mlrnmurn irnolmo roto
'""" ;r:;;;)
C7
needed 10 completely rcp,orcnt signal Is
S(n (0
(C) 95 o
( l{!~O 2 II r
'-I
3.31 The input 10 a matche I f tor • by

(j
st)=
0
10sin(2n x icfl) 0 < I< 10 4
0tf'llrtl\SO
r.oc

- ~ -+-lof-l...--L-_~ ~ ~- - ! (kHz)
-2 - 1.5 -1 0 1 15 2 The peak amplitude of the f,lter 011tpu1 1..,
(a) 1 kHz (b) 2 kHz (a) 10 vol ts (b) 5 volts
(c) 3 kHz (d) none of these (c) 10 m,lhvolts (d) 5 m1ll1volt..,
[1997 : 1 Mark] [1 999 : 2 Mar~sj

~ !n a P?M system with uniform quantization, Q Four independent messages have bandwid'hs 01
Q IncreasIng the number of bits from 8 to 9 will reduce Q 100 Hz, 100 Hz, 200 Hz, and 400 Hz, respecbely
the quantization noise power by a factor of Each is sampled at th e Nyquist rate, and tne
(a) 9 (b) 8 samples are Time Division Multiplexed (TOM) and
(c) 4 (d) 2 transmitted. The transmitted sample rate (1n Hz) ,s
[1998: 1 Mark] (a) 1600 (b) 800
(c) 400 (d) 200
3.26 Flat top sampling of low pass signals
(a) gives rise to aperture effect [1999 : 2 Marks]
(b) implies oversampling 3.33 In a digital communication system employing
(c) leads to aliasing Frequency Shift Keying (FSK), the Oand 1 bit are
(d) introducing delay distortion represented by sine waves of 10 kHz and 25 kHz
[1998 : 1 Mark] respectively. These waveforms wi ll be orthogonal

@ Quadrature multiplexing is
(~ ~ FDM
for a bit inteNal of
(a) 45 µsec (b) 200 µsec
(c) 50 µ sec (d) 250 µ sec
(6f ea~TDM
[2000 : 2 Marks]
(c) a combination of FDM and TOM
(d) quite different from FDM and TOM 3.34 A video transmission system tr ansm its 625
[1998 : 1 Mark] picture frames per second . Each frame consists
of a 400 x 400 pixel grid with 64 intensity levels
Q Compression in PCM refers to relative compression
Va, per pixel. The data rate of the system is
(a) 16 Mbps (b) 100 Mbps
(a) higher signal amplitudes
(c) 600 Mbps (d) 6.4 Gbps
(b) lower signal amplitudes
[2001 : 2 Marks]
(c) lower signal frequencies
(d) higher signal frequencies ~ The Nyquist sampling interval , for the signal
[1998: 1 Mark] sinc(700t) + sin c(500t) is
r : : \ e Nyquist sampling frequency (in Hz) of a signal 1 n
(a) sec (b) sec
4
'01ven by 6 x 10 sin c3 (400 f)*106 si11 c3 (100 t) is 350 350
(a) 200 (b) 300 \ n
(d) -sec
(c) 1500 (d) 1000 175
[1999 : 2 Marks] [2001 : 2 Marks]
 7
I communication Systems I S67
.•,P~f}S'I
,,.,
ounn
::m~~~~~~===------ ------
. . over a communicar
. g transmission
. 10n (a) 0.768V (b) 48 X 10 > l7
.96 channel, bit errors occur ind_ep~ndently with (C) 12 X 106 \J2 (d) 3 072 V
3 babilitY p. If a block of n bits 1s transmitted [2003 : 2 Marks)
pro robabilitY of at most one bit error is equal t '
the P n o 3.42 If Eb, the energy per bit of a binary d1g1tal s1g na'
(a) 1 - (1 - p) is 1o-5 watt-sec and the one-sided power spectral
(b)P+(n-1)(1-P) 6
density of the white noise, N0 = 10- W/HZ. then
c) np(1-Pt-1
the output SNR of the matched filter 1s
~d) (1 - p)n + np(1 - pt-1
(a) 26 dB (b) 10 dB
[2001 : 2 Marks]
(c) 20 dB (d) 13 dB
[2003 : 2 Marks]
consider a sample signal
3,37 6
y (t) == 5 X 1o- x (t) L n=- B(t - nT
,t<>O
5) 3.43 The input to a linear delta modulator having a step-
size !!,. = 0.628 is a sine wave w ith frequency fm
3
where x(t) == 1ocos(81t x 10 )t and T5= 100µsec. and peak amplitude Em. If the sampling frequency
When y ( t) is passed through an ideal lowpass f = 40 kHz, the combination of the sine-wave
5
filter with a c utoff frequency of 5 kHz, the output frequency and the peak amplitude, where slope
of the filter is overload will take place is
3
(a) 5 x 10-6cos(81t x 10 )t Em fm
3
(b) 5 x 10- 5cos (81t x 10 )t (a) 0.3 V 8 kHz
3
. (c) 5 x 10- 1cos(81t x 10 )t (b) 1.5 V 4 kHz
(d) 1ocos (81t x 103)t [2002 : 1 Mark] (c) 1.5 V 2 kHz
(d) 3.0 V 1 kHz
3.38 For a bit-rate of 8 Kbps , the best possible values [2003 : 2 Marks]
of the transmitted frequencies in a coherent binary
3.44 If S represents the carrier synchronization at the
FSK system are
receiver and p represents the bandw idth
(a) 16 kHz and 20 kHz
efficiency, then the correct statement for the
(b) 20 kHz and 32 kHz
coherent binary PSK is
(c) 20 kHz and 40 kHz
(d) 32 kHz; and 40 kHz [2002 : 1 Mark] (a) p = 0.5, S is required
(b) p = 1.0, S is required
3
3.39 A signal x(t) = 100 cos(24n x 10 )t is ideally (c) p = 0.5, Sis not required
sampled with a sampling period of 50 µsec and (d) p = 1.0, Sis not required
then passed through an ideal lowpass filter with [2003 : 2 Marks]
cutoff frequency of 15 kHz. Which of the following
3.45 A signal is sampled at 8 kHz and is quantized
frequencies is/are present at the fi lter output?
using 8-bit uniform quantizer. Assuming SNR q for
(a) 12 kHz only a sinusoidal signal, the correct statement for PCM
(b) 8 kHz only signal with a bit rate of R is
(c) 12 kHz and 9 kHz (a) R = 32 kbps, SNRq = 25.8 dB
(d) 12 kH z and 8 kHz [2002 : 2 Marks]
(b) R = 64 kbps, SNRq = 49.8 dB
3.40 At a given probability of error, binary coherent FSK (c) R = 64 kbps, SNRq = 55.8 dB
is inferior to binary coherent PSK by. (d) R = 32 kbps, SNRq = 49.8 dB
[2003 : 2 Marks]
(a) 6 dB (b) 3 dB
(c) 2 dB (d) 0 dB 3 .46 In a PCM system, if the code word length is
[ 2003 : 1 Mark] increased from 6 to 8 bits , the signal to
quantization noise ratio improves by the factor
3.41 A sinusoidal
· · signal with peak-to-peak amp 1·tI ud e
(a} 8/6 (b) 12
of 1.536 V is quantized into 128 levels using a
(c) 16 (d) 8
mid-rise uniform quantizer. The quantization-noise
[2004 : 1 Mark]
power 1s
 561 I

