GATE-2023

Electronics Engineering

Signals & Systems

Chapterwise & Topicwise

Contents
S.No. Topic Page No.

1. Basics of Signals & Systems .............................................................................................................. 1-11

2. LTI Systems Continuous and Discrete (Time Domain) ........................................................ 12-29

3. Fourier Series ..................................................................................................................................... 30-44

4. Fourier Transforms, Frequency Response and Correlation ................................................ 45-63

5. Laplace Transform ............................................................................................................................ 64-87

6. Z-Transform ..................................................................................................................................... 88-109

7. DTFS, DTFT and DFT ................................................................................................................... 110-121

8. Sampling ........................................................................................................................................ 122-131

9. Digital Filters ................................................................................................................................. 132-133

 1 Basics of Signals & Systems

ELECTRO NICS EN GINEERIN G E

(a) E (b)
(GATE Previous Years Solved Papers) 2
(c) 2E (d) 4E
Q.1 An excitation is applied to a system at t = T and [EC-2001 : 1 Mark]
its response is zero for – < t < T. Such a system
is a Q.6 Let P be linearity, Q be time-invariance, R be
causality and S be stability. A discrete-time
(a) non-causal system
system has the input-output relationship,
(b) stable system
(c) causal system x(n) n 1
(d) unstable system (n) = 0 n=0
[EC-1991 : 2 Marks] x(n + 1) n 1

where x(n) is the input and y(n) is the output.

Q.2 Which of the following signals is/are periodic?
The above system has the properties.
(a) s(t) = cos2t + cos3t + cos5t
(a) P, S but not Q, R
(b) s(t) = exp(j8 t)
(b) P, Q, S but not R
(c) s(t) = exp(–7t) sin10 t
(c) P, Q, R, S
(d) s(t) = cos2t cos4t
(d) Q, R, S but not P
[EC-1992 : 2 Marks]
[EC-2003 : 2 Marks]
Q.3 A system with an input x(t) and output y(t) is
Q.7 Consider the sequence
described by the relation: y(t) = tx(t). This system
is x [n] = [ 4 j 5, 1 + j 2, 4]
(a) linear and time-invariant
The conjugate antisymmetric part of the
(b) linear and time varying
sequence is
(c) non-linear and time-invariant
(a) [–4 –j2.5, j2, 4 – j2.5]
(d) non-linear and time-varying
(b) [–j2.5, 1, j2.5]
[EC-2000 : 1 Mark]
(c) [–j5, j2, 0]
Q.4 Let (t) denote the delta function. The value of (d) [–4, 1, 4]
[EC-2003 : 2 Marks]
3t
the integral (t ) cos dt is
2 Q.8 The function x(t) is shown in the figure. Even
and odd parts of a unit-step function u(t) are
(a) 1 (b) –1 respectively.
x(t)
(c) 0 (d)
2
1
[EC-2001 : 1 Mark]
t
Q.5 If a signal f(t) has energy E, the energy of the 0
singal f(2t) is equal to –1
 2 Electronics Engineering Signals & Systems

1 1 1 1 Which of the following descriptions

(a) , x(t ) (b) , x(t ) corresponds to a causal system?
2 2 2 2
1 1 1 1 (a) y(t) = x(t – 2) + x(t + 4)
(c) , x(t ) (d) , x(t )
2 2 2 2 (b) y(t) = (t – 4) + x(t + 1)
[EC-2005 : 1 Mark] (c) y(t) = (t + 4) + x(t – 1)
(d) y(t) = (t + 5) + x(t + 5) [EC-2008 : 1 Mark]
Q.9 The power in the signal
Q.14 Let x(t) be the input and y(t) be the output of a
s(t ) = 8 cos 20 t + 4 sin (15 t ) is continuous time system. Match the system
2
properties P1, P2 and P3 with system relations
(a) 40 (b) 41
R1, R2, R3, R4.
(c) 42 (d) 82
[EC-2005 : 1 Mark] Properties:
P1 : Linear but not time-invariant
Q.10 The dirac delta function (t) is defined as,
P2 : Time-invariant but not linear
1 t=0 P3 : Linear and time-invariant
(a) (t ) =
0 otherwise
Relation:
t=0 R1 : y(t) = t2 x(t)
(b) (t ) =
0 otherwise
R2 : y(t ) = t x(t )
1 t=0
(c) (t ) = and (t ) dt = 1 R3 : y(t ) = x(t )
0 otherwise
R4 : y(t) = x(t – 5)
t=0 (a) (P1, R1) (P2, R3), (P3, R4)
(d) (t ) = and (t ) dt = 1
0 otherwise (b) (P1, R2) (P2, R3), (P3, R4)
[EC-2006 : 1 Mark] (c) (P1, R3) (P2, R1), (P3, R2)
(d) (P1, R1) (P2, R2), (P3, R3)
Q.11 A system with input x[n] and output y[n] is given
[EC-2008 : 2 Marks]
as,
5 Q.15 The input x(t) and output y(t) of a system are
y [n] = sin n x (n)
6 related as,
The system is t
y(t ) = x ( ) cos(3 ) d
(a) linear, stable and invertible.
(b) non-linear, stable and non-invertible. The system is
(c) linear, stable and non-invertible. (a) time-invariant and stable.
(d) linear, unstable and invertible. (b) stable and non time-invariant.
[EC-2006 : 2 Marks] (c) time-invariant and not stable.
(d) not time-invariant and not stable.
Q.12 A Hilbert transformer is a
[EC-2012 : 2 Marks]
(a) non-linear system
(b) non-causal system Q.16 For a periodic signal,
(c) time-varying system v(t ) = 30 sin 100t + 10 cos 300t + 6 sin 500t +
(d) low-pass system 4
The fundamental frequency (in rad/sec) is
[EC-2007 : 2 Marks]
(a) 100 (b) 300
Q.13 The input and output of a continuous time (c) 500 (d) 1500
system are respectively denoted by x(t) and y(t). [EC-2013 : 1 Mark]
 GATE Previous Years Solved Paper 3

Q.17 The impulse response of a continuous time (a) T = 2 Ts (b) T = 1.2Ts

system is given by
(c) Always (d) Never
h(t) = (t – 1) + (t – 3)
[EC-2016 : 1 Mark]
The value of the step response at t = 2 is
(a) 0 (b) 1 Q.22 Consider a single input single output discrete-
(c) 2 (d) 3 time system with x[n] as input and y[n] as output,
[EC-2013 : 2 Marks] where the two are related as,

n x[n ] for 0 n 10
Q.18 A discrete-time signal x[n] = sin( 2n), n being an y[n] =
x[n] x[n 1] otherwise
integer, is
(a) periodic with period Which one of the following statements is true
(b) periodic with period 2 about the system?
(c) periodic with period /2 (a) It is causal and stable.
(d) not periodic (b) It is causal but not stable.
[EC-2014 : 1 Mark] (c) It is not causal but stable.
(d) It is neither causal nor stable.
Q.19 The waveform of a periodic signal x(t) is shown
[EC-2017 : 1 Mark]
in the figure.
Q.23 The input x(t) and the output y(t) of a
x(t)
continuous-time system are related as,
3 t
y(t ) = x(u) du
–2 1 4 t T
t
–4 –1 2 3
The system is
–3
(a) linear and time-variant.
(b) linear and time-invariant.
t 1 (c) non-linear and time-variant.
A signal g(t) is defined by g(t ) = x .
2 (d) non-linear and time-invariant.
The average power of g(t) is ________ . [EC-2017 : 1 Mark]
[EC-2015 : 1 Mark]
Q.24 Let the input be ‘u’ and the output be ‘y’ of a
Q.20 Two sequence x1[n] and x2[n] have the same system, and the other parameters are real
energy. Suppose x1[n] = 0.5n u[n], where is a constants. Identify which among the following
positive real number and u[n] is the unit step systems is not a linear system:
sequence. Assume, d3 y d2 y dy
(a) 3
+ a1 2
+ a2 + a3 y
1.5 for n = 0, 1 dt dt dt
x2 [ n] =
0 otherwise du d2u
b3 u + b2 + b1 2
Then the value of is _______ . dt dt
[EC-2015 : 2 Marks] (with initial rest conditions)
t
Q.21 A continuous time function x(t) is periodic with (t )
(b) y(t ) = e u( ) d
period T. The function is sampled uniformly 0
with a sampling period Ts. In which one of the (c) y = au + b, b 0
following cases is the sampled signal periodic? (d) y = au [EC-2018 : 1 Mark]
 4 Electronics Engineering Signals & Systems

ELECTRICAL EN GINEERIN G (a) 5t (b) 5 u(t) – C

(GATE Previous Years Solved Papers) 5 5 u(t )

(c) t (d)
C C
+6 [EE-2002 : 1 Mark]
2t
Q.1 The value of the integral e (t 1) dt is
5 Q.5 Which of the following is true?
(a) A finite amplitude signal is always time
equal to ________ .
bounded.
[EE-1994 : 1 Mark]
(b) A time bounded signal always possesses
Q.2 The rms value of the periodic waveform e(t), finite energy.
shown in figure is (c) A time bounded signal is always zero
e(t ) outside the interval [–t0, t0] for some t0.
(d) A time bounded signal is always finite in
+A
amplitude.
[EE-2006 : 1 Mark]
0 t
T/2 T
Q.6 A continuous-time system is described by
–A y(t) = e–|x(t)|, where y(t) is the output and x(t) is
the input, y(t) is bounded

3 2 (a) only when x(t) is bounded.

(a) A (b) A (b) only when x(t) is non-negative.
2 3
(c) only for t 0 if x(t) is bounded for t 0.
1
(c) A (d) 2A (d) even when x(t) is not bounded.
3
[EE-2006 : 2 Marks]
[EE-1995 : 1 Mark]
Q.7 Given a sequence x[n], to generate the sequence
Q.3 What is the rms value of the voltage waveform y[n] = x[3 – 4n], which one of the following
shown in figure? procedures would be correct?
(a) First delay x[n] by 3 samples of generate
v(t )
z1[n], then pick every 4th samples of z1[n] to
+100 V
generate z2[n], and then finally time reverse
z2[n] to obtain y[n].
0 t (b) First advance x[n] by 3 samples of generate
/3 2 /3 2
z1[n], then pick every 4th samples of z1[n] to
generate z2[n], and then finally time reverse
–100 V
z2[n] to obtain y[n].
(c) First pick every 4th sample of x[n] to generate
200 100 v1[n], time-reverse v1[n] to obtain v2[n], and
(a) V (b) V
finally advance v 2[n] by 3 rd samples to
(c) 200 V (d) 100 V obtain y[n].
[EE-2002 : 1 Mark] (d) First pick every 4th sample of x[n] to generate
v1[n], time-reverse v1[n] to obtain v2[n], and
Q.4 A current impulse of 5 (t), is forced through a
finally delay v2[n] by 3rd samples to obtain
capacitor C. The voltage V c(t), across the
y[n].
capacitor is given by
[EE-2006 : 2 Marks]
 GATE Previous Years Solved Paper 5

Q.8 The function shown in the figure can be

f(t)
represented as,
4

2.7
t
–3.3 –1.3 –0.3 0.7 3.7 4.7

t –2
0 T 2T

(t T ) (t 2T ) [EE-2017 : 1 Mark]
(a) u(t ) u(t T ) + u(t T ) u(t 2T )
T T
Q.12 The signal energy of the continuous-time signal,
t t
(b) u(t ) + u(t T ) u(t 2T ) x(t) = [(t – 1) u(t – 1)] – [(t – 2) u(t – 2)]
T u
– [(t – 3) u(t – 3)] + [(t – 4) u(t – 4)], is
(t T ) (t 2T )
(c) u(t ) u(t T ) + u(t ) u(t )
11 7
T T
(a) (b)
(t T ) (t 2T ) 3 3
(d) u(t ) + u(t T ) 2 u(t 2T )
T T 1 5
(c) (d)
[EE-2014 : 2 Marks] 3 3
[EE-2018 : 2 Marks]
+
Q.9 The value of t Q.13 The symbol a and T, represent positive
e (2t 2) dt , where (t) is the
quantities, and u(t) is the unit step function.
Which one of the following impulse responses
Dirac delta function, is
is not the output of a causal linear time-invariant
1 2 system?
(a) (b)
2e e
(a) 1 + e–at u(t) (b) e+at u(t)
1 1 (c) e–a(t – T) u(t) (d) e–a(t + T) u(t)
(c) 2 (d) 2
e 2e [EE-2019 : 1 Mark]
[EE-2016 : 1 Mark]
Q.14 xR and xA are, respectively, the rms and average
Q.10 Consider the system with following input- values of x(t) = x(t – T), and similarly, yR and yA
output relation y[n] = (1 + (–1)n) x[n], where x[n] are, respectively, the rms and average values of
is the input and y[n] is the output. The system is y(t) = kx(t) k, T are independent of t.
(a) invertible and time invariant Which of the following is true?
(b) invertible and time varying (a) yA = kxA ; yR = kxR
(c) non-invertible and time invariant (b) yA = kxA ; yR kxR
(d) non-invertible and time varying (c) yA kxA ; yR kxR
[EE-2017 : 1 Mark] (d) yA kxA ; yR = kxR
[EE-2020 : 1 Mark]
Q.11 The mean square value of the given periodic
waveform f(t) is _______ .
6 Electronics Engineering Signals & Systems

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Basics of Signals & Systems

1. (c) 2. (a, b, d) 3. (b) 4. (a) 5. (b) 6. (a) 7. (a) 8. (a)

9. (a) 10. (d) 11. (c) 12. (b) 13. (c) 14. (a) 15. (d) 16. (a)

17. (b) 18. (d) 19. (2) 20. (1.5) 21. (b) 22. (a) 23. (b) 24. (c)
25. (0) 26. (c, d)

Solutions
EC Basics of Signals & Systems

1. (c) 4. (a)
For the given system if the response is zero prior 3t 3× 0
to the application of the excitation. Then such a (t ) cos dt = f (0) = cos = cos 0 = 1
2 2
system is called causal system.
5. (b)
2. (a, b, d)
(a) s(t) is periodic as the ratio of any two E= f (t )2 dt
frequencies = p/q is rotational.
Where p and q are integers.
E = f (2t )2 dt
(b) s(t) is periodic with =8
(d) 2 cosA cosB = cos(A – B) + cos(A + B)
dp dp
= f ( p )2 2t = p ; dt =
1 2 2
s(t ) = [cos 2t + cos 6t ]
2
E
So, s(t) is periodic with fundamental E =
2
frequency 2 rad/sec.
6. (a)
3. (b)
y(n – n0) = x(n – n0 + 1) (time varying)
ay1(t) = atx1(t) y(n) = x(n + 1) (depends on future)
ay2(t) = atx2(t) i.e., y(1) = x(2) (non-causal)
a[y1(t) + y2(t)] = a[tx1(t) + tx2(t)] For bounded input, system has bounded output.
So it is stable,
System is linear,
y(n) = x(n); n 1
y(t – t0) = (t – t0) x(t – t0) = 0; n=0
Delay is introduced. So, time varying. = x(n + 1), n –1
So, system is linear.
 GATE Previous Years Solved Paper 7

7. (a) 10. (d)

x(n) = [ 4 j 5, 1 + 2 j , 4]
(t )

x (–n) = [4, 1 2 j , 4 + j 5]
t

x( n ) x ( n )
xCAS (n) = 11. (c)
2
= [ 4 2.5 j , 2 j , 4 2.5 j ] 5
y(n) = sin n x(n)
6
Let, x(n) = (n)
8. (a)
y(n) = sin0 = 0 (Bounded)
u(t ) + u( t ) BIBO stable.
Even part =
2
u(t ) u( t ) 12. (b)
Odd part = The Hilbert transformer is characterised by the
2
impulse response,
u(t ) u(–t)
1
h(t) = t (– , )
1 1 t
t t
h(t) 0 for t < 0
Thus, the Hilbert transformer is a non-causal
system.
1
ue(t) = 13. (c)
2
A system is causal if the output at any time
ue(t )
depends only on values of the input at the
1/2 present time and in the past.

15. (d)
t
t
y = x ( ) cos(3 ) d
x(t )
uo(t) =
2 t t0
uo(t) y(t – t0) = x ( ) cos(3 ) d

1/2
y (t) for input x(t – t0) is
t t
0
–1/2 y (t) = x( t0 )cos 3 d

(t t0 )
9. (a) y (t) = x ( )cos 3( + t0 ) d

s(t) = 8 cos 20 t + 4 sin 15 t y (t) y(t – t0) so system is not time invariant for
2
input x( ) = cos(3 ) (bounded input),
= 8 sin20 t + 4 sin15 t
t
82 4 2 y(t) = cos2 (3) d as t
P= + = 32 + 8 = 40
2 2
So, for bounded input-output is not bounded
therefore system is not stable.
8 Electronics Engineering Signals & Systems

16. (a) Energy of x2[n] = 1.5 + 1.5 = 3

1 = 100, 2 = 300, 3 = 500 2 4

=3
HCF of ( 1, 2 and 3) = HCF (100, 300, 500),
3
= 100 rad/sec. 9
2 = = 1.5
4
17. (b)
Step response = Integration of impulse response, 21. (b)
t y(t ) A signal is said to be periodic if T/Ts is a
(t 1) dt = u(t – 1) rational number.
2
Here, T = 1.2 Ts
t 1
(t 3) dt = u(t – 3) T 6
t =
1 2 3 Ts 5
At, t=2 Which is a rational number.
y(t) = 1
22. (a)
18. (d) Since present output does not depend upon
x[n] = sin( 2 n) future values of input, the system is causal and
2
o = also every bounded input produces bounded
2 output, so it is stable.
N= m
o
23. (b)
where, m is the smallest integer that converts
t
2 / o into a integer value.
Given that, y(t) = x(u) du
2 2
N= 2
m= m t T

• Since the given system satisfies both

So, there exists no such integer value of m which
homogeneity and additivity properties, the
could make the N integer, so the system is not
system is linear.
periodic.
• Check for time invariance:
19. Sol. t to
x(t) = –3t, –1 < t < 1 y(t – t0) = x(u) du
t to T
t 1 3
x = (t 1) –1 < t < 1
2 2 • When the applied input is x(t – t0),
and T=6 t t to

2 y1(t) = x(u to ) du = x( ) d
3
1 3 t T t to T
Average power = (t 1) dt = 2
6 2 = y(t – t0)
1
System is time invariant.
20. Sol.

2
24. (c)
Energy of x1[n] = x1 [ n ]
y = au + b, b 0 is a non-linear system.
n=

2 1 n 1
= = 2
4 1
n=0 1
4
2 4
=
3
 GATE Previous Years Solved Paper 9

Answers
EE Basics of Signals & Systems

2. (b) 3. (d) 4. (d) 5. (c) 6. (d) 7. (b) 8. (a) 9. (a)

10. (d) 11. (6) 12. (d) 13. (a) 14. (b)

Solutions
EE Basics of Signals & Systems

1. Sol. Now we will take every 4th samples of z1[n]

Using, i.e. z2[n] = z1[4n] = x[4n + 3]
Now reverse (time reverse) z2[n] will give
x(t ) (t t0 ) dt = x(t0) y[n] = z2[–n] = x[–4n + 3]

8. (a)
We have,
6
2t
e (t t1 ) dt = e–2 1
5

2. (b) t
0 T 2T
1/2
T /2 2 T
1 2A 2 The given function can be realized as follows:
Rms value = t dt + A dt
T T
0 T /2 u(t) u(t – T)

1 1
2
= A
3

t t
3. (d) 0 0 T
u(t) u(t – T)
Rms value (Unit step function right shifted)
(Unit step function)
2 (i) (ii)
/3 2 /3
1
= (100)2 dt + ( 100)2 dt + (100)2 dt r(t)
0 /3 2 /3
1 1
1/2
1
= 100 2 = 100 V
t t
0 0 T
4. (d) r(t) = tu(t) (t T )
(Ramp function) u( t T )
T
t t
1 1 5 (Right shifted unit ramp function)
Vc (t ) = i(t ) dt = 5 (t ) dt = u(t ) (iii)
C C C
1
7. (b)
y[n] = x[3 – 4n] = x[–4n + 3]
t
So to obtain y[n] we first advance x[n] by 3 unit. 0 T 2T
i.e., z1[n] = x[n + 3] (t T )
u(t 2T )
T
(iv)
 10 Electronics Engineering Signals & Systems

Combining the functions (i), (ii), (iii) and (iv), Invertibility test:
we get the given function. x(n) y( n )
(n 1) [1 + ( 1)n ] ( n 1) = [1 + ( 1)1 ( n 1) = 0 (n 1) = 0
f(t)
2 (n 1) [1 + ( 1)n ]2 ( n 1) = [1 + ( 1)1 ]2 ( n 1) = 0.2 ( n 1) = 0

1 Thus, we are getting many to one mapping

between input and output. So, the system is non-
invertible.