3 .47 In IIID OUl~)Ut of a OM p och cnCOdcr Ill Or0t1f) I

c .)OSO<..--t.Jtivo pu a,o ol Of'.)f)OSfto f)Ol,11 ty during I n. • p

111110 mtorval 11 s Is t Th s loct,caw that during 2 Of.' 0
this 1111orval J PSK R
(a) tho mpul to the modul11tor is t!GSonuall;•
4 PQ.~ s
T.
constant
(b) the modulator ,s going through slof}e overload
u.
(c) lhe accumulator 1s In salurahon (a) 1-T, 2-P. 3-U, .: .5
(d) the speech 51gnal ,s berng sampled al the (b) 1-S 2-U 3-P, 4•T
Nyquist rate (c) 1-S, 2-P 3-U, t .Q
[2004 : 1 Mark] (d) 1-U, 2-A. 3-S 4-0
3.48 A source produces binary data at the rate of 3.51 Three ana)Og signals, having bandw,o:ns l20J11.
10 kbps. The binary symbols are represented as
600 Hz and 600 Hz a•e samp ed al :.he: r~,ie
shown in the figure.
Nyquist rates. encoded ••,1th 12 b; ,','O<ds anci
Binary 1 Binary 0
time d1vis1on multiplexed The b t ra·e f0< e
1v i - - - - - ,
multiplexed signal 1s
0 0.1 (a) 115.2 kbps (b) 28 8 ;.:bps
0 0.1 t (msec) t (msec)
-1 V ....__ __, (c) 57.6 kpbs (d) 38 4 kbps
[ 2004 : 2 Mar'r(s j
The source output is transmitted using two
3.52 Consider the signal x(t) shown 1n the figure Le•
modulation schemes, namely Binary PSK (BPSK)
h( t) denote the impulse response of tne f :er
and Quadrature PSK (QPSK). Let 8 1 and 82 be the
matched to x(t) , with h(t) being non-zero on~, ~
bandwidth requirements of BPSK and QPSK
the interval O to 4 sec. The slope of h(t) ,n ,t:e
respectively. Assuming that the bandwidth of the
interval 3 < t < 4 sec is
above rectangular pulses is 10 kHz, 8 1 and 8 2 are
(a) 8 1 = 20 kHz, 8 2 =20 kHz
x(t)
(b) 8 1 = 10 kHz, 82 = 20 kHz
(c) 8 1 = 20 kHz, 82 = 10 kHz
(d) 81 = 10 kHz, 82 = 10 kHz
[2004 : 2 Marks]

3.49 Consider a binary g~ munication system

with equally likely O's and 1's. When binary O is
transmitted the detector input can lie between the
levels -0.25V and +0:25V with equal probability: 1 -1
(a) -sec (b) -1 sec- 1
when binary 1 is transmitted, the voltage at the 2
detector can have any value between O and 1 V 1 -1
(c) --sec (d) 1 sec- 1
with equal probability. If the detector has a 2
threshold of 0.2 V (i. e., if the received signal is [2004 : 2 Marks]
greater than 0.2 V, the bit is taken as 1), the
3.53 A signal as shown in the figure is applied to a
average bit error probability is
matched filter. Wh ich of the following does
(a) 0.15 (b) 0.2
represent the output of this matched filter?
(c) 0.05 (d) 0.5
input
[2004 : 2 Marks]
1
3. 50 Choose the correct one from among the alternatives
a, b, c, dafter matching an item from Group 1 with
the most appropriate item in Group 2. -1
 tton System•

r..,,
Vftth
... ...
............
O ■ illaQ_.L

, ., ., ,,
p
,.,..t-v.

range l~• {] C

{a)~ (t>) o/2

(c) ti/12 (cf) v
(2006 2 Matks )

Common Data for Questions 3 57 and 3 58

Let g(t) = p(t)•p(t) wn.Jre • denotes convolut on ilnd
P<.t) = u(t)- u(t -1) with u(() betng the un:: step runct1cn

3.57 The impulse response of filter m.1tched to the 51gnal

s(t)=g(t)-o<t -2)•g<t) 1 (llV('nclr;

(cl (a) s( 1 - t) (b) s( 1 f)

(c) - s(t) (d) s( t)
[2006 2 Marks]
()Ulplll
3.58 An Amplitude Modulated signal 1s given as
xl'Jit) = 100(p(t) + OSg(t))COSt•>d 1n the
interval O~ t $ 1. One set of possible values of the
modulating signal and modulation index would be
(a) t, 0.5 (b) t, 1.0
2
[2005 : 2 Marks] (c) t, 2.0 (d) t , 0 5
[2006 : 2 Marks]
3.54 The minimum sampling frequency (in samples/sec)
3 .59 In delta modulation, the slope overload d1stort1on
required to reconstruct the following signal from
can be reduced by
its samples without distortion
(a) decreasing the step size
,(t).s(sin2::ooat)' + 7(sin2::ooat)' (b) decreasing the granular noise
(c) decreasing the sampling rate
would be (d} increasing the step size
(a) 2x 103 (b) 4 x 103 [2007 : 2 Marks]
3
(c) 6x 103 (d) 8 x 10
[2006 : 2 Marks] 3.60 The raised cosine pulse p(t) is used for zero ISi in
digital communications. The expression for p(t)
3,55 The minimum
. . step-size required for a with unity roll -off factor is given by
D~~~di ·
u ator operating at 32 K samples/ sec
1o tr~ck the signal (here u(t) is the unit-step sin41t Wt
p(t} = ( 2 2
) . T he value of p(t)
funct1on) 4nWt 1-1 6W t
i(t) ~125t
so th(u(t) _ u(t - 1)) + (250 - 125t)(u(t-1) - u(t - 2))
(a) 2-at slope-overload .1s avoided,
. 1 .
10
would be at t= 4°W IS
(c) 2-6 (b) 2...s
3.5 (d) ~ [2006 : 2 Marks] (a) -0.5
(b) 0
6 1nth (d) oo
e foliowing
· figure
. (c) 0.5
Const the minimum value of the [2007: 2 Marks}
ant
that y (t) "C" .
• whic h is to be added to y1(t) such
1 and Y2(t) are different, is

l
 s10 1 GATE Pr• vlou • YurJ Solved Papers : [Ij I
hla.r,1 l
3 ,eb 11 1110 ,1 con .. wll 1,on or J• , • ~,
3.61 Du ring tran smI ss1011 ovu r n cort i,tn tmrnry 1
comrr11 1mo ,11on" er,, , an Alt(. 1 ct
communIcatIon ch c11111ol . b1t orr or~ oc c 111
w11ld of ma foll<TIMIO 1.rn, mo ,
indopendonlly with probability p Tho probilh1hty IJ
(n) Proh,it>1lil 'f ol 1,; trrihol errOf IOf C<ir,
of AT MOST ono bit 1n orror in a block of n b1I!> I~,
1<, IO t✓O(
given by
(b) Probt1b1ht / of ', Jrr t.l{}I ow,, for Cor
(a) P'1
Is h1gt1or
(b) 1 - P'1
(c) Probabtlil/ ol sJrnt.,01 e•rc,r !JO
(c) np(1 - pr - 1+ (1 - p)n
the constella tions •ua ~ t,
(d) 1 - (1 - p)n [2007 : 2 Marks)
(d) The value of N0 111II determ ne itt ',h,~
3.62 In a GSM system , 8 channels can co-exist in constellations has a lo'ller pror.,:., ' ie ,
~ .., 1/'A
200 kHz bandwidth using TOMA. A GSM based error ·1 tr
cellular operator is allocated 5 MHz bandwidth.
[2007 : 2 ►1 afh
Assuming a frequency reuse factor of 1/5, i.e. a
five-cell repeat pattern, the maximum number of Statement for Linked Answer Questions 3 c:,, ••
simultaneous channels that can exist in one cell is
. ·VO <lI\(j'l
ant1zer has the probao
An input to a 6-level qu_ . •
1
'Jr.~?
(a) 200 (b) 40 function f(x) as shown ,n the figure. Decis,on bo "~- · ,
(c) 25 (d) 5 of th e quan tizer are chosen so as to rr·a/ ," .-,:• '=.
. r zi:: ·, r