11. Sol.
t Mean square value = Power of f(t)
0 T 2T
1 2
Mean square value = f (t ) dt
Therefore, T0 T0

(t T ) 1 2
f(t) = u(t ) u(t T ) + = [4 × 1 + 2 2 × 2]
T 4
(t 2t ) 16 + 8 24
u(t T ) u(t 2T ) = = =6
T 4 4
1 1
u(t ) u(t T ) + r (t T ) r(t 2T ) 12. (d)
T T
1 x(t) = r(t – 1) – r(t – 2) – r(t – 3) + r(t – 4)
or u(t ) u(t T ) + (t T ) u(t 2T ) u(t 2T )
T x(t)

9. (a) 1

t–1 –(t – 4)
t
To find value of e (2t 2) dt
t
1 2 3 4
1
Since, (t 1)
(2t – 2) =
2 2
x(t ) dt
Ex(t) =
Above integral can be written as,

1 1 t 1 2 3 4
e t
(t 1) dt = e =
2 2e = (t 1)2 dt + 12 dt + { (t 4)} 2 dt
2
1 2 3

10. (d) 1 1 2 5
= + 1+ = +1 =
3 3 3 3
Given relationship,
y(n) = [1 + (–1)n] x(n) 13. (a)
The invariant test: a and T represents positive quantities.
Delay of
u(t) is unit step function,
x( n) System y(n) y( n – 1)
‘1’
= [1 + (–1)
n–1
]
h(t) = 1 + e–at u(t), is non-causal
x(n – 1) Here ‘1’ is a constant and two sided so the
system will be non-causal, because for causal
Delay of System y (n) = [1 + (–1) ]
n system,
‘1’ x(n – 1) x(n – 1) h(t) = 0, t < 0
Since, y(n – 1) y (n) 0, t > 0
So, the system is time variant.
 GATE Previous Years Solved Paper 11

14. (a, b) Case (ii): When K is imaginary or complex or

Given, y(t) = Kx(t) ...(i) real and negative then,
then, Average of y(t) = K × Average of x(t) K K
Y A = KXA and YR KXR
From equation (i), Thus, option (b) is satisfied, option (a) and (b)
Power of y(t) = K 2 power of x(t ) both satisfies the given condition.

YR2 = K 2 X R2 [Q Power = rms2]

YR = K X R
Case (i): When K is real and positive then,
K =K
and YR = KXR
Thus option (a) is satisfied.
 2 LTI Systems Continuous and
Discrete (Time Domain)
ELECTRO NICS EN GINEERIN G 1 at
(a) ae–at (b) (1 e )
(GATE Previous Years Solved Papers) a
(c) a(1 – eat) (d) 1 – e–at
Q.1 The impulse response and the excitation
[EC-1998 : 1 Mark]
function of a linear time invariant causal system
are shown in Fig. (a) and (b) respectively. The Q.4 Let u(t) be the step function. Which of the
output of the system at t = 2 sec. is equal to waveform in the figure corresponds to the
h(t ) x(t)
convolution of u(t) – u(t – 1) with u(t) –u(t – 2)?
1 1

1 1/2
(a) (b)
0 6 t(sec) 0 2 6 t(sec)
0 1 2 0 1 2 3
1 t t
(a) 0 (b)
2
1 1
3
(c) (d) 1
2
[EC-1990 : 2 Marks] (c) (d)

Q.2 Let h(t) be the impulse response of a linear time 0 1.5 2 3 0 1 3

invariant system. Then the response of the t t
system for any input u(t) is [EC-2000 : 2 Marks]
t
(a) Q.5 The impulse response functions of four linear
h ( ) u(t )d
0 systems S1, S2, S3, S4 are given respectively by
h1(t) = 1 ; h2(t) = u(t)
t
d
(b) h ( ) u(t )d u(t )
dt h3 (t ) = ; h4(t) = e–3t u(t)
0 t+1
t t where u(t) is the unit step function. Which of
(c) h ( ) u(t )d dt these systems is time invariant, causal, and
0 0 stable?
t (a) S1 (b) S2
(d) h 2 ( ) u(t )d (c) S3 (d) S4
0
[EC-2001 : 2 Marks]
[EC-1995 : 1 Mark]
Q.6 Convolution of x(t + 5) with impulse function
Q.3 The unit impulse response of a linear time (t – 7) is equal to
invariant system is the unit step function u(t). (a) x(t – 12) (b) x(t + 12)
For t > 0, the response of the system to an
(c) x(t – 2) (d) x(t + 2)
excitation e–at u(t), a > 0 will be
[EC-2002 : 1 Mark]
 GATE Previous Years Solved Paper 13

Q.7 The impulse response h[n] of a linear time- (a) is positive and is positive.
invariant system is given by (b) is negative and is negative.
h[n] = u[n + 3] + u[n –2] – 2u[n – 7] (c) is positive and is negative.
where u[n] is the unit step sequence. (d) is negative and is positive.
The above system is [EC-2008 : 1 Mark]
(a) stable but not causal.
Q.10 A discrete-time linear shift-invariant system has
(b) stable and causal.
an impulse response h[n] with h[0] = 1, h[1] = –1,
(c) causal but unstable. h[2] = 2, and zero otherwise. The system is given
(d) unstable and not causal. an input sequence x[n] with x[0] = x[2] = 1 and
[EC-2004 : 1 Mark] zero otherwise. The number of non-zero samples
in the output sequence y[n], and the value of
Q.8 Which of the following can be impulse response
y[2] are, respectively
of a causal system?
(a) 5, 2 (b) 6, 2
h (t )
(c) 6, 1 (d) 5, 3
[EC-2008 : 2 Marks]
(a)
Q.11 A system is defined by its impulse response
t h(n) = 2n u(n – 2). The system is
(a) stable and causal.
h (t ) (b) causal but not stable.
(c) stable but not causal.
(b) (d) unstable and non-causal.
t [EC-2011 : 1 Mark]

Q.12 Let y[n] denotes the convolution of h[n] and g[n],

h (t ) where h[n] = (1/2)n u[n] and g[n] is a causal
sequence. If y[0] = 1 and y[1] = 1/2, then g[1]
(c) equals
1
t (a) 0 (b)
2
3
h (t ) (c) 1 (d)
2
[EC-2012 : 2 Marks]
(d)
Q.13 Two systems with impulse responses h1(t) and
t h2(t) are connected in cascade. Then the overall
impulse response of the cascaded system is
[EC-2005 : 1 Mark] given by

Q.9 The impulse response h(t) of a linear time- (a) product of h1(t) and h2(t)
invariant continuous time system is described (b) sum of h1(t) and h2(t)
by h(t) = exp( t) u(t) + exp( t) u(–t), where u(t) (c) convolution of h1(t) and h2(t)
denotes the unit step function, and and are (d) subtraction of h2(t) and h1(t)
real constants. This system is stable if [EC-2013 : 1 Mark]
 14 Electronics Engineering Signals & Systems

Q.14 Consider a discrete-time signal The impulse responses of the systems are
n for 0 0 10 h1(t) = 2 (t + 2) – 3 (t + 1)
x[n ] = h2(t) = (t – 2)
0 otherwise
If y[n] is the convolution of x[n] with itself, the If the input x(t) is a unit step signal, then the
value of y[4] is ________ . energy of y(t) is __________ .
[EC-2014 : 2 Marks] [EC-2017 : 2 Marks]

Q.15 The result of the convolution x(–t) (–t –t0) is Q.20 The output of y[n] of a discrete-time system for
(a) x(t + t0) (b) x(t – t0) an input x[n] is
(c) x(–t + t0) (d) x(–t – t0) y[n ] = max x [ k]
k n
[EC-2015 : 1 Mark]
The unit impulse response of the system is
Q.16 The impulse response of an LTI system can be
(a) 0 for all n.
obtained by
(b) 1 for all n.
(a) differentiating the unit ramp response
(c) unit step signal u[n].
(b) differentiating the unit step response
(d) unit impulse signal [n].
(c) integrating the unit ramp response
(d) integrating the unit step response [EC-2020 : 1 Mark]

[EC-2015 : 1 Mark] Q.21 For a unit step input u[n], a discrete-time LTI
system produces an output signal (2 [n + 1] +
Q.17 Which one of the following is an eigen function
of the class of all continuous-time, linear, time- [n] + [n – 1]). Let y[n] be the output of the system
invariant systems (u(t) denotes the unit-step n
1
function)? for an input u(n) . The value of y[0] is
2
(a) e j ot u(t ) (b) cos( ot)
________ .
(c) e j ot (d) sin ( ot) [EC-2021 : 2 Marks]
[EC-2017 : 1 Mark] Q.22 The output of four systems (S1, S2, S3 and S4)
Q.18 Two discrete-time signals x[n] and h[n] are both corresponding to the input signal sin(t) for all
non-zero only for n = 0, 1, 2 and are zero time t, are shown in the figure.
otherwise. It is given that,
sin(t ) S1 sin(–t)
x[0] = 1, x[1] = 2, x[2] = 1, h[0] = 1
Let y[n] be the linear convolution of x[n] and sin(t ) S2 sin(t + 1)
h[n]. Given that, y[1] = 3 and y[2] = 4, the value
sin(t ) S3 sin(2t)
of the expression (10 y[3] + y[4]) is ________ .
[EC-2017 : 2 Marks] sin(t ) S4
2
sin (t)

Q.19 Consider the parallel combination of two LTI

Based on the given information, which of the
systems shown in the figure.
four systems is/are definitely not LTI (linear and
h1(t) time-invariant)?
(a) S1 (b) S2
x(t) y(t) (c) S3 (d) S4
[EC-2022]
h2(t)
 GATE Previous Years Solved Paper 15

ELECTRICAL EN GINEERIN G t
(a) s (t ) u( ) d
(GATE Previous Years Solved Papers) 0

Q.1 s(t) is step response and h(t) is impulse response t

d
of a system. Its response y(t) for any input u(t) is (b) s (t ) u( ) d
dt
0
given by
t t t
d s(t ) u(
(a) s(t ) u( ) d (c) 1) d 1 d
dt 0 0
0
t t
(b) s (t ) u( ) d (d) s (t )2 u( ) d
0 0
t [EE-2000 : 1 Mark]
(c) s (t ) u( 1) d 1 d
00 Q.7 x[n] = 0, n < –1, n > 0, x[–1] = –1, x[0] = 2 is the
t input and y[n] = 0, n < –1, n > 2, y[–1] = –1 =
d
(d) h( t ) u( ) d [EE-1993 : 1 Mark] y[1], y[0] = 3, y[2] = –2 is the output of a discrete-
dt
0
time LTI system.
Q.2 If f(t) is the step-response of a linear time- The system impulse response h[n] will be
invariant system, then its impulse response is (a) h[n] = 0, n < 0, n > 2, h[0] = 1, h[1] = h[2] = –1
given by ______ . (b) h[n] = 0, n < –1, n > 1, h[–1] = 1, h[0] = h[1] = 2
[EE-1994 : 1 Mark] (c) h[n] = 0, n < 0, n > 3, h[0] = –1, h[1] = 2,
Q.3 The convolution of the function f1(t) = e–2t u(t) h[2] = 1
and f2(t) = et u(t) is equal to ______. (d) h[n] = 0, n < –2, n > 1, h[–2] = h[1] =
[EE-1995 : 1 Mark] h[–1] = –h[0] = 3
[EE-2006 : 2 Marks]
Q.4 The impulse response of an initially relaxed
linear system is e–2t u(t). To produce a response Q.8 If u(t), r(t) denote the unit step and unit ramp
of te–2t u(t), the input must be equal to functions respectively and u(t) r(t) their
convolution, then the function u(t + 1) r(t – 2)
1 2t
(a) 2 e–t u(t) (b) e u(t ) is given by
2
(c) e–2t u(t) (d) e–t u(t) 1
(a) (t 1) u (t 2)
2
[EE-1995 : 1 Mark]
1
Q.5 The unit impulse response of a system is given (b) (t 1) u (t + 2)
2
as,
1
c(t) = –4e–t + 6e–2t (c) (t 1)2 u(t 1)
2
The step response of the same system for t 0 is
(d) None of the above
equal to
(a) –3e–2t – 4e–t + 1 (b) –3e–2t + 4e–t + 1 Q.9 Let a signal a1 sin( 1t + 1) be applied to a stable
linear time invariant system. Let the
(c) –3e–2t – 4e–t – 1 (d) 3e–2t + 4e–t + 1
corresponding steady-state output be
[EE-1996 : 2 Marks]
represented as a2 F( 2t + 2). Then which of the
Q.6 Let s(t) be the step response of a linear system following statements is true?
with zero initial conditions. Then the response (a) F is not necessarily a ‘sine’ or ‘cosine’
of this system to an input u(t) is function but must be periodic with 1 = 2.
 16 Electronics Engineering Signals & Systems

(b) F must be a ‘sine’ or ‘cosine’ function with Q.13 A linear time invariant system with an impulse
a1 = a2. response h(t) produces output y(t) when input
(c) F must be a ‘sine’ function with 1 = 2 and x(t) is applied. When the input x(t – ) is applied
1 = 2. to a system with impulse response h(t – ), the
(d) F must be a ‘sine’ or ‘cosine’ function with output will be
1 = 2. (a) y( ) (b) y(2(t – ))
[EE-2007 : 1 Mark] (c) y(t – ) (d) y(t – 2 )
Q.10 The impulse response of a causal linear time- [EE-2009 : 1 Mark]
invariant system is given as h(t). Now consider
Q.14 A cascade of three linear time invariant systems
the following two statements:
is causal and unstable. From this, we concludes
Statement (I) : Principle of superposition holds.
that
Statement (II) : h(t) = 0 for t < 0.
(a) each system in the cascade is individually
(a) Statement (I) is correct and statement (II) is
causal and unstable.
wrong.
(b) Statement (II) is correct and statement (I) is (b) at least one system is unstable and at least
wrong. one system is causal.
(c) Both statement (I) and statement (II) are (c) at least one system is causal and all systems
wrong. are unstable.
(d) Both statement (I) and statement (II) are (d) the majority are unstable and the majority
correct. are causal.
[EE-2008 : 1 Mark] [EE-2009 : 2 Marks]
Q.11 A signal e– t sin ( t) is the input to a real linear Q.15 The system represented by the input-output
time invariant system. Given K and are
5t
constants, the output of the system will be of the
relationship y(t ) = x( ) d , t > 0 is
from Ke– t sin(vt + ) where
(a) need not be equal to but v equal to .
(b) V need not be equal to but equal to . (a) linear and causal
(c) equal to and v equal to . (b) linear but not causal
(d) need not be equal to and v need not to be (c) causal but not linear
equal to . (d) neither linear nor causal
[EE-2008 : 1 Mark] [EE-2010 : 1 Mark]
Q.12 A system with x(t) and output y(t) is defined by Q.16 Given the finite length input x[n] and the
the input-output relation: corresponding finite length output y[n] of an
2t LTI system as shown below, the impulse
Y (t ) = x( t ) d response h[n] of the system is

The system will be h[n]

x[n] = {1, –1} y[n] = {1, 0, 0, 0, –1}
(a) causal, time-invariant and unstable.
(b) causal, time-invariant and stable.
(c) non-causal, time-invariant and unstable. (a) h[n] = {1, 0, 0, 1} (b) h[n ] = {1, 0, 1}
(d) non-causal, time-variant and unstable.
(c) h[n ] = {1, 1, 1, 1} (d) h[n] = {1, 1, 1}
[EE-2008 : 2 Marks]
[EE-2010 : 2 Marks]
 GATE Previous Years Solved Paper 17

Q.17 Given two continuous time signals x(t) = e–t and Q.22 The impulse response of a continuous time
y(t) = e–2t which exist for t > 0, the convolution system is given by
z(t) = x(t) y(t) is h(t) = (t – 1) + (t – 3)
(a) e–t – e–2t (b) e–3t The value of the step response at t = 2 is
(c) e +t (d) e–t + e–2t (a) 0 (b) 1
[EE-2011 : 1 Mark] (c) 2 (d) 3

Q.18 Let y[n] denote the convolution of h[n] and g[n], [EE-2013 : 2 Marks]
where h[n] = (1/2)n u[n] and g[n] is a causal
Q.23 x(t) is non-zero only for Tx < t < Tx , and
sequence. If y[n] = 1 and y[1] = 1/2, then g[1]
equals similarly, y(t) is non-zero only for Ty < t < Ty .
(a) 0 (b) 1/2
Let z(t) be convolution of x(t) and y(t). Which
(c) 1 (d) 3/2
one of the following statements is true?
[EE-2012 : 2 Marks]
(a) z(t) can be non-zero over an unbounded
Q.19 The input x(t) and output y(t) of a system are interval.
related as, (b) z(t) is non-zero for t < Tx + Ty .
t
(c) z(t) is zero outside of Tx + Ty < t < Tx + Ty .
y (t ) = x( ) cos(3 ) d

(d) z(t) is non-zero for t > Tx + Ty .

The system is
(a) time-invariant and stable. [EE-2014 : 1 Mark]
(b) stable and not time-invariant.
Q.24 Consider an LTI system with impulse response
(c) time-invariant and not stable. h(t) = e–5t u(t). If the output of the system is
(d) not time-invariant and not stable. y(t) = e–3t u(t) – e–5t u(t) then the input, x(t) is
[EE-2012 : 2 Marks] given by
Q.20 The impulse response of a system is h(t) = t u(t). (a) e–3t u(t) (b) 2 e–3t u(t)
For an input u(t – 1), the output is (c) e–5t u(t) (d) 2 e–5t u(t)
[EE-2014 : 1 Mark]
t2 t (t 1)
(a) u(t ) (b) u(t 1)
2 2 Q.25 The impulse response g(t) of a system, G is as
shown in Fig. (a). What is the maximum value
(t 1)2 t2 1
(c) u(t 1) (d) u(t 1) attained by the impulse response of two
2 2
cascaded blocks of G as shown in Fig. (b)?
[EE-2013 : 1 Mark]

Q.21 Two systems are impulse response h1(t) and 1

h2(t) are connected in cascaded. Then the overall G G
impulse response of the cascaded system is
t (b)
given by 0 1
(a)
(a) product of h1(t) and h2(t)
(b) sum of h1(t) and h2(t) 2 3
(a) (b)
(c) convolution of h1(t) and h2(t) 3 4
(d) subtraction of h2(t) from h1(t) 4
(c) (d) 1
[EE-2013 : 1 Mark] 5
[EE-2015 : 1 Mark]
 18 Electronics Engineering Signals & Systems

Q.26 A moving average function is given by Q.30 Consider a causal LTI system characterized by
t differential equation,
1
y( t ) = u( ) d
T dy(t ) 1
t T + y(t ) = 3x(t )
dt 6
If the input ‘u’ is a sinusoidal signal of frequency
The response of the system to the input
1 x(t) = 3e–t/3 u(t), where u(t) denotes the unit step
Hz, then the steady-state, the output ‘y’ will
2T function, is
lag ‘u’ (in degree) by _______ . (a) 9 e–t/3 u(t)
[EE-2015 : 1 Mark] (b) 9 e–t/6 u(t)
(c) 9 e–t/3 u(t) – 6 e–t/6 u(t)
Q.27 For linear time invariant systems, that are (d) 54 e–t/6 u(t) – 54 e–t/3 u(t)
bounded input bounded output stable, which
[EE-2016 : 1 Mark]
one of the following statement is true?
(a) The impulse response will be integrable, but Q.31 Let z(t) = x(t) y(t), where ‘ ’ denotes
may not be absolutely integrable. convolution. Let c be a positive real-valued
(b) The unit impulse response will have finite constant. Choose the correct expression for z(ct).
support. (a) c x(ct) y(t) (b) x(ct) y(ct)
(c) The unit step response will be bounded. (c) c x(ct) y(ct) (d) c x(ct) y(t)
(d) The unit step response will be bounded. [EE-2017 : 1 Mark]
[EE-2015 : 2 Marks] Q.32 A continuous-time input signal x(t) is an eigen
Q.28 Consider a continuous-time system will input function of an LTI system, if the output is
x(t) and output y(t) given by (a) kx(t), where k is an eigen value.
y(t) = x(t) cos(t) (b) ke j t x(t), where k is an eigen value and e j t
The system is is a complex exponential signal.
(c) x(t) ej t, where e j t is a complex exponential
(a) linear and time-invariant.
signal.
(b) non-linear and time-invariant.
(d) k H( ), where k is an eigen value and H( ) is
(c) linear and time-varying
a frequency response of the system.
(d) non-linear and time-varying.
[EE-2018 : 1 Mark]
[EE-2016 : 1 Mark]
Q.33 Suppose for input x(t) a linear time-invariant
Q.29 The output of a continuous-time, linear time-
system with impulse response h(t) produces
invariant system is denoted by T[x(t)] where x(t)
output y(t), so that x(t) h(t) = y(t). Further, if
is the input signal. A signal z(t) is called eigen
signal of the system T, when T[z(t)] = yz(t), where x(t ) h(t ) = z(t ), which of the following
y is a complex number, in general, and is called
statements is true?
an eigen value of T. Suppose the impulse
(a) For some but not all t (– , ), z(t) y(t)
response of the system T is real and even. Which
of the following statements is true? (b) For all t (– , ), z(t) y(t)
(a) cos(t) is an eigen-signal but sin(t) is not. (c) For all t (– , ), z(t) y(t)
(b) cos(t) and sin(t) are both eigen-signals but (d) For some but not all t (– , ), z(t) y(t)
with different eigen values. [EE-2020 : 2 Marks]
(c) sin(t) is an eigen-signal but cos(t) is not.
(d) cos(t) and sin(t) are both eigen-signals with
identical eigen values.
[EE-2016 : 2 Marks]
 GATE Previous Years Solved Paper 19