[2007 : 2 Marks) entropy of the quantizer output. It is gi•,er ,, •.

consecutive decision boundaries are '-1' o and ; ~
3.63 In a Direct Sequence CDMA system the chip rate f (x)
is 1.2288 x 1rfa chips per second. If the processing
gain is desired to be AT LEAST 100, the data rate
(a) must be less than or equal to 12.288 x 103
bits per sec - ----- - a
(b) must be greater than 12.288 x 103 bits per sec
(c) must be exactly equal to 12.288 x 103 bits t------- b
per sec X
(d) can take any value less than 122.88 x 103 bits -5 -1 0 1 5
persec 3.66 The values of a and bare
[2007 : 2 Marks] (a) a= 1/6 and b = 1/12
(b) a= 1/5 and b = 3/40
Common Data for Questions 3.64 & 3.65: (c) a= 1/4 and b = 1/16
Two 4-ary signal constellations are shown. It is given
(d) a= 1/3 and b = 1/24 [2007 : 2 Marks]
that q,1 and q,2 constitute an orthonormal basis for the two
constellations. Assume that the four symbols in both 3.67 Assuming th at the reconstruction levels of ti's
the constellations are equiprobable. Let Nof2 denote the quantizer are the mid-points of the decis.o·
traJ d 'ty f h't Gaussian noise. boundaries , the rat io of signal power w
+i quantization noise power is
B 152 64
(a) - (b) -
, ' 1/' "\, 9 3
,'
---<I----+---¼--+-
'\ ,'
«1>,----4-____.,._
',\
__ «I>,
-2✓2;',,,,:!!~// o -a',,,,, ~~/a (c) 76 (d) 28
3
[2007 : 2 Marks]
'•' -✓2a
Constellation 1 Constellation 2 3.68 Consider a Binary Symmetric Channel (BSC)i'l"Ji
1
probability of error being p. To transmit a bit, say ·
3.64 The ratio of the average energy of Constellation 1
we transmit a sequence of three 1s. The receiVer
to the average energy of Constellation 2 is
(a) 4a 2 (b) 4
will interpret the received sequence to represent
(c) 2 (d) 8 1 if at least two bits are 1. The probability that[fie
[2007 : 2 Marks] · ed bit will be received in error is
 rnun1ca11u11 ~J>lefflS
I corn
(b) p'3
t ~t1. . srl( 1-P)
(d) p3 + ,o2(1 - p) (a) 46 <18
~ wrl [2008 : 2 Marks1 (c) 42 <18 (b) •iJ8d8
(<1},: I ·
p 85 band limited .to W, w, 2 w and
~ I Y are .to be multiplexed using Time 3.75 Cons,d (2009 . 2 M,.Ht.si
er the Plllse stm
Multiplexing (TDM?. :he minimum irnpufse resp0 pe s(t) f.'.l Sha',•,n Tho
required for transm1ss1on of this TDM Pulse is nse h(t) 0 ' the f1 ter mmchec to th s

S(I)
(b) 3 W
(d) 7 W
1 ------ --- ---·--
[2008 : 2 Marks]
tor Questions 3.70, 3.71 and 3 72
oats · .
A' . 1band limited to 4 kHz and peak voltag
,·~n~gna, . e
~ n+5 vand -5 V, 1s sampled at the Nyquist
I kt,!twee . d d
~ ""sarr,ple is quantize an represented by 8 bits.
eacn h(/)
I bits o and 1 are transmitted using bipolar
i70 If tne 5 the minimum bandwidth required for
pulse , . . .
ion free transm1ss1on 1s (a)
diSIO rt
(a) 64 kHZ (b) 32 kHz
(c) 8 kHz (d) 4 kHz
[2008 : 2 Marks]
-T 0
Assuming the signal to be uniformly distributed
1 h(I)
s.1 between its peak to peak value, the signal to noise
ratio at the quantizer output is 1 ----- -- --- - ---- - -- -
(a) 16 dB (b) 32 dB
(c) 48 dB (d) 64 dB (b)
[2008 : 2 Marks]
~quantization levels required to reduce
tion noise by a factor of 4 would be
(b) 512 h(~

~64
[2008 : 2 Marks]

Common data for Questions 3.73 and 3.74 (c)

iheamplitude of a random signal is uniformly distributed
between - 5 V and 5 V.
3·73 11 the signal to T
quantization noise ratio required in 0
uniformly quantizing the signal is 43.5 dB, the step h(/)
size of the quantization is approximately
(a) 0.0333 V (b) O.O5 V
(c) 0.0667 V (d) O.10 V
[2oo9 : 2 Marks] (d)
3·74 It th · IY
e Positive values of the signal are uniform
nd th e
quantized with a step size of 0.05 V, a 0
T 2T
negative values are uniformly quantized with a s~ep [2010 : 1 Mark]
size of o 1 . antizat1on
. · V, the resulting signal to qu
.s approximately
noise rat·10 1
 S72 I GATE Previous vurs So lved Papers : en I
,.,~O(
t•.
3. 76 The Nyquist sampling r t1t o tor th e n1qnnl 3 81 A t,111t1r/ 6Yfrlrnc1rlc d 1rir •I

protw1>1111y of 1/8 II tho t

s(I) = - · -
s1n(SOOnt) sin(700nr) b ,. ,11cl1 th111 f'(X • 0) • 0
· X IS QIVOl1 Y
nl nf
ol orror for'"' optunum 1nco1
(a) 400 Hz (b) 600 Hz (h) r,·,
(a) 7/80
(c) 1200Hz (d) 1400Hz (c) 9/ 10 (< I) 1/10
[2010 : 2 Marks]
[2012 2 I -
3. 77 An analog signal is band-limited to 4 kHz , ; 1'f
3 .82 ABPSK schemeoperatinqo,er ml, !/(, let
sampled at the Nyquist rate and the samples are with noise power spectral den~11 1 01 ft ,
quantized into 4 levels. The quantized levels are equ,probable signals 'd<
assumed to be independent and equally probable.
If we transmit two quantized samples per second, s1(t) = {2E sin(wcl) and s2 (/) "' f2E.,
\ T , fc,
the information rate is VT
(a) 1 biVsec (b) 2 bits/sec over the symbol internal (0, T). If the locai
. ~...c -•,
(c) 3 bits/sec (d) 4 bits/sec in a coherent receiver _is ah~ad in Phase t -;;
(2011 : 1 Mark] with respect to the received signal , the Pro~ ~
Linked Data Questions 3. 78 & 3. 79.
of error in the resulting system is °' "',
A four-phase and an eight-phase signal constellation are
shown in the figure below.
(a) a(}ff) (b) a(~}
Q
(c) a(J;f) (d) a(J¾,J
_, ,.
\ I
'' ... __ ..,. ,, I ,' [2012 : 2 Marks]
,,
I
' , I

'•, ...
-- _ 3.83 The bit rate of digital communication system 18
R kbits/s . The modulation used is 32-0AM. The
3. 78 For the constraint that the minimum distance minimum bandwid th required fo r IS i free
between pairs of signal points be d for both transmission is
constellations, the radii r1 and r2 of the circles are (a) R/10 Hz (b) R/10 kHz
(a) r1 = 0.707 d, r2 = 2.782 d (c) R/5 Hz (d) R/5 kHz
(b) r1 = 0.707 d, r2 = 1.932 d [2013 : 1 Mark]
(c) r1 = 0.707 d, r2 = 1.545 d
(d) r1 = 0.707d, r2 = 1.307 d
[2011 : 2 Marks]
3.84 Let gJ) = e-nt2, and h( t) is a filter matched to
g(t). If g(t) is applied qS input to h(t) , then the
3. 79 Assuming high SNR and that all signals are equally
Fourier transform of the output is
probable, the additional average transmitted
signal energy required by the 8-PSK signal to (a) e-nr2 (b) e-nr212
achieve the same error probability as the 4-PSK
(c) e-nl~ (d) e-2nr2
signal is
(a) 11 .90 dB (b) 8.73dB [2013 : 1 Mark]
(c) 6.79dB (d) 5.33 dB
Common Data For Questions 3.85 and 3.86:
[2011 : 2 Marks]
Bits 1 and 0 are transmitted with equal probability. At
3.80 In a baseband communications link, frequencies the receiver, the pdf of the respective received signals
upto 3500 Hz are used for signaling . Using a for both bits are as shown below:
raised cosine pulse with 75% excess bandwidth pdf of received
and for no inter-symbol interference, the maximum 1 signal for bit 0
possible signaling rate in symbols per second is
(a) 1750 (b) 2625
)
(c) 4000 (d) 5250
[2012 : 1 Mark] -1 0 2
 ..-.-....-a:attons,....
18 1, the BER WU be Uif
3 Bl 1,
1
(b) 4
1
(d) 16
1
a 90
pi (2013 : 2 Marks)

~ threshold to achieve minimum b t

, ::,. rate (BER) is
4
1 (b) 5
(1) 2
3
(d) -
(C) 1
2
(2013 : 2 Marks]