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC LTI Systems Continuous and Discrete (Time Domain)

1. (b) 2. (a) 3. (b) 4. (b) 5. (d) 6. (c) 7. (a) 8. (b)

9. (d) 10. (d) 11. (b) 12. (a) 13. (c) 14. (10) 15. (d) 16. (b)

17. (c) 18. (31) 19. (7) 20. (c) 21. (0) 22. (c, d)

Solutions
EC LTI Systems Continuous and Discrete (Time Domain)

1. (b) 2

For causal LTI system, y(2) = x( ) h(2 )d

0
h(t) = 0 for t < 0
y(t) = x(t) h(t) 1 1 1
So, y(2) = × 2× =
2 2 2
y(t) = x( ) h(t )d
2. (a)
t For LTI system,
y(t) = x( ) h(t )d y(t) = u(t) h(t)
0
As, x(t) = 0 ; t<0 y(t) = h( ) u(t )d
and h(t) = 0 ; t<0
Q u(t) = 0 ; for t < 0
h(2 – )
t
y(t) = h( ) u(t )d
1 0

3. (b)
–4 0 2
Given that:
Impulse response = h(t) = u(t)
x( )
So, transfer function,
1
H(s) =
s
1/2
Y (s )
H(s) =
2 X( s )
Y(s) = H(s) × H(s)
20 Electronics Engineering Signals & Systems

Given that, 6. (c)

x(t) = e–at u(t) x(t + 5) (t – 7) = x(t + 5 – 7)
1 = x(t – 2)
X(s) =
(s + a)
7. (a)
1 1
Y(s) = ×
s (s + a ) h( k ) =
1 1 1 k=
Y(s) =
a s s+a
u( k + 3) + u( k 2)
Taking inverse Laplace transform,
k= 3 k=2
1 at
y(t) = [1 e ]
a 2 u( k 7)
k =7
4. (b)
6 6
1 s = 1+ 1
u(t) – u(t – 1) = [1 e ]
s k= 3 k=2

1 = 10 + 5 = 15 <
[1 e 2 s ]
u(t) – u(t – 2) =
s For bounded input, bounded output. So system
Convolution in the domain = Multiplication in is stable.
s-domain Response depends on future value of input
1 signal i.e., u(n + 3). So, system is not causal.
[1 e s ] [1 e 2s
]
s2 8. (b)
1 2s s 3s For a causal system,
= [1 e e +e ]
s2 h(t) = 0 ; for t < 0
= tu(t) – (t – 1) u(t – 1) – (t – 2) u(t – 2) + (t – 3)
9. (d)
u(t – 3)
h(t) = e t u(t) + e t u(–t)
= r(t) – r(t – 1) – r(t – 2) + r(t – 3)

5. (d) For the system to be stable, h(t ) dt < .

h1(t) 0 ; for t < 0
For the above condition, h(t) should be as shown
Therefore s1 is non-causal
below.
h2(t) = u(t) = 0 for t < 0
h (t )
h2 (t ) dt = u(t ) dt = dt =
0 t – t
e u(–t) e u (t)
Therefore s2 is unstable and causal,
u(t ) t
h3(t) =
t+1
Therefore, < 0 and > 0.
At, t = –1
h3(t) = 10. (d)
h3(t) = 0 for t < 0 L1 = 3, L2 = 3
Therefore s3 is unstable and causal, Number of samples in output
h4(t) = e–3t u(t) = L1 + L2 – 1 = 5
s4 is time invariant, causal and stable.
 GATE Previous Years Solved Paper 21

h[–k]
y[n] = x[ k ] h[n k ]
k= 1/2 1

1/4
y[2] = x[ k ] h[2 k ]
k=

y(2) = 3

x [k ] h[k] y[0] = h [ k ] g[ k ]
k=
1 1 2
y[0] = h[0] g[0]
1
1 = 1 g[0]
k k g[0] = 1
0 1 2 0 1 2
–1
y[1] = h [1 k ] g[ k ]
k=
h[–k] h[2 – k]
y[1] = h[1] g[0] + h[0] g[1]
2 2
h[1 – k] will be zero for k > 1 and g[k] will be
1 1
zero. For k < 0 as it is causal sequence,
–1
k k 1 1
–2 0 0 1 2
= × 1 + 1 g[1]
–1 –1 2 2
g[1] = 0
11. (b)
13. (c)
h(n) = 2n u(n – 2)
The overall impulse response h(t) of the
For causal system,
cascaded system is given by
h(n) = 0 ; for n < 0
h(t) = h1(t) h2(t)
Hence given system is causal.
For stability: 14. Sol.
n ; for 0 n 10
2n = x[n] =
n=2 0; otherwise

So, given system is not stable.

y[n] = x( k ) x( n k )
12. (a) k=

n 10
1 x( k ) x(4 k )
h[n] = u [ n] y[4] =
2 k=0
g[n] = ?? = x(0) x(4) + x(1) x(3) + x(2) + x(3)
x(1) + 0
g(0) g(1)
= 0 + (1 × 3) + (2 × 2) + (3 × 1)
= 10

15. (d)
x(–t) (–t – t0) = x(–t) (t + t0)
y[n] = h [n k ] g[ k ] = x(–t – t0)
k=
22 Electronics Engineering Signals & Systems

16. (b) By plotting y(t), we get

Step response, y(t )
t
y(t) = h( ) d 2

d
h(t) = y (t ) –2 –1 0 2
dt t(sec)

17. (c)
–1
If the input to the system is eigen signal output
is also the same eigen signal. The energy of y(t) can be given as,

18. Sol.
Ey = y 2 (t ) dt
x[n] = {1, 2, 1}
h[n] = {1, a, b} By plotting y2(t), we get
y[n] = {A, 3, 4, B, C}
y2(t)
h
x 1 a b 4

1 1 a b
1
2 2 2a 2b
–2 –1 0 2 t(sec)
1 1 a b

Ey = Area under the plot of y2(t)

y[0] = 1; y[1] = 2 + a
= (4 × 1) + (3 × 1) = 7 J
y[2] = 1 + 2a + b
y[3] = a + 2b, y[4] = b 21. (0)
Given, y[1] = 2+a=3 a=1 1
y[2] = 1 + 2a + b = 4 b = 1 x(n) = u(n) X( z) = 1
1 z
So, y[3] = a + 2b = 3
y(n) = 2 [n + 1] + [n] + [n – 1]
and y[4] = b=1
Y(z) = 2z + 1 + z–1
So, 10y[3] + y[4] = 10 × 3 +1 = 31
Transform function,
19. Sol. Y ( z) 2 z + 1 + z 1
H(z) = =
Given that, X( z) 1
h1(t) = 2 (t + 2) – 3 (t + 1) 1 z 1
h2(t) = (t – 2) = (2z + 1 + z–1) (1 – z–1)
Overall impulse response is, = 2z + 1 + z–1 – 2 – z–1 – z–2
h(t) = h1(t) + h2(t) H(z) = 2z – 1 – z–2
= 2 (t + 2) – 3 (t + 1) + (t – 2) n
1
If input, x(t) = u(t), then the output will be If x(n) = u(n) for this system with
2
y(t) = x(t) h(t)
= u(t) * [2 (t + 2) – 3 (t + 1) H ( z) = 2z – 1 – z–2 then,
+ (t – 2)] find y (n )I.Z.T.
= 2u(t + 2) – 3u(t + 1) + u(t – 2)
h(n) = 2 (n + 1) – (n) – (n – 2)
 GATE Previous Years Solved Paper 23

22. (c, d)
2z 1 z 2
Y(z) = X ( z) H ( z) =
1 1 sin t S1 sin(–t) = –sin t
1 z
2
sin t S2 sin(t + 1)
y(n) = x(n) h(n)
n+1 n sin t S3 sin(2t)
1 1
= 2 u(n + 1) u(n)
2 2 1 cos 2 t
sin t S4 sin 2 (t ) =
2
n 2
1 Since, LTI system does not change the frequency
u(n 2)
2 of sinusoidal input. So, S3 and S4 are definitely
As we want y(0), it is due to first 2 terms of y(n), not LTI as input and output sinusoidal
1 0+1
1 0 frequencies are different.
y(0) = 2
2 2
= 1–1=0

Answers
EE LTI Systems Continuous and Discrete (Time Domain)

1. (a) 4. (c) 5. (b) 6. (b) 7. (a) 8. (c) 9. (d) 10. (d)

11. (c) 12. (d) 13. (d) 14. (b) 15. (b) 16. (c) 17. (a) 18. (a)

19. (d) 20. (c) 21. (c) 22. (b) 23. (c) 24. (b) 25. (d) 26. (90)

27. (d) 28. (c) 29. (d) 30. (d) 31. (a) 32. (a) 33. (b)

Solutions
EE LTI Systems Continuous and Discrete (Time Domain)

1. (a) 2. Sol.
y(t) = u(t) h(t) (always) d
f (t )
where h(t) is impulse response dt
Q s(t) is step response (given)
3. Sol.
d
s(t ) = h(t) Q f1(t) = e–2t u(t)
dt
d 1
F1(s) =
y(t) = u(t ) s(t ) s+2
dt
d and f2(t) = et u(t)
= [u(t ) s(t )] (Using property)
dt 1
F2(s) =
s 1
d
= u( ) s(t )d
dt 1
F1(s) F2(s) =
(s + 2) (s 1)
(By definition of convolution)
t 1/3 1/3
d = +
= u( ) s(t )d s+2 s 1
dt
0 Q F1(s) F2(s) = f1(t) f2(t)
(Q u(t) and s(t) are causal functions)
24 Electronics Engineering Signals & Systems

Multiplication in s-domain results in 8. (c)

convolution in time domain.
u(t + 1) r(t – 2) = r( 2) u(t + 1 )d
1/3 1/3
f1(t) f2(t) = L 1 +
s+2 s 1 ( t + 1)
z(t) = r( 2) d
1 2t 1
= e + et u(t ) (t + 1)
3 3 z(t) = ( 2) d
2
For, (t + 1) 2
4. (c)
2 ( t + 1)
1 1 (t + 1)
H(s) = L [ e 2t u(t )] = z(t) = 2 = [ 2
4 ]2
s+2 2 2
2
2t 1
Y(s) = L [t e u(t )] = 1
(s + 2)2 For t 1= [(t + 1)2 4(t + 1) 4 + 8]
2
Y (s ) 1
X(s) = = 1 2
X( s ) s + 2 z(t) = [t + 2t + 1 4t 4 + 4]
2
Hence, input x(t) = L –1 [X(s) = e–2t u(t)]
For t 1
5. (b) 1
z(t) = (t 1)2
Impulse response = 4 e–t +6 e–2t 2
t t t For t 1
Step response = c(t ) dt = 4 e t dt + 6 e 2t
dt 1
0 0 z(t) = (t 1)2 u(t 1)
2
= 4e–t – 3e–2t – 1
Hence, option (b) is correct. 9. (d)

7. (a) Input |H| fH Output

For finite duration convolution,
x[n] have M terms The effect of linear time invariant system is best
h[n] have N terms understood by considering that, at any given
then, y[n] should have terms (M + N – 1) input frequency 1, the output is sinusoidal at
Here, x[n] = { 1, 2} same frequency, as the input applied, with
amplitude given by output = H × input and
y[n] = { 1, 3, 1, 2}
phase equal to 2 = H + 1, where H is the
So, h[n] should have only 3 terms and h[n] have
magnitude of the frequency response and H is
value starting from origin or [n = 0] only because
y[n] have value start from [n = –1]. its phase angle. Both H and H are functions
Consider option (I). of frequency.
–1 1
1 –1 2 10. (d)

–1 1 –2 Since system is causal therefore its impulse

response will be zero for t < 0.
–1 1 –2 So statement-II is correct.
since system is linear it also satisfies the
principle of superposition.
y[n] = { 1, 3, 1, 2}
So both the statement are correct.
So option (a) is correct.
 GATE Previous Years Solved Paper 25

11. (c)
e at j H( + j 0 ) j 0t
= H( +j 0 ) [e e
t LTI system t 2j
e sin( ot ) Ke sin( vt + )
h(t ) = real signal
j H( +j 0) j 0t
For real h(t): e e ]
h(t) = h (t) = Ke– t sin(vt + )
Applying Laplace transform,
12. (d)
H(s) = H (s )
H(s + j ) = H ( – j ) [Put s = + j ] First of all we will check for causality,
= 0, = –a 2t
y(t) = x( ) d
H(– + j 0) = H (– – j 0)
So for, t = –2
+ j 0)
H( + j 0 ) e j H(
4
y(–2) = x( ) d
j H( j 0)
= H( j 0) e
So output depends on future values of input
By comparison,
along with past and present values of input so
H( +j 0) = H( j 0) ...(i) system is non-causal.
and H(– + j 0) = – H(– – j 0) ...(ii) Let us find output for shifted input x(t – t0),
Now system input is, 2t
y (t) = x( t0 ) d
x(t) = e– t sin 0t
(2 t t0 )
1 t y (t) = x( ) d ...(i)
= e ( e j 0t e j 0t
)
2j
Now shift the output by t0
1 t( + j 0) (2 t t0 )
= e et ( j 0 )
then, y(t – t0) = x( ) d ...(ii)
2
1 so from equations (i) and (ii),
= [ x1 (t ) x2 (t )] y (t) y(t – t0)
2j
Therefore system is time variant.
+ j 0)
where, x1(t) = et ( Non-causal and time variant is present only in
t( j 0)
option (d).
and x2(t) = e
For x1(t) system output is, 13. (d)
Case-1 : Y(s) = H(s) X(s)
y1(t) = H ( s ) s = j 0
e( j 0 )t
Case-2 : Input = x(t – ) X(s) e–s
= H( j 0) e( j 0 )t Impulse response = h(t – ) H(s) e–s
Thus, the overall output y(t) is, Output : Y (s) = X(s) e–s H(s) e–s
Y (s) = X(s) H(s) e–s2t = Y(s) e–2sT
1
y(t) = [ y1 (t ) y2 (t )] y(t) = y(t – 2 )
2j
[Q By using property of linear system] 14. (b)
1
[H( + j 0 ) e( + j 0 )t
H( j 0 ) e( j 0 )t
] Since in cascade overall impulse response,
= 2j
h(t) = h1(t) h2(t) h3(t)
e at h 1 (t), h 2 (t), h 3 (t) are impulse response of
+ j 0)
= [ H( +j 0 ej H(
ej 0t
individual systems.
2j
Since initial point where h(t) is non-zero is t 0
H( j 0 ) e j H( j 0)
e j 0t ] and since in convolution initial point,
= t1 + t2 + t3
26 Electronics Engineering Signals & Systems

where, t1, t2, t3 are initial points of h1(t), h2(t), 17. (a)
h3(t) respectively. 1
1
So for it to be greater than or equal to zero at x(s) = , y( s ) =
s+1 s+2
least one of them t1, t2 or t3 must be +ve i.e. greater
z(s) = x(s) y(s)
than zero so atleast one of them must be causal.
1 1 1 1
Similarly if one (atleast) of the system become = × =
s+1 s+2 s+1 s+2
unstable then overall system will become
z(t) = e–t – e–2t
unstable.
18. (a)
15. (b)
Given, g[n] is a causal sequence.
Integrator is always a linear system.
g[n] will be zero for n < 0,
5t
Since, y(t) = x( ) d t>0
y[n] = g[ k ] h[n k ]
For t = 1 k=
5 n
y(t) = x( ) d 1
and h[n] = u [ n]
2
here value at t = 1 depends on future values like
at t = 2, 3, .... of input x(t). n
1
u[n]
So, it is a non-causal system. h[n] = 2
0 n<0
16. (c)
1 1
x[n] = {1, 1}, M = 2 h[0] = 1, h [1] = , h [2] =
2 4
y[0] = g[0] h[0] + g[1] h[–1] + 1...
y[n] = {1, 0, 0, 0, 1}, N 1 = 5 1 = g[0] × 1 + 0 + 0 + ...
g[0] = 1
Since output has number of elements, y[1] = g[0] h[1] + g[1] h[0] + g[2] h[–1] +...
N1 = M + N – 1 1 1
= 1 + g[1] 1 + 0
where N is number of elements in impulse 2 2
response h[n]. g[1] = 0
5 = 2+N–1
19. (d)
N=4
t
Let it be a1, a2, b1, b2 y = x( ) cos(3 ) d
y[n] = [ a1 ,( a2 a1 ), (b1 a2 ), (b2 b1 ), b2 ] t t0
y(t – t0) = x( ) cos(3 ) d
comparing with,
y (t) for input x(t – t0) is
y[n] = [1, 0, 0, 0, 1] t
y (t) = x( t0 ) cos 3 d
a1 = 1
( t t0 )
a2 = a1 = 1 y (t) = x( ) cos 3( + t0 ) d
b1 = a1 = 1
y (t) y(t – t0) so system is not time invariant for
b2 = b1 = 1
input x( ) = cos(3 ) bounded input,
Therefore,
t
y(t) = cos2 (3 ) d as t
h[n] = {1, 1, 1, 1}
So, for bounded input, output is not bounded
therefore system is not stable.
 GATE Previous Years Solved Paper 27

20. (c) Also, h(t) = e–5t u(t)

h(t) = t u(t) 1
H(s) =
1 (s + 5)
H(s) =
s2 Y (s)
=
2
× (s + 5)
X(s) =
Y (s) 1 H (s) (s + 3) (s + 5)
= 2
U (s) s 2
= (s + 3)
1 e s
Y(s) = Input = x(t) = 2e–3t u(t)
s2 s
(t 1)2 25. (d)
y(t) = u (t 1)
2 g(t) = u(t) – y(t –1)
1 e s
21. (c) G(s) =
s s
In cascade connection,
or, G(s) × G(s) = g(t) g(t)
H1(s) H2(s) g(t) g(t) g(t)

H(s) = H1(s) H2(s)

1 1
h(t) = h1(t) h2(t)

22. (b)
t
For step response, impulse response can be 1
0 1 2
integrated.
Maximum value = 1
y(t) = u(t – 1) + u(t – 3)
y(2) = u(1) + u(–1) 26. Sol.
= 1+0=1 System input: sin 0t

23. (c)
1
f0 = Hz
2T
x(t) is non-zero for Tx < t < Tx and y(t) is non- Therefore,

zero for Ty < t < Ty 2

0 = 2 0= = rad/sec.
2T T
Q z(t) = x(t) y(t) t
1 1 t
then the limits of the resultant signal is, y(t) = sin 0 d = [ cos 0 t ]t T
T T
Tx + Ty < t < Tx + Ty t T

i.e. z(t) is non-zero for Tx + Ty < t < Tx + Ty .

1
= [cos 0 t cos[ 0 (t T )]]
T
24. (b) t
0 =
Impulse response of an LTI system = transfer T
1 t
Y (s ) = cos at cos
function = = H (s) T T
X( s )
1 2
where, y(t) = e–3t u(t) – e–5t u(t) = [cos 0t + cos 0 t ] = cos 0 t
T T
1 1 2 2
Y(s) = = = sin( 0 t 90°)
(s + 3) ( s + 5) (s + 3) (s + 5) T
= Output is lagging by 90° w.r.t. input.
28 Electronics Engineering Signals & Systems

28. (c) 30. (d)

y(t) = x(t) cos(t) The differential equation,
To check linearity, dy(t ) 1
+ y(t ) = 3 x(t)
y1(t) = x1(t) cos(t) dt 6
[y1(t) is output for x1(t)]
1
y2(t) = x2(t) cos(t) So, sY (s ) + Y (s ) = 3 X(s)
6
[y2(t) is output for x2(t)]
3 X(s )
So, the output for (x1(t) + x2(t)) will be Y(s) =
1
y(t) = [x1(t) + x2(t)] cos(t) s+
6
= y1(t) + y2(t)
So, the system is linear to check time invariance. 9
X(s) =
The delayed output, 1
s+
y(t – t0) = x(t – t0) cos(t – t0) 3
The output for delayed input, 9
y(t, t0) = x(t – t0) cos(t) So, Y(s) =
1 1
s+ s+
Since, y(t – t0) y(t, t0) 3 6
System is time varying.
54 54
=
29. (d) 1 1
s+ s+
6 3
Given that impulse response is real and even.
Thus, H(j ) will also be real and even. So, y(t) = (54e–16t – 54e–1/3t) u(t)

j ot LTI j ot
31. (a)
e H(j o) e
system Time scaling property of convolution.
If, x(t) y(t) = z(t)
–j o t LTI –j o t
e H(–j o) e 1
system
Then, x(ct) y(ct) = z(ct )
c
Since, H(j ) is real and even. z(ct) = c × x(ct) y(ct)
Thus, H(j 0) = H(–j 0)
32. (a)
e jt + e jt
Now cos(t) is input i.e. is input/ Eigen function is a type of input for which output
2
is constant times of input.
output will be
i.e.,
H ( j 1) e jt + H ( j 1) e jt e jt + e jt
= H ( j 1)
at
e = x(t) H(s) y(t )
2 2
= H(j1) cos(t) where, x(t) = system input = eigen function
H(s) = transfer function of system
e jt e jt y(t) = system output
If sin(t) is input i.e. is input/output
2
Here, y(t) = H ( s ) s = a e at
will be
= K x(t)
H ( j 1) e jt + H ( j 1) e jt e jt + e jt
= H ( j 1) 2j
where, K = eigen value = H (s) s = a
2
= H(j1) sint x(t) = eigen function input = H (s) s = a
So, sin(t) and cos(t) are eigen signals with same
eigen values.
 GATE Previous Years Solved Paper 29

33. (b) Case-2:

Since, y(t) = x(t) + h(t) x (t ) h (t )

and z(t) = x(t ) × h(t ) 1 1

Case-1:
t t
0 1 0 1
x(t ) h(t )

1
then, y(t) = z(t)
0 1
t t
0 1
–1
1

then, y(t) and z(t)

t
0 1 2
y(t ) z (t )

1 2 Thus, z(t) y(t) for all ‘t’

–1
t
0

t
–1 0 1 2
 3 Fourier Series

ELECTRO NICS EN GINEERIN G Q.5 The trigonometric Fourier series of a periodic

time function can have only
(GATE Previous Years Solved Papers)
(a) cosine terms
Q.1 A half-wave rectified sinusoidal waveform has (b) sine terms
a peak voltage of 10 V. Its average value and the (c) cosine and sine terms
peak value of the fundamental component are (d) dc and cosine terms
respectively given by
[EC-1998 : 1 Mark]
20 10 10 10
(a) V, V (b) V, V Q.6 A periodic signal x(t) of period T0 is given by
2 2
10 20 1, t < T1
(c) V, 5 V (d) V, 5 V x( t ) = T0
0 , T1 < t <
[EC-1987 : 2 Marks] 2
The dc component of x(t) is
Q.2 Fourier series of the periodic function (period 2 )
defined by T1 T1
(a) (b)
T0 (2T0 )
0, <x<0
f (x) =
x, 0<x< 2T1 T0
(c) (d)
T0 T1
1 1
is + [cos(n ) 1] cos(nx ) cos(n )sin(nx ) [EC-1998 : 1 Mark]
4 n=1 n2 n
Q.7 The Fourier series representation of an impulse
By putting x = in the above, one can deduce train denoted by
1 1 1
that the sum of the series 1 + 2
+ 2
+ + ... s(t ) = (t nT0 )
3 5 72
n=
is
1 j 2 nt
[EC-1993 : 2 Marks] (a) exp
T0 n= T0
Q.3 The Fourier series of an odd periodic function,
contains only 1 j nt
(b) exp
(a) odd harmonics (b) even harmonics T0 n= T0
(c) cosine terms (d) sine terms
1 j nt
[EC-1994 : 1 Mark] (c) exp
T0 n= T0
Q.4 The trigonometric Fourier series of an even
function of time does not have 1 j 2 nt
(d) exp
(a) the dc term T0 n= T0
(b) cosine terms [EC-1999 : 2 Marks]
(c) sine terms
(d) odd harmonic terms Q.8 One period (0, T) each of two periodic
[EC-1996 : 1 Mark] waveforms W1 and W2 are shown in the figure.
 GATE Previous Years Solved Paper 31

The magnitudes of the nth Fourier series Q.12 The Fourier series of a real periodic function
coefficients of W1 and W2, for n 1, n odd, are has only
respectively proportional to P. Cosine terms if it is even.
Q. Sine terms if it is even.
R. Cosine terms if it is odd.
1 1
S. Sine terms if it is odd.
W1 W2 Which of the above statements are correct?
T (a) P and S (b) P and R
0 T/2 T 0 T/2
(c) Q and S (d) Q and R
[EC-2009 : 1 Mark]
–1 –1
Q.13 The trigonometric Fourier series of an even
function does not have the
(a) n 3 and n 2 (b) n 2 and n 3 (a) dc term (b) cosine terms
(c) sine terms (d) odd harmonic terms
1 2 4 2
(c) n and n (d) n and n [EC-2011 : 1 Mark]

[EC-2000 : 2 Marks] Q.14 Consider the periodic square wave in the figure
shown.
Q.9 Which of the following cannot be the Fourier
x(t)
series expansion of a periodic signal?
(a) x(t) = 2 cost + 3 cos3t 1

(b) x(t) = 2 cos t + 7 cost

(c) x(t) = cost + 0.5 t
0 1 2 3 4
(d) x(t) = 2 cos 1.5 t + sin3.5 t
[EC-2002 : 1 Mark]
–1
Q.10 The Fourier series expansion of a real periodic
signal with fundamental frequency fo is given by The ratio of the power in the 7th harmonic to the
power in the 5th harmonic for this waveform is
g p (t ) = cn e j 2 nf ot
closest in value to ______ .
n=
[EC-2014 : 1 Mark]
It is given that, c3 = 3 + j5. Then c–3 is
Q.15 The magnitude and phase of the complex
(a) 5 + j3 (b) –3 – j5
Fourier series coefficient ak of a periodic signal
(c) –5 + j3 (d) 3 – j5
x(t) are shown in the figure. Choose the correct
[EC-2003 : 1 Mark] statement from the four choices given.
Q.11 Choose the function f(t), – < t < , for which a Notation: C is the set of complex, number, R is
Fourier series cannot be defined. the set of purely real numbers, and P is the set of
(a) 3 sin(25t) purely imaginary numbers.
(b) 4 cos(20t + 3) + 2 sin(710t) ak
3 3
(c) exp ( t ) sin (25t ) 2 2

(d) 1 1

[EC-2005 : 1 Mark] k
–5 –4 –3 –2 –1 0 1 2 3 4
 32 Electronics Engineering Signals & Systems

ak Q.18 Let x(t) be a periodic function with period T = 10.

–5 –4 –3 –2 –1 1 2 3 4 The Fourier series coefficients for this series are
k
0 denoted by ak, that is
2
– jk t
x( t ) = ak e T
(a) x(t) R k=
(b) x(t) P The same function x(t) can also be considered
(c) x(t) (C – R) as a periodic function with period T = 40. Let bk
(d) the information given is not sufficient to be the Fourier series coefficients when period is
draw any conclusion about x(t). taken as T .
[EC-2015 : 1 Mark]
If ak = 16 , then bk is equal to
Q.16 A periodic signal x(t) has a trigonometric Fourier k= k=
series expansion:
(a) 256 (b) 64
x(t ) = a0 + ( an cos n o t + bn sin n ot )
(c) 16 (d) 4
n=1 [EC-2018 : 1 Mark]
If x(t) = –x(–t) = –x(t – / o), we can conclude
Q.19 The exponential Fourier series representation
that
of a continuous-time period signal x(t) is defined
(a) an are zero for all n and bn are zero for n
as,
even.
(b) an are zero for all n and bn are zero for n odd. ak e jk ot
x(t) =
(c) an are zero for n even and bn are zero for n k=
odd. where, o is the fundamental angular frequency
(d) an are zero for n odd and bn are zero for n of x(t) and the coefficients of the series are ak.
even. The following information is given about x(t)
[EC-2017 : 1 Mark] and ak :

Q.17 Let x(t) be a continuous time periodic signal with I. x(t) is real and even, having a fundamental
fundamental period T = 1 seconds. Let {ak} be period of 6.
the complex Fourier series coefficients of x(t), II. The average value of x(t) is 2.
where k is integer valued. Consider the k, 1 k 3
following statements about x(3t): III. ak =
0, k>3
I. The complex Fourier series coefficients of
x(3t) are {ak} where k is integer valued. The average power of the signal x(t) (Rounded
off to one decimal place) is ______ .
II. The complex Fourier series coefficients of
x(3t) are {3ak} where k is integer valued. [EC-2021 : 2 Marks]
III. The fundamental angular frequency of x(3t) Q.20 The frequency response H(f) of a linear time
is 6 rad/s. invariant system has magnitude as shown in
For the three statements above, which one of the the figure.
following is correct?
|H(f )|
(a) Only II and III are true.
1
(b) Only I and III are true.
(c) Only III is true. f (Hz)
– 0
(d) Only I is true. [EC-2017 : 2 Marks]
 GATE Previous Years Solved Paper 33

Statement-I : The system is necessarily a pure Q.3 Fourier series for the waveform, f(t) shown in
delay system for inputs which are band limited figure is
to – f .
f (t )
Statement-II : For any wide sense stationary
input process with power spectral density SX(f),
the output power spectral density SY(f) obeys
–1 1 2 3
SY(f ) = SX(f) for – f . Which one of the t
following combinations is true? –1
(a) Statement I is correct, Statement II is correct.
(b) Statement I is correct, Statement II is
8 1 1
incorrect. (a) 2
sin( t ) + sin(3 t ) + sin(5 t ) + ...
9 25
(c) Statement I is incorrect, Statement II is
correct. 8 1 1
(b) 2
sin( t ) cos(3 t ) + sin(5 t ) + ...
(d) Statement I is incorrect, Statement II is 9 25
incorrect.
8 1 1
[EC-2022] (c) 2
cos( t ) + cos(3 t ) + cos(5 t ) + ...
9 25

ELECTRICAL EN GINEERIN G 8 1 1
(d) 2
cos( t ) sin(3 t ) + sin(5 t ) + ...
(GATE Previous Years Solved Papers) 9 25
[EE-2002 : 1 Mark]
Q.1 A periodic rectangular signal, x(t) has the
waveform as shown in figure. Frequency of the Q.4 The rms value of the periodic waveform given
fifth harmonic of its spectrum is in figure is
x(t)
I

+6 A

T
t
–4 –2 0 2 4 t(ms) T/2

–6 A

(a) 40 Hz (b) 200 Hz

(c) 250 Hz (d) 1250 Hz (a) 2 6 A (b) 6 2 A
[EE-1996 : 1 Mark]
4
Q.2 If an a.c. voltage wave is corrupted with an (c) A (d) 1.5 A
3
arbitrary number of harmonics, then the overall
[EE-2004 : 2 Marks]
voltage waveform differs from its fundamental
frequency component in terms of Q.5 The rms value of the resultant current in a wire
(a) only the peak values which carries a d.c. current of 10 A and a
(b) only the rms values sinusoidal alternating current of peak value 20 A
(c) peak and rms values is

(d) all the three measures (peak, rms and (a) 14.1 A (b) 17.3 A
average values) (c) 22.4 A (d) 30.0 A
[EE-2000 : 1 Mark] [EE-2004 : 2 Marks]
 34 Electronics Engineering Signals & Systems

Q.6 For the triangular wave form shown in the Q.10 Let x(t) be a periodic signal with time period T.
figure, the rms value of the voltage is equal to Let, y(t) = x(t – t0) + x(t + t0) for some t0. The
V(t) Fourier series coefficients of y(t) are denoted by
bk. If bk = 0 for all odd K. Then t0 canbe equal to
1V
T T
(a) (b)
8 4

T/2 T 3T/2 2T
t T
(c) (d) 2T
2
1 1 [EE-2008 : 2 Marks]
(a) V (b) V
6 3
Q.11 The Fourier series coefficients, of a periodic
1 2 signal x(t) expressed as,
(c) V (d) V
3 3
x(t) = k=
ak e j 2 kt /T
[EE-2004 : 2 Marks]
are given by
Q.7 The Fourier series for the function f(x) = sin2x is a –2 = 2 – j1
(a) sinx + sin2x (b) 1 – cos2x a –1 = 0.5 + j0.2
(c) sin2x + cos2x (d) 0.5 – 0.5 cos2x a0 = j2
[EE-2005 : 2 Marks] a1 = 0.5 – j0.2
a2 = 2 + j1
Q.8 x(t) is real valued function of a real variable with
period T. Its trigonometric Fourier series and ak = 0 for k > 2
expansion contains no terms of frequency Which of the following is true?
= 2 (2k)/T, k = 0, 1, 2, ..... Also, no sine terms (a) x(t) has finite energy because only finitely
are present. Then x(t) satisfies the equation many coefficients are non-zero.
(a) x(t) = –x(t – T)
(b) x(t) has zero average value because it is
(b) x(t) = x(T – t) = –x(t) periodic.
T (c) The imaginary part of x(t) is constant.
(c) x(t ) = x(T t ) = x t
2 (d) The real part of x(t) is even.
T [EE-2009 : 2 Marks]
(d) x(t ) = x(t T ) = x t
2
Q.12 The second harmonic component of the periodic
[EE-2006 : 1 Mark]
waveform given in the figure has an amplitude
Q.9 A signal x(t) is given by of
T 3T
1, <t
4 4
x( t ) = +1
3T 7T
1, <t
4 4 t
T/2 T
Which among the following gives the
fundamental Fourier term of x(t)? –1
4 t t
(a) cos (b) cos + (a) 0 (b) 1
T 4 4 2T 4
2
4 t t (c) (d) 5
(c) sin (d) sin
T 4 4 2T 4
[EE-2010 : 1 Mark]
[EE-2007 : 2 Marks]
 GATE Previous Years Solved Paper 35

Q.13 The Fourier expansion: Q.17 The signum function is given by

x
f (t ) = a0 + an cos n t + bn sin n t , x 0
n=1 sgn ( x ) = x
0, x=0
of the periodic signal shown below will contain
the following non-zero terms: The Fourier series expansion of sgn (cos(t)) has
(a) only sine terms with all harmonics.
f (t )
(b) only cosine terms with all harmonics.
(c) only sine terms with even numbered
t harmonics.
0
(d) only cosine terms with odd numbered
harmonics.
(a) a0 and bn, n = 1, 3, 5, ....
[EE-2015 : 2 Marks]
(b) a0 and an, n = 1, 2, 3, ....
(c) a0, an and bn, n = 1, 2, 3, .... Q.18 Let f(x) be a real, periodic function satisfying
f(–x) = –f(x). The general form of its Fourier series
(d) a0 and an, n = 1, 3, 5, ....
representation would be
[EE-2011 : 1 Mark]
(a) f ( x ) = a0 + a
k =1 k
cos( kx )
Q.14 For a periodic signal

v(t ) = 30 sin 100t + 10 cos 300t + 6 sin 500t + , (b) f (x) = b

k =1 k
sin ( kx )
4
the fundamental frequency in rad/sec is (c) f ( x ) = a0 + a
k = 1 2k
cos( kx )
(a) 100 (b) 300
(d) f ( x ) = a
k = 0 2k + 1
sin (2 k + 1) x
(c) 500 (d) 1500
[EE-2013 : 1 Mark] [EE-2018 : 1 Mark]

Q.15 Let, g : [0, ) [0, ) be a function defined by

+
k
g(x) = x – [x], where [x] represents the integer Q.19 Let the signal x(t ) = ( 1)k t be
k= 2000
part of x. (That is, it is the largest integer which
is less than of equal to x.) The value of the
passed through an LTI system with frequency
constant term in the Fourier series expansion of
response H( ), as given in the figure.
g(x) is _______ .
H( )
[EE-2014 : 2 Marks]
1
Q.16 For a periodic square wave, which one of the
following statements is true?
(a) The Fourier series coefficients do not exist.
–5000 5000
(b) The Fourier series coefficient exist but the
The Fourier series representation of the output
reconstruction converges at most point. is given as,
(c) The Fourier series coefficient exist and the (a) 4000 + 4000 cos(2000 t) + 4000 cos(4000 t)
reconstruction converges at no points.
(b) 2000 + 2000 cos(2000 t) + 2000 cos(4000 t)
(d) The Fourier series coefficient exist and the
(c) 4000 cos(2000 t)
reconstruction converges at every point.
(d) 2000 cos(2000 t)
[EE-2014 : 1 Mark]
[EE-2017 : 2 Marks]
 36 Electronics Engineering Signals & Systems

t t , t 0 If, f (t ) = a0 + an cos nt + bn sin nt

Q.20 Consider g(t ) = , where
t t , otherwise n=1 n=1

A sin t , 0 t
t R. Here t represents the largest integer less f (t ) =
0, <t<2
than or equal to t and t denotes the smallest The Fourier series coefficients a1 and b1 of f(t)
integer greater than or equal to t. The coefficient are:
of the second harmonic component of the A A
(a) a1 = 0 ; b1 = (b) a1 = ; b1 = 0
Fourier series representing g(t) is _______ .
[EE-2017 : 1 Mark] A A
(c) a1 = 0 ; b1 = (d) a1 = ; b1 = 0
2 2
Q.21 A periodic function f(t), with a period of 2 , is
[EE-2019 : 2 Marks]
represented as its Fourier series.

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Fourier Series

1. (c) 2. (Sol.) 3. (d) 4. (c) 5. (c) 6. (c) 7. (d) 8. (c)

9. (b) 10. (d) 11. (c) 12. (a) 13. (c) 14. (0.51) 15. (a) 16. (a)

17. (b) 18. (c) 19. (32) 20. (a)

Solutions
EC Fourier Series

1. (c) Time period = To

V(t) 1
fo = = fundamental frequency
To
Vm
= 1st harmonics
t Fourier series of v(t) is given by
–T0 –T0/4 T0/4 T0
v(t) = a0 + a1 cos( ot) + a2 cos(2 ot) + ....
To a 0 = dc value = average value
V(t) = Vm cos( ot) t
4
 GATE Previous Years Solved Paper 37

1 2. Sol.
a0 = v(t ) dt
To f(x)
To

To /4
1
= Vm cos( o t ) dt
To
To /4 x
–
To /4
1 Vm [sin o t ] To /4 Time period = T = – ( ) = 2
a0 = At the point of discontinuity,
To o
x =
Vm 2 To 2 To
a0 = sin sin f(x) expressed in Fourier series converge to the
2 To 4 To 4
To media value = /2. From the given trigonometric
To
form of Fourier series.
Vm
a0 = sin sin At, x =
2 2 2
2
f( ) = + s=
Vm V 4 2
= [1 ( 1)] = m × 2
2 2 1 1
where, s = 1+ 2
+ + ....
Vm 3 52
a0 =
2
s = =
Vm 10 2 4 4
So, average value = a0 = = V
2
s=
2 8
a1 = v(t ) cos( o t ) dt
To
To 3. (d)
To /4 The Fourier series of an odd periodic function
2
a1 = Vm cos( o t ) cos( o t ) dt does not contain the dc term (ao = 0) and cosine
To
To /4 terms. The Fourier series of an odd periodic
function contains only same terms.
To /4
2
a1 = Vm cos 2 ( o t ) dt 4. (c)
To
To /4
The trigonometric Fourier series of an even
To /4 function of time contains only dc term and
2Vm (1 + cos 2 ot)
a1 = dt cosine term. The trigonometric Fourier series of
To 2
To /4 an even function of time does not contain sine
terms.
To /4 To /4
2Vm
a1 = dt + cos 2 ot dt 5. (c)
To
To /4 To /4
The trigonometric Fourier series of a periodic
Vm To To V T time function can have only cosine and sine
a1 = × = m× o terms.
To 4 4 To 2

Vm 6. (c)
a1 =
2 x( t)
Vm = 10
10 1
So, a1 = =5V
2 t
–To/2 –T1 T1 To/2
38 Electronics Engineering Signals & Systems

Given periodic signal x(t) has time period To , Again integrating,

To /2
1
xdc = x(t ) dt
To
To /2

T
1 1 1 T1
= 1 dt = [t ] T 1
To To 1
T1 Cn
n2
1 2T
xdc = [Tt1 ( T1 )] = 1 9. (b)
To To
x(t) = 2 cos t + 7 cost
7. (d) 2
T1 = =2

s(t) = c k e jk ot
2
k= T2 = =2
1
To /2 T1 1
1 jk ot 1 = = irrational
ck = (t ) e dt = T2
To To
To /2
x(t) is not periodic and does not satisfy
1 Dirchlet condition.
s(t) = e jk ot
To k= 10. (d)

8. (c) c–k = c k
For impulse train the Fourier series coefficient c3 = 3 + j5
are: c–3 = c 3 = 3 j5

11. (c)
All other functions are either periodic or
constant function.

12. (a)
1 The Fourier series of a real periodic function
Cn =
Ts
has only cosine terms if it is even and only sine
terms if it is odd.

13. (c)
Trigonometric Fourier series of an even function
has dc and cosine terms only.

14. Sol.
A sin n o /2
Cn = x(t )
T n o /2
1
2A
= sin n o
Tn o 2
t
0 1 2 3 4
1
Cn
n
–1
 GATE Previous Years Solved Paper 39

From Parseval’s theorem,

X( ) = 2 ak ( k ) T +
k= 1 2 2
Pav = x(t ) dt = ak
T
1 0 k=
For rectangle pulse, ak
k 3
2
7th harmonic power = (2 a7)2 = ak
k= 3
5th harmonic power = (2 a5)2
Ratio = 3
2 2
2 = a0 +2 ak
2
a7 5 25 k =1
= = = 0.51
a5 7 49 = (2)2 + 2[(1)2 + (2)2 + (3)2]
= 4 + 28 = 32 Watts
15. (a)
20. (a)
ak – even symmetry
ak – odd symmetry ( can be – ) x(t ) Delay y(t ) = x(t – Td)
Td
x(t) is real.