Lei a[v) be the BER of a BPSK system over an M

,J7 AWGN channel w ith two-sided noise power [20 14 • 2 Mnrks. Sot-4]
spectral density N0 /2. The parameter y is a functoo 3 92 ~ srusooa1 sigiai a 2 kHz trequeocy
d bit energy and noise power spectral density. a cter.a modulator Tho sampling • • ard stcr>Sl.iie
Asystem with two independent and identical ~ a
ltie delta rrod 1.,,.'°' are 20oco samp
AWGN channels with noise power spectral densrty second and O 1 v, resPt,""Ctively To prover t po
O'ler'.oad, lhe maxm.:m ampl tude cA the ... nusoldal
NJ2 is shown in the figure. The BPSK demodulator Slgnal (;n 1/olts) IS
receives the sum of outputs of both the channels
1 1
(a) - {o) -
AWGN 21t 1t
Channlll-1
2
0/1 BPSK IT\ -~ (c) - (d) r.
~ - 1t
ModlU!or
[2015: 1 Mark, Set-1]
AWGN
Channel-2
3 .93 The input Xro 81riary Symme~,c Cnanne (SSC)
sho.vn 1n tne figure s 1 'Nltn prooaotl,ry of O a
If the BER of this system is q bji), then the
The cross-over probab1hry is 1/7. If t.'1e receI ,Ed
value of bis - - bit Y = 0. me cona,t,onal probao1hr; mat 1' was
[2014 : 2 Marks, Set-1] transmmea 1s _ _ __
3-88 Coh · ·
erent orthogonal binary FSK modulation 1s
used to transmit two equiproba ble symbol P[X= OJ= 02
waveforms s1(t) = a.cos2nf t & s (t) = acos2nfi,
Wher 1 2 .h
e ex== 4 mV. Assume an AWGN channel wit
· P[X=1] = 08
N -si·d ed noise power spectral density
two
0 6.7
2 == 0.5 x 10- 12 W/ Hz. Using an optimal receiver [2015 : 2 Marks, Set-1 }
2
. The transmitted signal In a GSM system s of
= _ 1_ -Je
and th _!:!- 3 94
e relation 0( v) 2 du, the bit error · 00 kHz bandwidth and 8 users share a common
2
Probabl1 J2ii. v bandwidth using TOMA If at a g,ven time 12 users
ity for a data rate of 500 kbps is
are talking in a cell, the total bandwidth of the
(a) 0:2)
(b) 0(2✓2) signal received by the base station of the cell wtll
(c) 0:4) be at least (1n kHz) - - - -·
2)
(d) 0(4 ✓ (2015 : 2 Marks, Set- 1)
[2014 : 2 Marks, Set- 21
 en
s14 l
-------------- -
GATE p,.vlous YHrs Solved Papen :
- - -
3.98 The modulatton scheme co~n ~! ,
I
""'~
1ransm1ss on from GSM mob e 1 ' -~
3.95 A source emits bit O w11h probability 23 and bit 1 (a) 4-0AM ~
(b) 16-PSK
2 (c) Walsh-Hadamard orthogona
with probability
3. The emitted bits are
(d) Gaussian Minimum Shift Key ~
"9 G1,,.
communicated to the receiver. The receiver [2015 · 1 •;: : -
. Man~
decides for either 0 or 1 based on the received .d b d ~ .,
value R. It is given that the conditional density 3.99 Cons1 er a. inary ata transmiss l()f' a• •"•
functions of R as 56 kbps using baseband binary Pu se :~=·i: <ii
modulation (PAM) that is designed a: ..,. ·~~
. d . ~ n~
_! -3SrS1 ra ise -cos ine spect rum The tral"\s,,., " •~:•
~ 0 (r) = 4' and bandwidth (in kHz) required of a rol -o" ss ;:-
{
0, otherwise 0.25 is _ _ . ' fac:c, :i

..! -1s,ss
lRt,(r) = 6 •
[2016 : 1 Mark S .
, e,·1]
{
0, otherwise
The minimum decision error probability is 3.100 An analog pulse s( t) is transmittea
1 additive white Gaussian noise (A WGN) C:Oh\e· a:
ar,,,
(a) O (b) 12 The received signal is r{ t) = s( t) + rt._ t\ \~her~ t;
is additive white Gaussian noise with t;., ;
1 1 Po1,e-
(c) - (d) 6
9 . No
spectral density - - The received s·1gna s
[2015: 2 Marks, Set-1] 2

3.96 A sinusoidal signal of amplitude A is quantized passed through a filter with impulse res PQnse
by a uniform quantizer. Assume that the signal h(.t). Let Es and Eh denote the energies of t'e
utilizes all the representation levels of the pulse s(t) and the filter h(t) , respectively. Whsn
quantizer. If the signal to quantization noise ratio the signal to noise ratio (SNR) is maximizea a'.
is 31.8 dB, the number of levels in the quantizer the output of the filter (SNRmaJ. which of te
is__. following holds?
[2015 : 1 Mark, Set-2]

3.97 Consider a binary, digital conmunication system

n n
which uses pulses g,_ and -{/.. for transmitting E
(b) Es= Eh; SNRmax = _ s
bits over an AWGN channel. If the receiver uses 2N0
a matched filter, which one of the following pulses
will give the minimum probability of bit error? (c) Es > Eh; SNRmax > 2Es
g(t) g(t)
No