16. (a) Y(f) = X ( f ) e j 2 fTd

Signal has odd and half wave symmetries.
Y( f ) j 2 fTd
So all an are zero and bn are zero for n even. H(f) = =e
X( f )
17. (b)
H( f ) = 1
Initially, T = 1 sec, so o = 2 rad/sec.
When x(t) is compressed by 3, frequency will H(f) = –2fT d
expand by same factor but there is no change in Given that, input is band limited to – f .
values of ak. Magnitude response of the system given,
So, both statement I and II are correct. |H(f )|

18. (c)
1
Change in only time period or frequency does
not change in the value of Fourier series f (Hz)
– 0
coefficients.
So, bk = ak Based on the magnitude response given,
statement (I) is correct.
bk = ak = 16 Note : Phase response of the system is not given
k= k= in the question without knowing phase
response of the system we can not comment
19. (32)
about exact nature of the system,
From II, a0 = 2
2
From III, a 1 = 1, a2 = 2, a3 = 3 (PSD)output = ( PSD)input H ( f )
From I, since x(t) is real and even ak = a–k SY(f) = SX(f ); – f
a 1 = a–1 = 1 Statement (II) is correct).
and a 2 = a–2 = 2 Hence option (a) is correct.
and a 3 = a–3 = 3
40 Electronics Engineering Signals & Systems

Answers
EE Fourier Series

1. (d) 2. (c) 3. (c) 4. (a) 5. (b) 6. (a) 7. (d) 8. (c)

9. (a) 10. (b) 11. (c) 12. (a) 13. (d) 14. (a) 15. (0.5) 16. (b)

17. (d) 18. (b) 19. (c) 21. (c)

Solutions
EE Fourier Series

1. (d) 5. (b)
1 Rms value of d.c. current = 10 A
Q Fundamental frequency = Hz
4 × 10 3 Rms value of alternating current

5 20
= = 10 2 A
Frequency of 5th harmonic = 3 2
4 × 10
= 1250 Hz Rms value of resultant current

= 102 + (10 2 )2 A
3. (c)
Q f(t) is an even function with half wave = 17.32 A
symmetry. 6. (a)
d.c. term as well as sine terms = 0
From the wave symmetry,
Only the cosine terms with odd harmonics will
1/2
be present. T /2 T
1 2t 2
Option (c) is correct. Vrms = dt + 0 dt
T T
0 T /2
4. (a)
T /2
1 /2 2 1 4t 2 4 T3 1
1
T Vrms = = =
Irms = i 2 (t ) T
0 T2 3T 3 8 6
T
0
1
T T
Vrms = V
2 6
1 12t
2
I rms = dt + 6 2 dt
T T
0 T /2 7. (d)
f(n) = sin 2x
T /2 2 T
1 144(t ) 36 For finding the Fourier series expansion,
= 2
dt + dt
T T T
0 T /2 2 1 cos 2 x
f(x) = sin x =
2
T /2
144 t 3 36 T = 0.5 – 0.5 cos2x
= +
T3 3 0
T 2
8. (c)
144 36 Since trigonometric Fourier series has no sine
= + = 24
8× 3 2 terms and has only cosine terms therefore this
will be an even signal i.e. it will satisfy,
Irms = 2 6 A
x(t) = x(–t)
 GATE Previous Years Solved Paper 41

or we can write, 1
C1 = [2 e j /4 + 2 e j /4 ]
x(t – T) = x(–t + T) j2
But signal is periodic with period T.
2 j /4 2 j /4 2 1 j
Therefore, C1 = e = e =
j 2 2
x(t – T) = x(t)
Therefore, Comparing it with
x(t) = x(T – t) ...(i) a1 jb1
C1 =
Now since signal contains only odd harmonics 2 2
i.e. no terms of frequency, 4 4
a1 = , b1 =
2 ×2k 2 2
=
; k = 0, 1, 2, 3, 4
T Fundamental Fourier term will be
i.e. no even harmonics. = a1 cos 0t + b1 sin 0t
This means signal contains half wave symmetry 4 4
this implies that, = cos t + sin t
2 T 2 T
T
x(t) = x t ...(ii) 4
2 = cos t
T 4
From equations (i) add (ii),
T 10. (b)
x(t) = x(T t ) = x t
2 y(t) = x(t – t0) + x(t + x0)
9. (a) Since, x(t) is periodic with period T.
Therefore, x(t – t0) and x(t + t0) will also be
According to definition of signal given in
periodic with period T.
question the x(t) will be as,
jk 0t0
bk = ak e + ak e jk 0t0
x(t )
(By property)
1
ak is Fourier series coefficient of signal x(t).

Therefore, bk = ak [ e jk 0t0 + e jk 0t0 ]

t
–T/4 3T/4 7T/4
bk = 2 ak cosk 0t0
since, bk = 0 for odd k
–1

i.e., k 0t0 = k
So, it is periodic with period, 2
where, k = ±1, ±3, ±5....
7T T
T0 = = 2T
4 4
0t0 =
2
Fundamental angular frequency,
2 2 T T
= = t0 = ×
=
0 = 2 2 4
T0 2T T
Now, 11. (c)
C1 = (exponential series coefficient for k = 1)
j 2 0t
x(t) = a 2 e + a 1 e j 0t + a0
1 j 0t
C1 = x (t ) e dt
T0
T0
+ a1 e j 0t + a2 e j 2 0t

j 2 0t
1 3T /4 j 0t 7T /4 j 0t
= (2 j) e + (2 + j ) e j 2 0t
C1 = e dt + 1e dt
T0 T /4 3T /4
+ (0.5 + j 0.2) e j 0t + (0.5 j0.2) e j 0t + 2 j
42 Electronics Engineering Signals & Systems

j 2 0t
+ e j 2 0t ) + j [ e j 2 0t e j 2 0t ] Then function g(x) will be
= 2 (e
+ 0.5( e j 0t +e j 0t
) j 0.2 ( e j 0t e j 0t
)+2j g (x )

= 4 cos2 0t – 2 sin2 0t + cos 0t

1
+ 0.4 sin 0t + 2j
= Real [x(t)] + j Img[x(t)]
where, x
–1 0 1 2 3
Real [x(t)] = 4 cos2 0t – 2 sin2 0t + cos 0t
+ 0.4 sin 0t The value of the constant term (or) dc term in
= neither even nor odd signal the Fourier series expansion of g(x) is
Img [x(t)] = 2 = constant T
1
a0 = f ( x ) dx
12. (a) T
0
The given signal is odd as well as having half Area in one period
wave symmetry. =
one period
So it has only sine terms with odd harmonics.
1
So for second harmonic term amplitude = 0. = × 1 × 1 = 0.5
2
13. (d)
16. (b)
Let, x(t) = Even and Hws
x(t)

1 t
t
1

T/2 –T/2 0 T/2 T

0 T
Reconstruction of signal by its Fourier series
T/4 3T/4
coefficient is not possible at those points where
Fs expansion of x(t) contains cos terms with odd signal is discontinuous.
harmonics. In the above figure, at integer multiples of ‘T/2’,
f (t )
signal recovery is not possible by using its
coefficient.
Therefore, reconstruction of x(t) by using its
2 coefficient is possible at most of the points except
t
those instants where x(t) is discontinuous.
Now, f(t) = 1 + x(t) 17. (d)
Fs of f(t) contains dc and cos terms with odd
Sgn(x)
harmonics.

15. Sol.
Given function, x
g(x) = x – [x]
where [x] is a integer part of x.
 GATE Previous Years Solved Paper 43

So, cos(t) is Since, x(t) is even-half wave symmetric. So, its

expansion will contain only odd harmonics of
cos(t)
cos. Therefore, coefficient of fundamental
1 harmonic is
T0 /2
t 2
a1 = x(t ) cos 0t dt
T
T0 /2
–1
Sgn cos(t)
T /2
1 4 0
= (t ) cos 0 t dt
T0
0
T0 /2
t 4
= (t ) cos 0 dt
T0
–1 0

T0 /2
4 4
So, sgn (cost) is a rectangular signal which is = (t ) dt = = 4000
T0 T0
even and has half wave symmetry. 0
So, Fourier series will have only cosine terms Now, frequency components available in
with odd harmonics only. expansion are:
0, 3 0,.....
18. (b)
2000 , 6000 , .....
Given that, f(–x) = –f(x) As, LTI system given in the question will pass
So, function in an odd function. upto 5000 rad/sec frequency component of
So the Fourier series will have sine term only so, input.
So, output will have only one component of
f(x) = bx sin( kx )
frequency 2000 rad/sec.
k=1
Thus, y(t) = expansion of output
19. (c) = a1 cos 0t
= 4000 cos2000 t
K
x(t) = ( 1)K t
k= 2000 20. Sol.

t t , t 0
K Given that, g(t) =
= ( 1) (t KT ) t t , otherwise
k=
where,
1
where, T= t = greatest integer less than or equal to ‘t’.
2000
x(t) t = smallest integer greater than or equal to ‘t’.
Now,
–T T 3T 1
0 2T t 0 t<0
1
g(t ) t t g(t ) t t
Time period: t–0 0, 1 0 t+0 –1, 0 0

1 t–1 1, 2 1 t+1 –2, –1 –1

T0 = 2T = t–2 2, 3 2 t+2 –3, –2 –2
1000
t–3 3, 4 3 t+3 –4, –3 –3
2
0 = = 2000 rad
T0
44 Electronics Engineering Signals & Systems

g(t) A A
= 2 2 sin 2 t dt = [1 cos 2t ] dt
2
1 2 3 0 0
– – –

t
–4 –3 –2 –1
t A sin 2t
0 4 3 2 1 = t
+
3
+
2
+
1 2 2 0
t
t

A A
Since, g(t) is non-periodic. So there is no Fourier = [( 0) (0 0)] =
2 2
series expansion of this signal and hence no
T
need to calculate harmonic here. 2 0
a1 = f (t ) cos 0 t dt
T0
0
21. (c)
2
2 2
T0 = 2 0 = =1 = f (t ) cos t dt
T0 2
0
T0
2 1
Now, b1 = f (t ) sin 0t dt = A sin t cos t dt
T0
0 0

2
2 A
= f (t ) sin t dt [Q = sin 2t dt = 0
2 0 = 1] 2
0 0

1
= A sin t sin dt
0
 4 Fourier Transforms, Frequency
Response and Correlation
ELECTRO NICS EN GINEERIN G List-II

(GATE Previous Years Solved Papers) 1. Gaussian function

2. Rectangular pulse
Q.1 Specify the filter type if its voltage transfer 3. Triangular pulse
function H(s) is given by 4. Ramp function
2
K(s + 2 5. Zero
o)
H (s) =
[EC-1995 : 2 Marks]
s2 + o s + 2o
Q
Q.5 The Fourier transform of a real valued time
(a) all pass filter (b) low pass filter signal has
(c) band pass filter (d) notch filter (a) odd symmetry
[EC-1988 : 2 Marks] (b) even symmetry
(c) conjugate symmetry
Q.2 The magnitude and phase functions for a
distortionless filter should respectively be (d) no symmetry

(Magnitude) (Phase) [EC-1996 : 1 Mark]

(a) Linear Constant Q.6 The function f(t) has the Fourier transform g( ).
(b) Constant Constant
(c) Constant Linear The Fourier transform of g(t ) = g(t ) e j t
dt
(d) Linear Linear
[EC-1990 : 2 Marks] is
Q.3 If G(f ) represents the Fourier’s transform of a 1 1
(a) f( ) (b) f( )
signal g(t) which is real and odd symmetric in 2 2
time, then (c) 2 f (– ) (d) none of these
(a) G(f ) is complex. [EC-1997 : 1 Mark]
(b) G(f ) is imaginary.
Q.7 If the Fourier transform of a deterministic signal
(c) G(f ) is real.
g(t) is G(f ), then
(d) G(f ) is real and non-negative.
1. the Fourier transform of g(t – 2) is
[EC-1992 : 2 Marks]
2. The Fourier transform of g(t/2) is
Q.4 Match each of the items, A, B and C with an (a) G(f) e–j(4 f) (b) G(2f )
appropriate item from 1, 2, 3, 4 and 5. (c) 2G(2f) (d) G(f – 2)
List-I [EC-1997 : 2 Marks]
A. Fourier transform of a Gaussian function
Q.8 The amplitude spectrum of a Gaussian pulse is
B. Convolution of a rectangular pulse with
(a) uniform (b) a sine function
itself
(c) Gaussian (d) an impulse function
C. Current through an inductor for a step input
[EC-1998 : 1 Mark]
voltage
 46 Electronics Engineering Signals & Systems

Q.9 The Fourier transform of a function x(t) is X(f ). Q.14 A linear phase channel with phase delay Tp and
The Fourier transform of dx(t)/dt will be group delay Tg must have
dX( f ) (a) Tp = Tg = constant
(a) (b) j2 fX(f)
dt (b) Tp f and Tg f
X( f ) (c) Tp = constant and Tg f
(c) jfX(f ) (d)
jf (d) Tp f and tg = constant (f denotes frequency)
[EC-1998 : 1 Mark] [EC-2002 : 1 Mark]

Q.10 The Fourier transform of a voltage signal x(t) is Q.15 The Fourier transform F{e–t u(t)} is equal to

X(f ). The unit of X( f ) is 1 1

. Therefore, F is
(a) Volt (b) Volt-sec 1 + j2 f 1 + j2 t
(c) Volt/sec (d) Volt2
(a) e f u(f ) (b) e –f u(f )
[EC-1998 : 1 Mark]
(c) e f u(–f ) (d) e –f u(–f )
Q.11 A signal x(t) has a Fourier transform X( ). If x(t) [EC-2002 : 1 Mark]
is a real and odd function of t, then X( ) is
Q.16 Let x(t) be the input to a linear, time-invariant
(a) a real and even function of .
system. The required output is 4x(t – 2). The
(b) an imaginary and odd function of .
transfer function of the system should be
(c) an imaginary and even function of .
(a) 4 e j4 f (b) 2e –j8 f
(d) a real and odd function of .
(c) 4 e –j4 f (d) 2e j8 f
[EC-1999 : 1 Mark]
[EC-2003 : 1 Mark]
Q.12 The input to a channel is a bandpass signal. It
is obtained by linearly modulating a sinusoidal Data given below for two questions. Solve the problem
carrier with a single-tone signal. The output of and choose the correct answer.
the channel due to this input is given by The system under consideration is an RC low-pass filter
1 (RC-LPF) with R = 1.0 k and C = 1.0 µF.
y (t ) = cos(100t 10 6 ) cos(106 t 1.56)
100 Q.17 Let H(f ) denote the frequency response of the
The group delay (tg) and the phase delay (tp) in RC-LPF. Let f1 be the highest frequency such
seconds, of the channel are
H( f1 )
(a) tg = 10–6, tp = 1.56 that, 0 f f1 ; 0.95. Then f1 (in Hz)
H (0)
(b) tg = 1.56, tp = 10–6
(c) tg = 10–8, tp = 1.56 × 10–6 is
(d) tg = 108, tp = 1.56 (a) 327.8 (b) 163.9
[EC-1999 : 1 Mark] (c) 52.2 (d) 104.4
[EC-2003 : 2 Marks]
3t 2
Q.13 The Fourier transform of the signal x(t ) = e
Q.18 Let tg(f) be the group delay function of the given
is of the following form, where A and B are RC-LPF and f2 = 100 Hz. Then tg(f2) (in ms), is
constants: (a) 0.717 (b) 7.17
Bf 2 Bf 2
(a) Ae (b) Ae (c) 71.7 (d) 4.505
2 [EC-2003 : 2 Marks]
(c) A + B f (d) Ae–Bf
[EC-2000 : 1 Mark]
 GATE Previous Years Solved Paper 47

Q.19 The Fourier transform of a conjugate symmetric Q.23 The output y(t) of a linear time invariant system
function is always is related to its input x(t) by the following
(a) imaginary equation:
(b) conjugate antisymmetric y(t) = 0.5x(t – td + T) + x(t – td) + 0.5x(t – td – T)
(c) real The filter transfer function H( ) of such a system
(d) conjugate symmetric is given by
[EC-2004 : 1 Mark] j td
(a) (1 + cos T ) e
Q.20 A rectangular pulse train s(t) as shown in the (b) (1 + 0.5 cos T ) e j td
figure is convolved with the singal
cos2(4 × 103t). The convolved signal will be a (c) (1 cos T ) e j td

s (t ) (d) (1 0.5 cos T ) e j td

1 [EC-2005 : 2 Marks]

Q.24 Let x(t) X(j ) be Fourier transform pair. The

t Fourier transform of the signal x(5t – 3) in terms
0
0.1 msec of X(j ) is given as,
(a) DC (b) 12 kHz sinusoid j3 j3
1 5
j 1 5 j
(c) 8 kHz sinusoid (d) 14 kHz sinusoid (a) e X (b) e X
5 5 5 5
[EC-2004 : 2 Marks]
1 j3 j 1 j3 j
Q.21 Let x(t) and y(t) (with Fourier transforms X(f) (c) e X (d) e X
5 5 5 5
and Y(f) respectively) be related as shown in
[EC-2006 : 1 Mark]
the figure. Then Y(f) is
y( t)
Q.25 The 3-dB bandwidth of the low-pass signal
x (t ) e–t u(t), where u(t) is the unit step function, is
1
–2 –1 0 given by
t
1 1
(a) Hz (b) 2 1 Hz
2 2
(c) (d) 1 Hz
t –1
–2 0 2 [EC-2007 : 2 Marks]

1 f 1 f Q.26 The signal x(t) is described by

(a) X e j 2 f (b) X e j2 f
2 2 2 2 1 for 1 t +1
x( t ) =
f f j2 f 0 otherwise
(c) X e j2 f
(d) X e
2 2
Two of the angular frequencies at which its
[EC-2004 : 2 Marks] Fourier transform becomes zero are
Q.22 For a signal x(t) the Fourier transform is X(f). (a) ,2 (b) 0.5 , 1.5
Then the inverse Fourier transform of X(3f + 2) (c) 0, (d) 2 , 2.5
is given by [EC-2008 : 2 Marks]
1 t j3 t 1 t j 4 t /3 Statement for Linked Answer Questions (27 and 28):
(a) x e (b) x e
2 2 3 3
The impulse response h(t) of a linear time-invariant
(c) 3x(3t) e–j4 t (d) x(3t + 2) continuous time system is given by h(t) = exp(–2t) u(t),
[EC-2005 : 2 Marks] where u(t) denotes the unit step function.
 48 Electronics Engineering Signals & Systems

Q.27 The frequency response H( ) of this system in Causal : The system is causal.
terms of angular frequency , is given by H( ) = LP : The system is low pass.
1 LTI : The system is linear and time-invariant.
sin( )
(a) (b) (a) Causal, LP
1 + j2
(b) BIBO, LTI
1 j
(c) (d) (c) BIBO, Causal LTI
2+ j 2+ j
(d) LP, LTI [EC-2009 : 2 Marks]
[EC-2008 : 2 Marks]
Q.31 The Fourier transform of a signal
Q.28 The output of this system, to the sinusoidal input
h(t) is H(j ) = (2 cos ) (sin2 )/
x(t) = 2 cos(2t) for all time t, is
The value of h(0) is
(a) 0
(b) 2–0.25 cos(2t – 0.125 ) 1 1
(a) (b)
4 2
(c) 2–0.5 cos(2t – 0.125 )
(c) 1 (d) 2
(d) 2–0.5 cos(2t – 0.25 )
[EC-2012 : 2 Marks]
[EC-2008 : 2 Marks]
Q.32 A continuous, linear time-invariant filter has
Q.29 A function is given by
an impulse response h(t) described by
f(t) = sin2t + cos2t
Which of the following is true? 3 for 0 t 3
h( t ) =
(a) f has frequency components at 0 and 0 otherwise
1/2 Hz. When a constant input of value 5 is applied to
(b) f has frequency components at 0 and 1/ Hz. this filter, the steady-state output is ______ .
(c) f has frequency components at 1/2 and [EC-2014 : 1 Mark]
1/ Hz.
Q.33 For a function g(t), it is given that,
(d) f has frequency components at 0, 1/2
+
and 1/ Hz. j t 2 2
g(t ) e dt = e
[EC-2009 : 1 Mark]

Q.30 Consider a system whose input and output y t

for any real value . If y(t ) = g(t ) d , then
are related by the equation,

+
y(t ) = x(t ) h(2 ) d
y(t ) dt _______ .

h (t ) (a) 0 (b) –j
j j
(c) (d)
2 2
t
0 [EC-2014 : 2 Marks]

Q.34 The value of the integral sin c 2 (5t ) dt is

where h(t) is shown in the graph.
Which of the following four properties are
possessed by the system? BIBO : Bounded input ________ .
gives a bounded output. [EC-2014 : 2 Marks]
 GATE Previous Years Solved Paper 49

Q.35 The phase response of a passband waveform at 3

the receiver is given by (c) ak cos( k ot + k )
k=1
(f) = –2 (f – fc) – 2 fc
where fc is the center frequency, and and are 2
positive constants. The actual signal (d) ak cos( k ot + k )
k=1
propagation delay from the transmitter to
receiver is [EC-2016 : 2 Marks]

(a) (b) sin(t ) sin(t )

+ + Q.39 If the signal x(t ) = with
t t
(c) (d)
denoting the convolution operation, then x(t) is
[EC-2014 : 1 Mark]
equal to
Q.36 A real-valued signal x(t) limited to the frequency sin(t ) sin(2t )
(a) (b)
W t 2 t
band f is passed through a linear time
2 2
2 sin(t ) sin(t )
(c) (d)
invariant system whose frequency response is t t
W [EC-2016 : 1 Mark]
e j4 f , f
2
H( f ) = Q.40 A continuous time signal x(t) = 4 cos(200 t) +
W
0, f >
2 8 cos(400 t), where ‘t’ is in seconds, is the input
The output of the system is to a linear time invariant (LTI) filter with the
impulse response,
(a) x(t + 4) (b) x(t – 4)
(c) x(t + 2) (b) x(t – 2) 2 sin (300 t )
, t 0
[EC-2014 : 1 Mark] h (t ) = t
600 , t=0
sin(4 t ) Let y(t) be the output of this filter. The maximum
Q.37 The energy of the signal x(t ) = is ___ .
4 t
value of y(t ) is ________ .
[EC-2016 : 1 Mark]
[EC-2017 : 2 Marks]
Q.38 A network consisting of a finite number of linear
resistor (R), inductor (L), and capacitor (C) Q.41 Consider an LTI system with magnitude
elements, connected all in series of all in response,
parallel, is excited with a source of the form
f
3 1 , f 20
where, ak 0, 0 H( f ) = 20
ak cos( k ot ) o
k =1 0, f > 20
The source has non-zero impedance. Which one and phase response,
of the following is a possible form of the output arg {H(f )} = –2f
measured across a resistor in the network? If the input to the system is
3
x(t ) = 8 cos 20 t + + 16 sin 40 t + + 24 cos 80 t +
(a) bk cos( k ot + k ) where, bk ak, k 4 8 16
k=1
then the average power of the output signal y(t)
3 is _______ .
(b) bk cos( k ot + k ) where, bk 0, k
k=1
[EC-2017 : 2 Marks]
 50 Electronics Engineering Signals & Systems

Q.42 The input 4 sinc(2t) is fed to a Hilbert ELECTRICAL EN GINEERIN G

transformer to obtain y(t), as shown in the figure
(GATE Previous Years Solved Papers)
below:
sin ( x ) 1
Here, sin c( x ) = Q.1 Let, x(t ) = rect t (where, rect(t) = 1 for
x 2
The value (accurate to two decimal places) of
1 1
x and zero otherwise). Then
2 2 2
y(t ) dt is ______ .
sin( x )
sin c( x ) = , the Fourier transform of
Hilbert y (t )
x
4 sinc(2t)
transform x(t) + x(–t) will be given by
[EC-2018 : 2 Marks]
(a) sin c
2
Q.43 X( ) is the Fourier transform of x(t) shown
(b) 2 sin c
2 2
below. The value of X( ) d (rounded off
(c) 2 sin c cos
2 2
to two decimal places) is _______ .