. 2Eh
(d) Es< Eh, SNRmax = -
No
(a) (b)
[2016: 2 Marks, Set- 11
0
3.101 An ideal band-pass channel 500 Hz- 2000 Hz is
g(t) g(t) 1
deployed for communication. A modem:
designed to transmit bits at the rate o'
c'
4800 bits/s using 1WAM. The roll--0ff fact<1 ,.
(c) (d) trum trti,
a pulse with a raised cosine spec
0 1 t utilizes the entire frequenCY band is ~ l
(2015 : 2 Marks, Set-2] [2016: 2 Marl<S,
 ntcatlon Systems
11111
I
cort1 - - - - - - -- - __
is sampled at 8 kHz and - 3.105 I .
/
s(i'eC~into signal
pCM tormat using. 8 bits/sample n d d1g11n1 ro,, rn • 1 1 1 c::~==-=
I m
r pr
~
d :.nd'd~• data 1s_ transmitted through 0·
118 r.pCiv• t,ennel via 4-IeveI r . M. The minim □'A
pulso sh
upf.t th/JUJ
sample, has
~ t100 m
8 at tt ,e ,ecc ,c, om ,
,
f r,CJ <i H ) equ1·red f t um symbol Fourier trnns'orm P(f) f
(In I< z r or ransmission symbolss are Iran rr- lt(.d at the rate of 2000

◄ ◄
per second 1
"' :: .~· [20 6 M cases is th '" ich of .he t1.eto?
or ,n,erle•ence ry:, ilQ
ca'i,r • : , ark, Set-2] e inter simoo .

ase.band digital communicatio P(fl

·ne.rf b n
.4' ~ t>1 rnploys the signal (a)
IV" s'15ts'fl e
+.o ~ t ~ T5

l
12 l'(l<HL)
-1 .2 0
p(t}::: ✓ Ts Pf/)
O, otherwise

transrnission of bits. The graphical {b)

f(kHZ)
~:~resentation ~f the matched filter output }'{ t) -1 2 --0.8 o 08 12

for this signal will be P(f)

y(t)

(c)
(a) .2--..J...1_ _...1.0_ _..L-l
--1--': 1 1.2- - !(kHz)

0 T, P(f)

y(t)

__ __ __,
(d)
!(kHz)
(b} 0 12
_... ........ -1.2
[2017 : 1 Mark, Set-1]
2T,
0
3.106 Which one of the following statements about
y(I)
differential pulse code modulation (OPCM) is

true?
(c) (a) The sum of message signal sample with its
prediction is quantized.
r. 2r. (b) The message signal sample is directly
0
quantized, and its prediction is not used.
y(t)
(c) The difference of message signal sample
and a random signal is quantized.
(d) (d) The difference of message signal sample
with its prediction is quantized.
[2017 : 1 Mark, Set-1]
0 T/2 T5
[2016 : 1 t,1ark, Set-3]
_107 In binary frequency shift keying (FSK), the given
3·104 The bit· error probability of a memoryless binary 3 signal waveforms are
symmetric channel is 1o- If 1a5 bits are sent 5. u (t) == 5cos(2()()001tt) ; 0 ~ t 5. T and
over th is channel then the probability that not u (t)0 == 5cos(22{)()01t t) ; 0 5. t 5. T,
more than one bi~ will be in error is ---=·3
r201 6 : 2 Marks, set- 1
1
 S76 j
GATE Previo us Years So lved Papers : m I
MADE
EAs,.,1
- - - - - -- - ·- 3 . 111 /\ r.111()I£ 1>11 , oqu.1ll 1 hkolytob oo ttr111 I
where T Is the b1t-duratIon 1ntorval and / 1s 1r1 t,on t aero ,s :in ,1dci111vo •t1h1te 0J ,J, JS<: 1o~~
seconds . Both u0(1) and u1(1) aro zoro ours,cJo (AWC-,N) ch,mnol With f)O tlr•r S .. an '
~ Pectrr
the interval O s t s T With a matched filt er Nol? 8 1nAry Sl(Jnallnq , With o - , P<t) J ~c
11
(correlator) based receiver, tho smallest positive ' fl(J
q(I) , Is used for tho transrn I55I0n ,llonc
•
,
value of T(in milliseconds) required to have u0 ( I) optimal receiver that mmim,, e th J ,,
and u1(t) uncorrelated is probability. " e b t-e,,,./(
(a) 0.25 ms (b) 0.5 ms Let cp 1(/),cp 2(1) form an orthonormar s1gna
(c) O. 75 ms (d) 1.0 ms If we choo~e p(I) =_cp 1(1) and q(t).. &et
[2017 : 2 Marks, Set-1] would obtain a certain bit-error prob b <9,(/J i.
a iht1P
3. 108 A sinusoidal message signal is converted to a If we keep p(I) = <p 1(/) , but take ~ t) == [£ ...
PCM signal using a uniform quanti zer. The
for what value of E would we obtain th ~/ti
required signal to quantization noise ratio (SQNR) bit-error probability Pb? e sa'1",e
at the output of the quantizer is 40 dB. The (a) 0 (b) 3
minimum number of bits per sample needed to (c) 1 (d) 2
achieve the desired SQNR is [2019 : 2 Marks]
[2017 : 1 Mark, Set-2]
3.112 A vo ice signal m(t) is in the frequency range
. 5
3.109 A random variable X takes values -0.5 and 0.5 kHz to 15 kHz . The signal is amp i·tI Ude.