(d) sin c sin [EE-2008 : 2 Marks]

x (t ) 2 2
3
Q.2 A signal x(t) = sinc( t) where is a real constant
2
1 sin( x )
sin c( x ) = is the input to a linear time
t x
–3 –2 –1 0 1 2 3 4
invariant system, whose impulse response
[EC-2020 : 2 Marks] h(t) = sinc( t), where is a real constant. If min
( , ) denotes the minimum of and and
Q.44 Consider a real valued base band signal x(t),
similarly max( , ) denotes the maximum of
band limited to 10 kHz. The Nyquist rate for the
and , and K is a constant, which one of the
t following statements is true about the output of
signal, y(t ) = x(t ) x 1 + is
2 the system?
(a) It will be of the form K sinc ( t) where
(a) 20 kHz (b) 30 kHz
= min ( , ).
(c) 15 kHz (d) 60 kHz
(b) It will be of the K sinc( t) where = max( , ).
[EC-2021 : 1 Mark]
(c) It will be of the form K sinc( t).
Q.45 The Fourier transform X(j ) of the signal, (d) It cannot be a sinc type of signal.
[EE-2008 : 2 Marks]
t
x( t ) = is ________ .
(1 + t 2 )2 Q.3 x(t) is a positive rectangular pulse from t = –1 to
t = +1 with unit height. The value of
(a) e (b) e
2j 2 2
X( ) d {where X( ) is the Fourier
(c) e (d) e [EC-2022]
2j 2 transform of x(t)} is
 GATE Previous Years Solved Paper 51

(a) 2 (b) 2 (c) g(t) would be proportional to f(t) only if f(t)

(c) 4 (d) 4 is a sinusoidal function.
[EE-2010 : 2 Marks] (d) g(t) would never be proportional to f(t).
[EE-2014 : 2 Marks]
Q.4 The Fourier transform of a signal h(t) is H(j ) =
(2 cos ) (sin2 )/ . The value of h(0) is Q.7 A function f(t) is shown in the figure.

1 1 f(t)
(a) (b)
4 2 1/2
(c) 1 (d) 2
[EE-2012 : 2 Marks] –T/2
t
–T/2 0
Q.5 A continuous time LTI system with system
function H( ) has the following pole-zero plot. –1/2
For this system, which of the alternatives is true?
The Fourier transform F( ) of f(t) is
(a) real and even function of .
2 (b) real and odd function of .
(c) imaginary and odd function of .
1 (d) imaginary and even function of .
(0, 0) [EE-2014 : 1 Mark]

Q.8 A signal is represented by

1, t <1
x( t ) =
0, t >1
(a) H (0) > H ( ) ; >0 The Fourier transform of the convolved signal
(b) H ( ) has multiple maxima, at 1 and 2 t
y(t ) = x(2t ) x is
2
(c) H (0) < H ( ) ; >0

(d) H ( ) = constant ; – < < 4

(a) 2
sin sin (2 )
2
[EE-2014 : 2 Marks]
4
Q.6 Let f (t) be a continuous time signal and let F( ) (b) 2
sin
2
be its Fourier transform defined by
4
(c) 2
sin (2 )
j t
F( ) = f (t ) e dt
4
(d) 2
sin 2
jut
and g(t) = F (u) e du [EE-2014 : 1 Mark]

Q.9 A differentiable non-constant real even function

What is the relationship between f(t) and g(t)?
x(t) has a derivative y(t), and their respective
(a) g(t) would always be proportional to f(t).
Fourier transforms are X( ) and Y( ). Which of
(b) g(t) would be proportional to f(t) if f(t) is an
the following statements is true?
even function.
 52 Electronics Engineering Signals & Systems

(a) X( ) and Y( ) are both real. Which one of the following statements is true?
(b) X( ) is real and Y( ) is imaginary. (a) x1(t) and x2(t) are complex and x1(t), x2(t) is
(c) X( ) and Y( ) are both imaginary. also complex with non-zero imaginary part.
(d) X( ) is imaginary and Y( ) is real. (b) x1(t) and x2(t) are real and x1(t), x2(t) is also
[EE-2014 : 2 Marks] real.
(c) x1(t) and x2(t) are complex but x1(t), x2(t) is
Q.10 Consider a signal defined by
real.
e j 10t for t t (d) x1(t) and x2(t) are imaginary but x1(t), x2(t)
x(t ) =
0 for t > 1 is real.

Its Fourier transform is [EE-2016 : 2 Marks]

2 sin( 10) sin( 10) Q.12 Suppose the maximum frequency in a brand-
(a) (b) 2 e j 10
10 10 limited signal x(t) is 5 kHz. Then, the maximum
2 sin j 10 2 sin frequency in x(t) cos(2000 t), in kHz is _____ .
(c) (d) e
10 [EE-2016 : 1 Mark]
[EE-2015 : 2 Marks]

Q.11 Suppose x 1 (t) and x 2 (t) have the Fourier sin 2 t

Q.13 The value of the integral 2 dt is
transforms as shown below. t

X1(j ) equal to
(a) 0 (b) 0.5
1
(c) 1 (d) 2

0.5
[EE-2016 : 2 Marks]

0.3
Q.14 The Fourier transform of a continuous time
signal x(t) is given by
1
–1 0 1 2 X( ) = < <
(10 + j )2

X2(j ) where j = 1 and denotes frequency. Then

1 the value of ln x(t ) at t = 1 is ______ .

(upto one decimal place). (ln denoets the
0.5
logarithm to base e).
0.3 [EE-2018 : 2 Marks]

–2 –1 0 2
 GATE Previous Years Solved Paper 53

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Fourier Transforms, Frequency Response and Correlation

1. (d) 2. (c) 3. (b) 4. (A-1, B-3, C-4) 5. (c) 6. (c) 7. (Sol.)

8. (c) 9. (b) 10. (b) 11. (b) 12. (c) 13. (b) 14. (a) 15. (c)

16. (c) 17. (c) 18. (a) 19. (c) 20. (a) 21. (b) 22. (b) 23. (a)

24. (a) 25. (a) 26. (a) 27. (c) 28. (d) 29. (b) 30. (b) 31. (c)

32. (45) 33. (b) 34. (0.2) 35. (c) 36. (d) 37. (0.25) 38. (a) 39. (a)

40. (8) 41. (8) 42. (8) 43. (56.61) 44. (b) 45. (a)

Solutions Fourier Transforms, Frequency Response and Correlation

EC

1. (d) 2. (c)
For distortionless transmission,
K (s 2 + 2o )
Given that, H(s) = H ( f ) = constant
s2 + o s + 2o
Q and H(f ) = –kf where k is a constant
S=j So, the magnitude function of a distortionless
filter should be constant.
2
K (0 + o) The phase function of a distortionless filter
H(0) = 2
=K
0+0+ o should be linear.
H(0) K
3. (b)
2
Fourier transform of real and odd symmetric
Ks 2 1 + o
s signal is imaginary and odd function of
H( ) = 2 frequency.
1
s2 1+ o + 2o
Q s s 4. Sol.
(A) – 1,(B) – 3, (C) – 4
K (1 + 0)
H( ) = =K (A) Fourier transform of a Gaussian function is
1+ 0+ 0
H( ) 0 also a Gaussian function,
Here, the given transfer function is a notch filter. t2 F.T. f2
e e
54 Electronics Engineering Signals & Systems

(B) Convolution of a rectangular pulse with 9. (b)

itself is a triangular pulse.
F.T.
(C) Current through an inductor for a step input x( t ) X( f )
voltage is ramp function, d F.T.
x (t ) ( j 2 f ) X( f )
1 dt
i= V dt
L
1 10. (b)
i= u(t ) dt
L By definition Fourier transform,
1
i = r (t ) j 2 ft
L X(f ) = x(t ) e dt

5. (c)
x(t) Volts
Fourier transform of real valued signal has
So, X(f) Volt-sec
conjugate symmetry,
11. (b)
j t Example:
[X(j ) ] = x( t ) e dt
A sin ot = x(t)
(Real and Odd function)
j( )t
X (j ) = x(t ) e dt x( ) = Aj [ ( + o) – ( – o)]
(Imaginary any odd function of )
X (j ) = X(–j )
12. (c)
6. (c)
1 6
According to the duality property of Fourier y(t) = cos(100t 10 )
100
transform,
cos(106t – 1.56)
F.T.
f (t ) g( ) Comparing with,
F.T. y(t) = cos k(t – tg) cos[ c(t – tp)]
g( t ) 2 f( )
1
y(t) = cos[100(t 10 8 )]
7. Sol. 100
1-A, 2-C cos[106 (t – 1.56 × 10–6)]
F.T. tp = 1.56 × 10–6 sec
g( t ) G( f )
tg = 10–8 sec
F.T.
g(t 2) e j 2 2 f G( f ) = G( f ) e j (4 f )
13. (b)
t F.T. 1 f
g G = 2 G(2 f ) Normalized Gaussian function has Gaussian
2 1/2 1/2
Fourier transform.
8. (c) 14. (a)
A normalized Gaussian pulse is defined as, ( ) = – to
t2 ( )
x(t) = e = to
tp =
t2 F.T. f2
x(t) = e e = X( f )
d ( )
tg = = to
t2 d
So, X(f ) = e is also a Gaussian pulse in
tp = tg = to = constant
frequency domain.
 GATE Previous Years Solved Paper 55

15. (c) 18. (a)

f(t) = e–t u(t) 1
H( ) =
1 1 + j RC
F(f ) =
1 + j2 f ( ) = –tan–1 RC
From duality,
10 3
F.T. tg = 6
F (t ) f( f) 1 + 10 × 4 2 × 10 4
1 = 0.717 ms
F(t) =
1 + j2 t
20. (a)
f(–f) = ef u(–f)
To = 0.1 × 10–3 = 10–4
F.T. f
F (t ) e u( f ) 1
fo = = 10 4 = 10 kHz
To
16. (c)
cos2(4 × 103t) has frequency = 4 kHz
y(t) = 4x(t – 2)
Y(s) = 4e–2s X(s)
Y (s )
= 4e –2s
X( s )
H(f ) = 4e–2xj 2 f = 4e–j4 f –4 kHz 0 4 kHz

17. (c)
R
+ +
0 10 kHz 30 kHz
Vi C Vo
So, only at ‘0’, we get output after convolution.
– – (Only odd harmonics are present).

1
H(f ) =
1 + j 2 fRC
H(0) = 1
H ( f1 ) 1
= 0.95 Constant in time domain.
H (0) 1+ 4 2 2 2 2
f1 R C
1 21. (b)
2 2 0.9025
1+ 4 f 1 ( RC )2 y(t) = –x[2(t + 1)]

x(t – to) j 2 fto

2 2 X( f ) e
1.108 1+ 4 f 1 ( RC )2
to = –1
2 2
0.108 4 f 1 ( RC )2
1 f
x (at) = X
0.108 a a
f 12 2
4 ( RC )2 a = –2
0.329 1 f
f1 Y(f ) = X e j2 f
3 2 2
2 × 10
f1 max = 52.2 Hz
56 Electronics Engineering Signals & Systems

22. (b) X( ) = 0

2 1 t j 4 t /3 ej e j
=0
X 3 f+ = x e
3 3 3 1
ej =0
Applying scaling and shifting property. ej
e2j – 1 = 0
23. (a)
e2j = 1
y(t) = 0.5x(t – td + T) + 0.5x(t – td – T) + x(t – td) ej = ±1
Taking Fourier transform, = ,2
j ( td + T )
Y( ) = 0.5 e + 0.5 e j ( td T ) + e j td X( )
27 (c)
Y( ) j td h(t) = exp(–2t) u(t)
= e 0.5 e j T + 0.5 e j T + 1
X( )
j t
Y( ) H( ) = h (t ) e dt
j td
= e [cos T + 1]
X( )
2t j t (2 + j )t
H( ) = (1 + cos T ) e j td = e e dt = e dt
0 0
24. (a)
1 (2 + j )t 1
x(t) X(j ) = e =
2+ j 2+ j
j3
3 1 j 5
x 5 t = X e 28. (d)
5 5 5
Input, x(t) = 2 cos(2t)
Using scaling and shifting property.
Frequency response,
25. (a) X( ) = 2 [ ( – 2) + ( + 2)]
1 1
Laplace transform of e t u(t ) = H( ) =
2+ j
s+1
1 Output,
Q magnitude at 3-dB frequency = Y( ) = H( ) X( )
2
1 1 1
= = 2 [ ( 2) + ( + 2)]
2 2 2+ j
1+
= 1 rad 2 2
= ( 2) + ( + 2)
2 + j2 2 j2
1
f= Hz
2
= [ ( 2) + ( + 2)] j [ ( 2) ( + 2)]
2 2
26. (a)
cos 2t sin 2t
y(t) = +
2 2
j t
X( ) = x( t ) e dt
2 1 1
= cos 2t + sin 2t
2 2 2
1 j t 1
j t e
= e dt = 1
j = cos(2t 0.25 )
1 1 2
j
e ej 1 j = 2–0.5 cos(2t – 0.25 )
= = = (e e j )
j j j
 GATE Previous Years Solved Paper 57

29. (b) 32. Sol.

1 3 , for 0 t 3
(1 cos 2t ) + cos 2t
f(t) = h(t) =
2 0, otherwise
Frequency components are h(t) = 3[u(t) – u(t – 3)]
f1 = 0
3s
1 e
2 1
2 H(s) = 3
f2 = = = Hz s s
2 2
h(t )
30. (b)
3
y(t) = x(t ) h(2 ) d

t
It is not low pass filter. 0 3
But the system is LTI and BIBO.
Q x(t) = 5
31. (c) 5
Then, X(s) =
2 cos sin 2 s
= H(j ) ...(1) We know that,
3s
2 sin 2 1 e 5
H(j ) = 2 cos Y(s) = H (s) X(s ) = 3
2 s s s
= H1(j ) H2(j ) ...(2) Steady-state output
where, H1(j ) = 2 cos 3s
15 (1 e )
2 sin 2 = lim s Y (s ) = lim s
H2(j ) = = 2 Sa(2 ) s 0 s 0 s s
2
3s
h2(t) = 15 lim 3e [Q L-Hospital’s rule]
s 0

1/2 = 45
2 Sa[2 ]
33. (b)
t
–2 2 2
G( j ) = e 2
H1(j ) = 2 cos = + (ej
) e–j
So, H( j ) = 2 cos H2(j )
g(t ) dt = G(j0) = 0
= (ej + e–j ) H2(j )
H( j ) = ej H2( j ) + e–j H2(j ) y(t) = g(t) u(t)
h(t) = h2(t + 1) + h2(t – 1) Y(j ) = G(j ) U(j )
h( t) 1
Y(j ) = G( j ) + ( )
1 j

2 2
1/2 e 2 2
h(t) = = + e ( )
j
t
–3 –1 1 3
1
So, h(0) = 1 y(t ) dt = Y ( j 0) = = j
j
58 Electronics Engineering Signals & Systems

34. Sol. 37. Sol.

X( )
sin c 2 (5t ) dt
1
Let, f(t) = sinc(5t) sin at
= Sa (5 t) [Q sinc(t) = Sa( t)] t
–a a
1
F( ) = Rect
5 10 X( )

2 1/4
ESDf = F( )
sin 4 t 1 sin 4 t
= ×
F ( ) = G10 ( ) = F( ) 4 t 4 t
–4 4

1/5
1 2
Energy, Ex(t) = X( ) d
2
–5 5
4 2
1 1
1 = d
Ef = f 2 (t ) dt = ESD d 2
4
4
2
1 1 1
5 2 = × [8 ] = = 0.25 J
1 1 2 16 4
sin c 2 (5t ) dt = d
2 5
5
38. (a)
1 1 1
= × × 10 =
2 25 5 A sin ot LTI system B sin( ot + )

35. (c) When a sinusoidal input is given to LTI system,

Group delay is given by the output is also a sinusoidal with change in
magnitude and the phase shift offered by LTI
1 d (f) 1
= = [ 2 ] system.
2 df 2
Group delay = 39. (a)

36. (d) X1( )

j4 f 1
e , f
2 sin at
Given, H(f ) = x1 (t ) =
t
0, f > –a a
2
So, H( f ) = 1
X1( )

H( ) = e–j2 f 1
2
h(t) = (t – 2) sin t
x1 ( t ) =
As we know that, t
–1 1
y(t) = x(t) h(t)
or, y(t) = X( ) H( ) x(t) = x1(t) x1(t)
y(t) = x(t) (t – 2) X( ) = X1( ) X1( )
y(t) = x(t –2)
 GATE Previous Years Solved Paper 59

X1( ) X1 ( ) CTFT 4 f f
4 sin c(2t ) rect = 2 rect
1 1 2 2 2

× 2
X( f ) df = 2 × (2)2 = 8
–1 1 –1 1

X( ) 2
So, y(t ) dt = 8
1
X(f )
=
2
–1 1

sin t f (Hz)
x(t) = –1 0 1
t

40. Sol. 43. Sol.

x(t) = 4 cos200 t + 8 cos400 t y (t )

H(f )
3
2
2 sin 300 t F.T. 2
h(t) = (rad/sec)
t
–300 0 300 1
So, y(t) = 8 cos200 t t
–3 –2 –1 0 1 2 3 4
y(t ) max = 8

2
41. Sol. X( ) d

H(f )
2 2
1 = 2 x(t ) dt = 2 y(t ) dt
Since,
0
2
f = 2×2 y(t ) dt
–20 0 20
2

1 1 0
So, y(t) = × 8 cos(20 t + ) = 2×2 (t + 2)2 dt + (2t + 3)2 dt
2
2 1
= 4 cos(20 t + ),
1 0
42 (t + 2)3 (2t + 3)3
So, Py = =8W = 4 +
2 3 3× 2
2 1
42. Sol.
1 0 33 1 1 26
Hilbert transform does not alter the amplitude = 4 + =4 +
3 6 3 6
spectrum of the signal.
1 26 1 13
2
y(t ) dt =
2
x(t ) dt =
2
X ( f ) df = 4 + =4 × ×
So, 3 6 3 3

x(t) = 4 sinc(2t) 14 56
= 4 × =
3 3
CTFT
sin c(t ) rect (f )
60 Electronics Engineering Signals & Systems

44. (b) Maximum frequency of y(t) is,

Given, x(t) band limited to 10 kHz, fm = 15 kHz
Nyquist rate = 2fm = 30 kHz
X( f )
45. (a)
t
Consider, x(t) = e
By taking Fourier transform,
f
–10 kHz 10 kHz t F.T. 2
e
1+ 2
t
Given, y(t) = x(t ) x 1 + Applying differentiation in frequency,
2
F.T. d
Assume, x1(t) = x(t) t x( t ) j X( )
d
t
and x2(t) = x 1 + F.T. d 2
2 te t j
d 1+ 2
y(t) = x1(t) x2(t)
Apply Fourier transform from convolution in t F.T. [0 2(2 )
te j 2 2
frequency domain property, (1 + )
Y(f ) = X1(f) X2(f )
t F.T. 4
X1(f ) = X(f ) te 2 2
(1 + )
t
X2(t) = x +1 Applying duality property,
2
4 jt F.T.
X2(f ) = 2ej4 f X(2f) 2 2
2 ( )e
(1 + t )
X1 ( f ) X2 ( f )
t F.T. 2 e
2 2 4j
(1 + t )
t F.T.
f f 2 2
e
–10 kHz 0 10 kHz –5 kHz 0 5 kHz (1 + t ) 2j

Y(f )

f
–15 kHz 0 15 kHz
(–10 kHz – 5 kHz) (10 kHz + 5 kHz)
 GATE Previous Years Solved Paper 61

Answers
EE Fourier Transforms, Frequency Response and Correlation

1. (c) 2. (a) 3. (d) 4. (d) 5. (c) 6. (b) 7. (c) 8. (a)

9. (b) 10. (a) 11. (c) 12. (6) 13. (d) 14. (10)

Solutions Fourier Transforms, Frequency Response and Correlation

EE
1. (c)
x(t) = rect [t – 1/2]
1/
x(t) = 1 ; 0 t 1
= 0; otherwise H(j ) =
–
j t
F[x(t)] = x(t ) e dt
y(t) = h(t) x(t)
1
Y(j ) = X(j ) H(j )
j t 1 j
= 1e dt = (1 e )
j
0 1/

2 ej /2
ej /2
j t /2 Y(j ) = if, <
= xe
2j
–

sin /2 j /2
= e
/2
1/
F.T.
F[ x(t ) + x( t ) X( ) + X( ) or Y(j ) = if, <
sin /2 j /2
= [e + ej /2
] –
/2
1
sin /2 so, y(t) = Sa ( t) if, <
= 2 cos
/2 2
1
y(t) = sin c ( t ) if, <
= 2 sin c cos
2 2
1
2. (a) or y(t) = sin c ( t ) if, <
x(t) = sinc( t) = Sa[ t]
So, output is of the form k sinc( t)
h(t) = sinc( t) = Sa[ t]
where, = min( , )
x(t) = Sa[ t]
3. (d)
1/ Symmetrically located pole and zero
All pass filter
X(j ) =
(Constant magnitude)
–
(– )
h(t) = Sa[ t]
62 Electronics Engineering Signals & Systems

4. (d) 7. (c)
By using Parseval’s theorem, Given signal f(t) is an odd signal. Hence F( ) is
2 1 imaginary and odd function of .
X( ) d = 2 1 dt = 2 × 2
1
8. (a)
2
X( ) d =4 Given signal can be drawn as,
x(t)
5. (c)
2 sin 2 1
H(j ) = cos

t
ej + e j
2 sin 2 –1 1
=
2 2 For signal,
1 j 2 sin 2 j 2 sin 2
= e +e f (t )
2
1
2 sin 2
Let, X( ) =

t
1; 2<t<2 –T/2 T/2
Then, x(t) =
0 ; otherwise F( ) is given by
By time shifting property,
Gate width
1 2
h(t) = ( x (t + 1) + x(t 1))
2
T
1 1 F( ) = T Sa
h(0) = [ x(1) + x( 1) = [1 + 1] = 1 2
2 2
Area of gate
6. (b) pulse

Given that, f (t ) F( ) Therefore,

x(t) X( ) = 2 Sa( )
jut Now, x(t) X( )
and g(t) = F( u) e du
Then by time scaling,

1 1
F( ) e j t
d x( at ) X
Now, f(t) = a a
2

x(2t ) Sa ...(i)
2 f(t) = F( ) e j t
d 2
t
u, and x 4 Sa (2 ) ...(ii)
2

2 f(t) = F(u) e jut du t

Now, y(t) = x(2t ) x
2
= g(t) Convolution in time domain multiplication in
g(t) = 2 f(–t) = 2 f(t) frequency domain,
If f(t) is even then,
Y( ) = 4 Sa Sa (2 )
f(t) = f(–t) 2
g(t) f(t) if f(t) is even.
 GATE Previous Years Solved Paper 63

By observing X1( ) and X2( ), we can say,

4 sin X2( ) = X1(– )
2 sin (2 )
Y( ) = IFT
2
2 x2(t) = x1(–t) ...(i)
Now, X1( ) is real. Therefore, x1(t) will be
4
Y( ) = 2
sin sin(2 ) conjugate symmetric.
2
i.e., x1(t) = x1 ( t )
9. (b)
t = –t
For real even function x(t) the Fourier transform
x1(–t) = x1 (t )
X( ) is always real even. y(t) is a derivative of
x(t) which is a real odd function because Now, x1(t) x2(t) = x1(t) x2(–t) [From (i)]
derivative of even function is an odd function = x 1 (t ) x 1 ( t )
and hence, Fourier transform Y( ) is imaginary 2
odd. = x1 (t ) = real function

10. (a) 12. Sol.

Maximum possible frequency of x(t) (2000 t)
e j 10t for t 1 = (f1 + f2)
Since, x(t) =
0 for t > 1 = 5 kHz + 1 kHz
= 6 kHz
j t
as, X( ) = x(t ) e dt
13. (d)
The Fourier transform of
1 1
so, X( ) = j 10t j t jt(10 )
e e dt = e dt 2 sin(t /2)
2 rect
1 1 t
1 sin(2 t )
= [ e j(10 )
e j(10 )
] rect
j(10 ) t 4

2 sin( 10) sin (2 t ) j t

X( ) = So, e dt = rect
( 10) t 4
Putting = 0 in above equation,
11. (c)
sin (2 t )
By observing X1(j ) and X2(j ) we can say that dt = 1
t
they are not conjugate symmetric. Since the
Fourier transform is not conjugate symmetric sin (2 )
the signal will not be real. So x1(t), x2(t) are not 2 dt = 2
t
real. Now the Fourier transform of x1(t) x2(t)

1 14. Sol.
will be X1 ( j ) X 2 ( j ) and by looking at
2 By taking inverse Fourier transform,
x(t) = te–10t u(t)
X1(j ) and X2(j ) we can say that X1(j ) X2(j )
Now, x(t)t = 1 = 1 × e–10 × 1 = e–10
will be conjugate symmetric and thus x1(t) × x2(t)
10
will be real. Thus, ln{ x(t )} = ln ( e ) = 10 = 10
 5 Laplace Transform

ELECTRO NICS EN GINEERIN G (a) 2[1 – e–2t] u(t) (b) 4[e–t – e–2t] u(t)

(GATE Previous Years Solved Papers) (c) sin2t (d) (1 – 4e–4t) u(t)
[EC-1990 : 2 Marks]
Q.1 Laplace transform of the functions tu(t) and u(t)
sin(t) are respectively. Q.6 The pole-zero pattern of a certain filter is shown
in figure. The figure must be of the following
1 s 1 1
(a) 2
, 2 (b) , 2 type.
s s +1 s s +1
1 1 s j
(c) 2
, 2 (d) s , 2
s s +1 s +1 +j2
[EC-1987 : 2 Marks]
+j1
Q.2 The Laplace transform of a function f(t) u(t),
–1
where f(t) is periodic with period T, is A(s) times
the Laplace transform of its first period. Then, +1

(a) A(s) = s –j 1

1
(b) A(s) = –j 2
(1 exp( Ts ))
1 (a) low-pass (b) high-pass
(c) A(s) =
(1 + exp( Ts)) (c) all-pass (d) band-pass
(d) A(s) = exp (Ts) [EC-1991 : 2 Marks]
[EC-1988 : 2 Marks] Q.7 The voltage across an impedance in a network
Q.3 The transfer function of a zero-order hold is is V(s) = Z(s) I(s), where V(s), Z(s) and I(s) are the
Laplace transform of the corresponding time
1 exp( Ts) 1
(a) (b) functions v(t), z(t) and i(t). The voltage v(t) is
s s
(a) v(t) = z(t) i(t)
1
(c) 1 (d) t
[1 exp( Ts)]
(b) v(t ) = i( ) z(t )d
[EC-1988 : 2 Marks] 0
t
Q.4 The transfer function of a zero-order hold is
(c) v(t ) = i( ) z(t + ) d
1 exp( Ts ) 1 0
(a) (b)
s s (d) v(t) = z(t) + i(t)
1 [EC-1992 : 2 Marks]
(c) 1 (d)
[ exp( Ts)]
Q.8 The Laplace transform of te periodic function
[EC-1988 : 2 Marks] f(t) described by the curve below, i.e.,
Q.5 The response of an initially relaxed linear sin t if (2n 1) t 2n ( n = 1, 2, 3...)
f (t ) =
constant parameter network to a unit impulse 0 otherwise
applied at t = 0 is 4e–2 u(t). The response of this is ________ .
network to a unit step function will be
 GATE Previous Years Solved Paper 65

f(t) (c) ratio of the Laplace transform of the output

and that of the input with all initial
conditions zeros.
2 3 4 5 6 7 8
t
0 (d) none of these
[EC-1995 : 1 Mark]
[EC-1993 : 2 Marks] Q.14 The final value theorem is used to find the
(a) steady state value of the system output.
K
Q.9 If F(s ) = L [ f (t ) = then, lim f (t ) (b) initial value of the system output.
( s + 1) ( s 2 + 4) t
(c) transient behaviour of the system output.
is given by (d) none of these
K [EC-1995 : 1 Mark]
(a) (b) zero
4
Q.15 The inverse Laplace transform of the function,
(c) infinite (d) undefined
[EC-1993 : 2 Marks] s+5
is
(s + 1) (s + 3)
Q.10 The Laplace transform of a unit ramp function
starting at t = a, is (a) 2e–t – e–3t (b) 2e–t + e–3t
as (c) e–t – 2e–3t (d) e–t + 2e–3t
1 e
(a) (b) [EC-1996 : 2 Marks]
( s + a )2 ( s + a )2
Q.16 The Laplace transform of e t cos( t) is equal to
e as a
(c) (d) (s + ) (s + )
s2 s2 (a) (b)
[EC-1994 : 1 Mark] (s 2
) + 2 (s + )2 + 2

Q.12 Indicate whether the following statement is 1

(c) (d) none of these
True/False. Give reason for your answer. If G(s) (s )2
[EC-1997 : 1 Mark]
1
is a stable transfer function, the F(s) = is
G( s) Q.17 Match each of the items 1, 2 on the left with
always a stable transfer function. most appropriate item A, B, C or D on the right.
In the case of a linear time invariant system
[EC-1994 : 1 Mark]
1. Poles in the right half plane implies.
(2 s + 1) 2. Impulse response zero for t 0 implies
Q.12 If L [ f (t )] = 2
, then f(0+) and f( ) are
s + 2s + 5 (a) Exponential decay of output

given by (b) System is causal.

(a) 0, 2 respectively (b) 2, 0 respectively (c) No stored energy in the system.

(c) 0, 1 respectively (d) 2/5, 0 respectively (d) System is unstable.

[Note : ‘L’ stands for ‘Laplace transform of’] [EC-1997 : 2 Marks]

[EC-1995 : 1 Mark]
Q.18 If L[ f (t )] = , then the value of
Q.13 The transfer function of a linear system is the (s2 + 2
)
(a) ratio of the output, vo(t) and input vi(t).
(b) ratio of the derivatives of the output and lim f (t )
t
the input.
 66 Electronics Engineering Signals & Systems

(a) cannot be determined Q.23 The transfer function of a system is given by

(b) is zero 1
H (s) = 2
(c) is unity s (s 2)
(d) is infinite The impulse response of the system is
[EC-1998 : 1 Mark] (a) (t2 e–2t) u(t) (b) (t e–2t) u(t)

Q.19 The transfer function of a zero-order hold system (c) (te–2t) u(t) (d) (te–1t) u(t)
is ( denotes convolution, and u(t) is unit step
function)
1 sT 1 sT
(a) (1 + e ) (b) (1 e ) [EC-2001 : 1 Mark]
s s
Q.24 The Laplace transform of a continuous-time
1 1 sT
(c) 1 e sT (d) 1 + e
s s 5 s
signal x(t) is X (s ) = . If the Fourier
2
[EC-1998 : 1 Mark] s s 2
transform of this signal exists, then x(t) is
Q.20 If L [f(t) = F(s), then L [ f(t – T)] is equal to
(a) e2t u(t) – 2e–t u(t)
(a) esT F(s) (b) e–sT F(s)
(b) –e2t u(–t) + 2e–t u(t)
F(s ) F(s)
(c) (d) (c) –e2t u(–t) – 2e–t u(t)
1 + esT 1 e sT
(d) e2t u(–t) – 2e–t u(t)
[EC-1999 : 1 Mark]
[EC-2002 : 2 Marks]
Q.21 Given that,
Q.25 The Laplace transform of i(t) is given by
2
s+2 s +1 2
L [ f (t )] = 2
, L [ g(t )] = , I (s) =
s +1 (s + 3) (s + 2) s (1 + s )
t As t , the value of i(t) tends to
h(t ) = f ( ) g(t )d
(a) 0 (b) 1
0
(c) 2 (d)
L[h(t)] is
[EC-2003 : 1 Mark]
2
s +1
(a) Q.26 Consider the function f(t) having Laplace
s+3
transform:
1
(b) F(s ) = o , Re[s] > 0
s+3
s2 + 2
o
s2 + 1 s+2 The final value of f(t) would be
(c) +
(s + 3) (s + 2) s2 + 1 (a) 0 (b) 1
(d) None of the above (c) –1 f( ) 1 (d)
[EC-2000 : 1 Mark] [EC-2006 : 2 Marks]

Q.22 A linear time invariant system has an impulse Q.27 If the Laplace transform of a signal y(t) is
response e2t, for t > 0. If initial conditions are
1
zero and the input is e3t, the output for t > 0 is Y (s) = , then its final value is
s (s 1)
(a) e3t – e2t (b) e5t
(a) –1 (b) 0
(c) e3t + e2t (d) none of these
(c) 1 (d) unbounded
[EC-2000 : 2 Marks]
[EC-2007 : 1 Mark]
 GATE Previous Years Solved Paper 67

Q.28 Given that, F(s) is the one-side Laplace transform Q.33 The unilateral Laplace transform of f(t) is

t 1
. The unilateral Laplace transform of
of f(t), the Laplace transform of f ( ) d is s2 + s + 1
0
t f(t) is
1 s 2s + 1
(a) sF(s) – f(0) (b) F (s ) (a) (b)
s 2
(s + s + 1) 2
(s + s + 1)2
2

s
1 s 2s + 1
(c) F( ) d (d) [ F( s ) f (0)] (c) (d)
s 2
(s + s + 1) 2 (s + s + 1)2
2
0
[EC-2009 : 2 Marks] [EC-2012 : 1 Mark]

Q.29 A continuous-time LTI system is described by Q.34 The impulse response of a system is h(t) = tu(t).
For an input u(t – 1), the output is
d 2 y(t ) dy(t ) dx(t )
+4 + 3 y(t ) = 2 + 4x(t )
dt 2 dt dt t2 t (t 1)
(a) u(t ) (b) u (t 1)
Assuming zero initial conditions, the response 2 2
y(t) of the above system for the input (t 1)2 t2 1
x(t) = e–2t u(t) is given by (c) u (t 1) (d) u (t 1)
2 2
(a) (et – e3t) u(t) (b) (e–t – e–3t) u(t) [EC-2013 : 1 Mark]
(c) (e–t + e–3t) u(t) (d) (et + e3t) u(t)
Q.35 Assuming zero initial condition, the response
[EC-2010 : 2 Marks]
y(t) of the system, given below to a unit step
Q.30 If the unit step response of a network is (1 – e– t), input u(t) is
then its unit impulse response is
U(s ) 1/s Y(s )
(a) e– t (b) –1 e– t

(c) (1 – –1) e– t (d) (1 – ) e– t

(a) u(t) (b) tu(t)
[EC-2011 : 1 Mark]
t2
(c) u(t ) (d) e–t u(t)
Q.31 An input x(t) = exp(–2t) u(t) + (t – 6) is applied 2
to an LTI system with impulse response [EC-2013 : 1 Mark]
h(t) = u(t). The output is
Q.36 Which one of the following statements is not
(a) [1 – exp(–2t)] u(t) + u(t + 6)
true for a continuous time causal and stable LTI
(b) [1 – exp(–2t)] u(t) + u(t – 6) system?
(c) 0.5[1 – exp(–2t)] u(t) + u(t + 6)
(a) All the poles of the system must lie on the
(d) 0.5[1 – exp(–2t)] u(t) + u(t – 6) left side of the j -axis.
[EC-2011 : 2 Marks] (b) Zero of the system can lie anywhere in the
s-plane.
2 (s + 1)
Q.32 If F(s ) = L [ f (t )] then the initial and (c) All the poles must lie within s = 1.
s 2 + 4s + 7
final values of f(t) are respectively (d) All the roots of the characteristic equation
must be located on the left side of the j -axis.
(a) 0, 2 (b) 2, 0
[EC-2013 : 1 Mark]
2 2
(c) 0, (d) ,0
7 7 Q.37 A system is described by the differential
[EC-2011 : 2 Marks] equation,
 68 Electronics Engineering Signals & Systems

For the above system

d2 y dy
+5 + 6 y(t ) = x(t ) (a) only S1 and S2 are true.
dt 2 dt
(b) only S2 and S3 are true.
Let x(t) be a rectangular pulse given by
(c) only S1 and S3 are true.
1, 0<t<2
x (t ) = (d) S1, S2 and S3 are true.
0 , otherwise
[EC-2014 : 2 Marks]
dy
Assuming that, y(0) = 0 and = 0 at t = 0, the Q.41 The unilateral Laplace transform of f (t) is
dt
1
Laplace transform of y(t) is . Which one of the following is the
2
s +s+1
e 2s 1 e 2s
(a) (b) unilateral Laplace transform of g(t) = t f(t) ?
s (s + 2) (s + 3) s (s + 2) (s + 3)
s (2s + 1)
e 2s 1 e 2s (a) (b)
(c) (d)
2
(s + s + 1) 2 (s + s + 1)2
2
(s + 2) (s + 3) (s + 2) (s + 3)
s 2s + 1
[EC-2013 : 2 Marks] (c) (d)
(s 2 + s + 1)2 (s + s + 1)2
2

Q.38 A system is described by the following [EC-2014 : 2 Marks]

differential equation, where u(t) is the input to
the system and y(t) is the output of the system, Q.42 A stable linear time invariant (LTI) system has

y (t ) + 5 y ( t ) = u ( t ) 1
a transfer function H (s ) = . To make
2
s +s 6
When y(0) = 1 and u(t) is a unit step function,
y(t) is this system causal it needs to be cascaded with
(a) 0.2 + 0.8e–5t (b) 0.2 – 0.2e–5t another LTI system having a transfer function
(c) 0.8 + 0.2e–5t (d) 0.8 – 0.8e–5t H1(s). A correct choice for H1(s) among the
[EC-2014 : 2 Marks] following options is
(a) s + 3 (b) s – 2
Q.39 The input –3e–2t u(t), where u(t) is the unit step
(c) s – 6 (d) s + 1
function, is applied to a system with transfer
[EC-2014 : 2 Marks]
s 2
function . If the initial value of the output Q.43 A causal LTI system has zero initial conditions
s+3
and impulse response h(t). Its output x(t) and
is –2, then the value of the output at steady state input y(t) are related through the linear constant
is _______ .
coefficient differential equation,
[EC-2014 : 1 Mark]
d 2 y(t ) dy(t ) 2
Q.40 Let h(t) denote the impulse response of a causal 2
+ + y(t ) = x(t )
dt dt
1 Let another signal g(t) be defined as,
system with transfer function . Calculate
s+1 t
2 dh(t )
the following three statements: g( t ) = h( ) d + + h (t )
dt
S1 : The system is stable. 0

h (t + 1) If G(s) is the Laplace transform of g(t), then the

S2 : is independent of t for t > 0.
h(t ) number of poles of G(s) is ________ .
S3 : A non-causal system with the same transfer [EC-2014 : 2 Marks]
function is stable.
 GATE Previous Years Solved Paper 69

Q.44 Input x(t) and output y(t) of an LTI system are Q.48 Let x(t) = s(t) + s(–t) with s(t) = e–4t u(t), where
related by the differential equation y (t) – y (t) – u(t) is unit step function. If the bilateral Laplace
6y(t) = x(t). If the system is neither causal nor transform of x(t) is
stable, the impulse response h(t) of the system is 16
X (s) = 2
4 < Re { s} < 4
1 3t 1 2t s 16
(a) e u( t) + e u( t )
5 5 Then the value of is _______ .
1 3t 1 2t [EC-2015 : 2 Marks]
(b) e u ( t) + e u( t )
5 5
Q.49 Consider the function g(t) = e–t sin(2 t) u(t)
1 3t 1 2t where u(t) is the unit step function. The area
(c) e u( t) e u(t )
5 5 under g(t) is _______ .
1 3t 1 2t [EC-2015 : 1 Mark]
(d) e u ( t) e u(t )
5 5
Q.50 The Laplace transform of the causal periodic
[EC-2015 : 2 Marks]
square wave of period T shown in the figure
Q.45 Consider the differential equation, below is

dx f (t )
= 10 0.2 x with initial conduction x(0) = 1.
dt
1
The response x(t) for t > 0 is
(a) 2 – e–0.2t (b) 2 – e0.2t
(c) 50 – 49e–0.2t (d) 50 – 49e0.2t
t
[EC-2015 : 2 Marks] 0 T/2 T 3T/2 2T