with probabilities l and ¾,respectively. The noisy modulated to generate an AM signal f{I) == A 1
+ m(t)) cos 2n f/, where fc = 600 kHz. The (
observation of Xis Y= X + Z, where Zhas uniform signal f(t) is to be digitized and archived. T~~
probability density over the interval (-1 , 1). X and is do~e by first sampl ing f(t) at 1.2 times t~:
Z are independent. If the MAP rule based detector Nyquist fr~quency, and then quantizing each
outputs X as sample using a 256-level quantizer. Finally, each
quantized sample is binary coded using Kbits
x= {-0.5, y < a.
where K is the minimum number of bits require~
0.5, y~ a.,
for the encoding . The rate , in Megabits per
then the value of a. (accurate to two decimal
second (rounded off to 2 decimal places), of the
places) is _ _
resulting stream of coded bits is _· _ Mbps.
[2018 : 2 Marks]
[2019 : 2 Marks]
3. 110 A bin~ source generates symbols Xe {-1, 1}
3.113 A random variable X takes values -1 and +1
which are transmitted over, a noisy channel. The
with probabilities 0.2 and 0.8. respective ly. It is
probability of transmitting X = 1 is 0.5. Input to
transmitted across a channel which adds noise
the threshold detecto r is R = X + N. The
N, so that the random variable at the channel
probability density function fN(n) of the noise N
output is Y = X + N. The noise N is independent
is shown below.
of X, and is uniformly distributed over the interval
fN(n)
[- 2 , 2]. The receiver makes a decision

0.5
x= {-1, if Y ~ e
+1 , if Y < e
where the threshold ee [- 1, 1] is chosen so as
-2 2 n to minimize the probability of error Pr [Xt X].
If the detecti on threshold is zero , then the The minimum probability of error, rounded off to
probab ility of error (correct to two decimal
1 decimal place, i s _.
places) is _ _ __ [2019 : 2 Marks]
[2018 : 1 Mark]
•••
MADEEASY I Communication Systems I 577

LifiWl§fj Dig ital Communication Systems

3.1 (c) 3 .2 (c) 3 .3 (c) 3.4 (b) 3 .5 (a) 3.6 (b) 3 .7 (b)
3.8 (3600) 3. 9 (sol.) 3 .10 (3600) 3.11 (c) 3 .12 (a) 3 .13 (a) 3.14 (sol.)
3.15 (a) 3 .1 6 (3) 3 .17 (b) 3.18 (c) 3.19 (c) 3 .20 (c) 3 .2 1 (c)
3 .22 (d) 3 .23 (c) 3 .24 (d) 3 .25 (c) 3 .26 (a) 3.27 (d) 3.28 (a)
3 .29 (c) 3 .30 (b) 3.31 (d) 3.32 (a) 3.33 (b) 3.34 (c) 3.35 (c)
3 .36 (d) 3 .37 (c) 3 .38 (a) 3 .39(d) 3.40 (b) 3.41 (c) 3 .42 (d)
3 .43 (b) 3.44 (b) 3.45 (b) 3.46 (c) 3.47 (a) 3.48 (c) 3.49 (a)
3 .50 (c) 3 .51 (c) 3 .52 (b) 3.53 (c) 3.54 (c) 3.55 (b) 3 .56 (b)
3 .57 (c) 3 .58 (a) 3.59 (d) 3.60 (c) 3.61 (c) 3.62 (b) 3 .63 (a)
3 .64 (b) 3.65 (a) 3.66 (a) 3.67 (d) 3.68 (a) 3.69 (d) 3 .70 (a)
3 .71 (c) 3 .72 (b) 3.73 (c) 3.74 (c) 3.75 (c) 3.76 (c) 3.77 (d)
3.78 (d) 3.79 (d) 3.80 (c) 3.81 (d) 3.82 (b) 3.83 (d) 3.84 (d)
3.85 (d) 3.86 (b) 3.87 (1.414) 3.88 (c) 3.89 (0 .25) 3.90 (200) 3.91 (17)

3 .92 (a) 3.93 (0.4) 3.94 (400) 3.95 (d) 3.96 (32) 3.97 (a) 3.98 (d)
3 .99 (35) 3.100 (a) 3.101 (0.25) 3 .102(16) 3.103(c) 3.104 (0 .7357) 3 .105 (b)
3.106 (d) 3.107 (b) 3.108 (7) 3.109(-0.50) 3 .110 (0 .125) 3 . 111 (b) 3 .112 ( 11 .8 1)
3 .113(0.10)