Q.46 The bilateral Laplace transform of a function, 1

(a) F(s ) = sT /2
1, if a t b 1+ e
f (t ) = is
0 , otherwise 1
(b) F(s ) =
s (1 + e sT /2 )
a b z
e ( a b)
(a) (b) 1
s s (c) F(s ) =
s (1 e sT /2 )
as bs ( a b)
e e e
(c) (d) 1
s s (d) F (s ) = sT
1 e
[EC-2015 : 1 Mark]
[EC-2016 : 2 Marks]
Q.47 Let the signal f(t) = 0 outside the interval [T1, T2]
Q.51 A first order low-pass filter of time constant T is
where T 1 and T 2 are finite. Furthermore,
excited with difference input signals (with zero
f (t ) < . The region of convergence (ROC) of initial conditions upto t = 0). Match the
the signal’s bilateral Laplace transform F(s) is excitation signals X, Y, Z with the
corresponding time responses for t 0:
(a) a parallel strip containing the j axis.
List-I List-II
(b) a parallel strip not containing the j axis.
X. Impulse P. 1 – e–t/T
(c) the entire s-plane.
Y. Unit step Q. t – T(1 – e–t/T)
(d) a half-plane containing the j axis.
Z. Ramp R. e–t/T
[EC-2015 : 1 Mark]
 70 Electronics Engineering Signals & Systems

(a) X - R, Y - Q, Z - P
G(s )
(b) X - Q, Y - P, Z - R that is, Y (s ) = . The forced response of the
s
(c) X - R, Y - P, Z - Q
system is
(d) X - P, Y - R, Z - Q
(a) u(t)
[EC-2016 : 2 Marks]
(b) 2u(t)
(c) u(t) – 2e–t u(t) + e–3t u(t)
2
Q.52 A signal 2 cos t cos( t ) is the input to (d) 2u(t) – 2e–t u(t) + e–3t u(t)
3
[EC-2019 : 1 Mark]
an LTI system with the transfer function,
H(s) = es + e–s Q.56 A system with transfer function
If Ckdenote the kth coefficient in the exponential 1
G(s) = ,a>0
Fourier series of the output signal, then c3 is (s + 1) (s + a)
equal to is subjected an input 5 cos3t. The steady-state
(a) 0 (b) 1
1
(c) 2 (d) 3 output of the system is cos(3t 1.892).
10
[EC-2016 : 2 Marks]
The value of a is ________ .
Q.53 Consider the following statements for [EC-2020 : 2 Marks]
continuous-time linear time invariant (LTI)
systems: Q.57 Let x1(t) = e–t u(t) and x2(t) = u(t) – u(t – 2), where
I. There is no bounded input bounded output u( ) denotes the unit step function. If y(t) denotes
(BIBO) stable system with a pole in the right the convolution of x 1 (t) and x 2 (t), then
half of the complex plane. lim y(t ) = ___ (Rounded off to 1 decimal place).
t
II. There is no causal and BIBO stable system
with a pole in the right half of the complex [EC-2022]
plane.
Which one among the following is correct? ELECTRICAL EN GINEERIN G
(a) Both I and II are true. (GATE Previous Years Solved Papers)
(b) Both I and II are not true.
Q.1 The Laplace transforma.tion of f(t) is F(s).
(c) Only I is true.
(d) Only II is true. Given, F( s ) = , the final value of f(t) is
s2 + 2
[EC-2017 : 1 Mark]
(a) infinity (b) zero
Q.54 The transfer function of a causal LTI system is
(c) one (d) none of these
H(s) = 1/s. If the input to the system is
[EE-1995 : 1 Mark]
x(t) = [sin(t)/ t] u(t), where u(t) is a unit step
function, the system output y(t) as t is Q.2 The Laplace transform of (t2 – 2t) u(t – 1) is
________ .
2 s 2 s 2 2s 2 s
[EC-2017 : 2 Marks] (a) e e (b) 3
e e
s 3
s 2 s s2
Q.55 Let Y(s) be the unit-step response of a causal
2 s 1 s
system, having a transfer function, (c) e e (d) None of the above
s3 s
3 s
G(s) = [EE-1998 : 2 Marks]
(s + 1) (s + 3)
 GATE Previous Years Solved Paper 71

Q.3 The output of a linear time invariant control 5

(a) 5 (b)
system is c(t) for a certain input r(t). If r(t) is 2
modified by passing it through a block whose 5
transfer function is e–s and then applied to the (c) (d) 0
3
system, the modified output of the system [EE-2004 : 1 Mark]
would be
Q.8 The Laplace transform of a function f(t) is
r(t ) r (t )
(a) (b) t 5s 2 + 23s + 6
1 + et 1+ e F( s ) =
s (s 2 + 2 s + 2)
(c) c(t – 1) (d) r(t) u(t – 1)
[EE-1998 : 1 Mark] As t , f(t) approaches
(a) 3 (b) 5
Q.4 A rectangular current pulse of duration T and
17
magnitude 1 has the Laplace transform (c) (d)
2
1 1 [EE-2005 : 2 Marks]
(a) (b) exp( Ts)
s s
Q.9 The integrator given by
1 1 t
(c) exp(Ts) (d) [1 exp( Ts)]
s s y(t ) = x( ) d
[EE-1999 : 2 Marks]
(a) has no finite singularities in its double sided
Q.5 Given the relationship between the input u(t) Laplace transform H(s).
and the output y(t) to be (b) produces a bounded output for every causal
t bounded input.
t( t )
y(t ) = (2 + t )e u( ) d
(c) produces a bounded output for every anti-
0
causal bounded input.
the transfer function Y(s)/U(s) is (d) has no finite zeroes in its double sided
2s Laplace transform H(s).
2e s+2
(a) (b)
s+3 (s + 3)2 [EE-2006 : 2 Marks]
2s + 5 2s + 7 Common Data Questions (10 and 11):
(c) (d)
s+3 (s + 3)2 Given: f(t) and g(t) as shown below.
[EE-2001 : 2 Marks] f (t ) g (t )

Q.6 Let Y(t) be the Laplace transform of the function 1 1

y(t), then the final value of the function is
(a) LimY (s) (b) Lim Y ( s)
s 0 s t t
0 1 0 3 5

(c) Lim sY (s) (d) Lim sY ( s)

s 0 s Q.10 g(t) can be expressed as:
[EE-2002 : 1 Mark] t
(a) g(t) = f(2t – 3) (b) g(t ) = f 3
2
5
Q.7 Consider the function, F( s ) = 3 t 3
2
s (s + 3s + 2) (c) g(t ) = f 2t (d) g(t ) = f
2 2 2
where F(s) is the Laplace transform of the [EE-2010 : 2 Marks]
function f(t). The initial value of (t) is equal to
 72 Electronics Engineering Signals & Systems

Q.11 The Laplace transform of g(t) is Q.15 Assuming zero initial condition, the response
1 3s 1 y(t) of the system given below to a unit step input
(a) (e e 5s ) (b) (e 5s
e 3s
)
s s u(t) is
3s 1 5s
e
(c) (1 e 2s
) (d) (e e 3s ) U(s) 1/s Y(s )
s s
[EE-2010 : 2 Marks] (a) u(t) (b) t u(t)
Q.12 Let the Laplace transform of a function f(t) which t2
(c) u(t ) (d) e–t u(t)
exists for t > 0 be F1(s) and the Laplace transform 2
of its delayed version f(t – ) be F2(s). F1 (s) be [EE-2013 : 1 Mark]

the complex conjugate to F1(s) with the Laplace Q.16 Which one of the following statements is not
true for a continuous time causal and stable LTI
F (s) F1 (s) system ?
variable set as s = + j . If G(s) = 2 2
,
F1 (s) (a) All the poles of the system must lie on the
left side of the j -axis.
then the inverse Laplace transform of G(s) is (b) Zero of the system can lie anywhere in the
(a) an ideal impulse (t). s-plane.
(b) an ideal delayed impulse (t – ).
(c) All the poles must lie within s = 1.
(c) an ideal step function u(t).
(d) All the roots of the characteristic equation
(d) an ideal delayed step function u(t – ).
must be located on the left side of the j -axis.
[EE-2011 : 2 Marks]
[EE-2013 : 1 Mark]
Q.13 The unilateral Laplace transform of f(t) is
Q.17 Consider an LTI system with transfer function:
1 1
2
. The unilateral Laplace transform of H (s) =
s +s+1 s (s + 4)
t f(t) is If the inputs to the system is cos(3t) and the
s 2s + 1 steady-state output is A sin(3t + ), then the
(a) (b)
2
(s + s + 1) 2
(s + s + 1)2
2 value of A is
1 1
s 2s + 1 (a) (b)
(c) (d) 30 15
(s 2 + s + 1)2 (s 2 + s + 1)2
3 4
[EE-2012 : 1 Mark] (c) (d)
4 3
Q.14 Consider the differential equation: [EE-2014 : 2 Marks]

d 2 y(t ) dy(t )
2
+2 + y(t ) = (t ) with t
dt dt Q.18 The Laplace transform of f (t ) = 2 is s–3/2.
dy
y(t ) t = 0 = 2 and =0
dt t = 0 The Laplace transform of g(t ) = 1/ t is
dy 5/2
The numerical value of is 3s
dt t = 0+ (a) (b) s–1/2
2
(a) –2 (b) –1 (c) s1/2 (d) s3/2
(c) 0 (d) 1 [EE-2015 : 1 Mark]
[EE-2012 : 2 Marks]
 GATE Previous Years Solved Paper 73

Q.19 The Laplace transform of f(t) = e2t sin(5t) u(t) is Q.24 The output response of a system is denoted as
5 5 y(t), and its Laplace transform is given by
(a) 2 (b) 2
s 4s + 29 s 5 10
Y (s) = 2
s 2 5 s(s + s + 100 2 )
(c) 2 (d)
s 4s + 29 s+5 The steady-state value of y(t) as,
[EE-2016 : 1 Mark]
1
(a) 100 2 (b)
10 2
Y (s) s
Q.20 The transfer function of system is = .
R(s ) s + 2 1
(c) 10 2 (d)
The steady-state output y(t) is A cos(2t + ) for 100 2
the input cos(2t). The value of A and , [EE-2019 : 1 Mark]
respectively are
Q.25 A system transfer function is
1 1
(a) , 45° (b) , + 45°
2 2 a1 s 2 + b1s + c1
H (s) =
(c) 2 , 45° (d) 2 , + 45° a2 s 2 + b2 s + c 2
[EE-2016 : 1 Mark] If a1 = b1 = 0, and all other coefficient are positive,
Q.21 Consider a causal LTI system characterized by the transfer function represents a
(a) high pass filter (b) low pass filter
dy(t ) 1
differential equation + y(t ) = 3x(t ). The (c) notch filter (d) band pass filter
dt 6
[EE-2019 : 1 Mark]
response of the system to the input
x(t) = 3e–t/3 u(t), where u(t) denotes the until step Q.26 The inverse Laplace transform of
function, is
s+3
(a) 9 e–t/3 u(t) H (s) = 2
for t 0 is
s + 2s + 1
(b) 9 e–t/6 u(t)
(c) 9 e–t/3 u(t) – 6 e–t/6 u(t) (a) 2t e–t + e–t (b) 3 e–t
(d) 54 e–t/6 u(t) – 54 e–t/3 u(t) (c) 3t e–t + e–t (d) 4t e–t + e–t
[EE-2016 : 1 Mark] [EE-2019 : 1 Mark]

Q.22 The solution of the differential equation, for t > Q.27 Consider a linear time-invariant system whose
0, y (t) + 2y (t) + y(t) = 0 with initial conditions input r(t) and output y(t) are related by the
y(0) = 0 and y (0) = 1, is (u(t) denotes the unit following differential equation,
step function).
d 2 y(t )
(a) t e–t u(t) (b) (e–t – te–t) u(t) + 4 y(t ) = 6 r (t )
dt 2
(c) (–e–t – te–t) u(t) (d) e–t u(t)
[EE-2016 : 1 Mark] The poles of this system are at
(a) +4j, –4j (b) +4, –4
Q.23 Consider a linear time-invariant system with
(c) +2, –2 (d) +2j, –2j
transfer function:
[EE-2020 : 1 Mark]
1
H (s ) =
(s + 1) Q.28 Which of the following statements is the about
If the input is cos(t) and the steady-state output the two sided Laplace transform?
is A cos(t + ), then the value of A is ______ . (a) It has no poles for any bounded signal that
[EE-2016 : 1 Mark] is non-zero only inside a finite time interval.
74 Electronics Engineering Signals & Systems

(b) If a signal can be expressed as a weighted Q.29 The causal realization of a system transfer
sum of shifted one sided exponential, then function H(s) having poles at (2, –1), (–2, 1) and
its Laplace transform will have no poles. zeroes at (2, 1), (–2, –1) will be
(c) It exists for every signal that may or may (a) unstable, complex, all pass
not have a Fourier transform. (b) unstable, real, high pass
(d) The number of finite poles and finite zeroes (c) stable, complex, low pass
must be equal. (d) stable, real, all pass
[EE-2020 : 1 Mark] [EE-2020 : 2 Marks]

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Laplace Transform

1. (c) 2. (b) 3. (a) 4. (a) 5. (a) 6. (c) 7. (b) 8. (Sol.)

9. (d) 10. (c) 11. (Sol.) 12. (b) 13. (c) 14. (a) 15. (a) 16. (a)

17. (1-D, 2-B) 18. (a) 19. (b) 20. (b) 21. (b) 22. (a) 23. (b)

24. (c) 25. (c) 26. (c) 27. (d) 28. (d) 29. (b) 30. (a) 31. (d)

32. (b) 33. (d) 34. (c) 35. (b) 36. (c) 37. (b) 38. (a) 39. (0)

40. (a) 41. (d) 42. (b) 43. (1) 44. (b) 45. (c) 46. (c) 47. (c)

48. (–2) 49. (0.155) 50. (b) 51. (c) 52. (b) 53. (d) 54. (0.5) 55. (a, c)
56. (4) 57. (0)

Solutions
EC Laplace Transform

1. (c) 2. (b)
L.T. 1 The given function represents a causal periodic
tu(t )
s 2 signal,
f(t) u(t) = 0 ; t<0
L.T. a
u(t )sin at Period of f(t) = T ; for t > 0
s + a2
2
Let, f1(t) = f(t) u(t); 0 t T
L.T. 1 1
u(t )sin t = = 0; otherwise
s 2 + 12 s2 + 1
 GATE Previous Years Solved Paper 75

f(t) u(t) 4. (a)

The impulse response h(t) of zero order hold
system is
h(t) = u(t) – u(t – T)
t
0 T 2T where, T is the sampling period,
1 1 Ts
f1(t) H(s) = e
s s
1 exp( Ts )
H(s) =
s
t
0 T 5. (a)
Given that, h(t) = 4e–2t
f(t) u(t) = f 1 (t nT ) 1
n=0 So, H(s) = 4
(s + 2)
Let, f1(t) F1(s)
Given that, x(t) = u(t)
nTs
f1(t – nT) e F1 (s ) 1
So, X(s) =
s
nTs
f(t) u(t) F( s ) = e F1 (s) We know that,
n=0
Y (s )
H(s) =
F1 (s ) X( s )
= Ts
1 e Y(s) = H(s) Y(s)
st
1 Transform of the 1 4 1 1 1
f(t) u(t) ×
Y(s) = × =2
1 e Ts period of f (t ) u(t )
(s + 2) s s s+2
1 Taking inverse Laplace transform of both side,
A(s) = Ts
1 e y(t) = 2[1 – e–2t] u(t)

3. (a) 6. (c)
A zero order hold system holds the input signal In the given pole-zero pattern, poles and zeros
value for a period of T, i.e. for an input of short are symmetrical about imaginary axis. This is
duration ( ) pulse, it produces an output pulse the case of all-pass filter.
of duration T, the sampling period.
7. (b)
T
s (t ) Zero order We know that, multiplication of two functions
hold in frequency (s) domain is equivalent to the
convolution in time domain.
Input, x(t) = (t)
t
So, X(s) = 1
So, v(t) = i( ) z(t )d
Output, y(t) = u(t) – u(t – T) 0
Ts Ts
1 e 1 e
So, Y(s) = = 8. Sol.
s s s
The given signal f(t) is a causal periodic signal
Transfer function,
with period,
Ts
Y (s) 1 e To = 2
H(s) = =
X( s ) s f1(t) = f(t) ; 0<t<2
= 0; otherwise
76 Electronics Engineering Signals & Systems

s
f(t) s 1+ e 1
F(s) = e 2
× 2 s
s +1 (1 e )
2 s
t s (1 + e )
0 F(s) = e 2 s s
( s + 1) (1 + e ) (1 e )
s
e
F(s) =
(s 2 + 1) (1 e s
)
sin(t ) u (t )

9. (d)
K
F(s) =
0 2 3 4
t (s + 1) ( s 2 + 4)
So, poles = –1; ±2j
Since all the poles are not in the left half of
s-plane. So, the final value theorem cannot be
sin(t – ) u (t – )
applied.

So, lim f (t ) is indeterminate.

t
t
0 2 3 4 5
10. (c)
L.T. 1
r (t )
s2
sin(t – 2 ) u(t – 2 )

L.T. 1 e as
r (t a ) e as × 2 = 2
s s
t
0 2 3 4 5 6 11. Sol.
If G(s) stable then all its poles must lie in the left
half of s-plane and there is no restriction on its
So, f1(t) = –[sin(t – ) u(t – ) + sin(t – 2 )
zeros, which can lie also in the right half of
1
sin(t ) u(t ) 2 1
s +1 s-plane. The inverse function F(s) = may
G( s)
e s
sin(t ) u(t ) or my not be stable. The zeros of G(s) may lie in
s2 + 1
the right half of s-plane.
e 2 s Hence the given statement is not true.
sin(t 2 ) u(t 2 )
s2 + 1
1
s 2 s
So, F(s) = is not always a stable transfer
e e G( s)
F1(s) = 2
+
s +1 s2 + 1 function.
s
s 1+e 12. (b)
F1(s) = e
s2 + 1 Initial value theorem,

F1 (s ) F1 (s ) 2s(s + 1)
F1(s) = = f(0+) = lim 2
1 e Tos
1 e 2 s s s + 2s + 5
 GATE Previous Years Solved Paper 77

2s2 + 2s 18. (a)

f(0+) = lim 2
s s + 2s + 5
F(s) =
2 s2 + 2
s2 2 +
s f(t) = sin t
f(0+) = lim
s 2 2 5 lim f (t ) = lie between – 1 to +1
s 1+ + 2
s s t

2 So, it can not be determined.

2+
2+0 s
f(0+) = lim = =2
s 2 5 1+0+0 19. (b)
1+ + 2
s s
The impulse response h(t) of zero-order hold
Final value theorem,
system is given by
s 2 (s + 1) h(t) = u(t) – u(t – T)
f( ) = lim sF(s ) = lim
s s 0 s2 + 2s + 5 where T is the sampling period,
f( ) = 0
1 1 sT 1 sT
H(s) = e = [1 e ]
13. (c) s s s

Y (s ) 20. (b)
H(s) =
X( s ) f(t – T) = F(s) e–sT
(Time shifting property of Laplace transform)
14. (a)
Final value theorem is used to find the final value 21. (b)
of the steady-state value of the system output, Convolution in time domain is multiplication
f( ) = lim sF(s) in s-domain.
s 0
1
L[h(t)] = L[ f (t )] × L[ g(t )] =
15. (a) s+3

(s + 5) 22. (a)
F(s) =
(s + 1) (s + 3) h(t) = e2t
2 1 1
F(s) = H(s) =
s+1 s+3 s+2
Taking inverse Laplace transform on both side, x(t) = e3t
f(t) = 2e–t – e–3t
1
X(s) =
16. (a) s 3

s Output = x(t) h(t)

L.T.
cos( t ) u(t )
s2 + 2 1
= X(s) H (s) =
(s )
(s 2) (s 3)
t L.T.
e cos( t ) u(t ) 2 2
(s + ) + 1 A B
= +
(s 2) (s 3) s 2 s 3
17. Sol.
A = –1
1-D, 2-B
B=1
(i) Poles lie in the right half of s-plane Output = [e3t – e2t]
unstable system.
(ii) h(t) = 0 for t 0 causal system.
78 Electronics Engineering Signals & Systems

23. (b) 28. (d)

Impulse response of system L–1[H(s)] t
1
1 1 1 L f( )d = [F(s) f (0)]
+2 t s
= × = (t e ) u(t ) 0
s 2 (s 2) s2 s 2
29. (b)
24. (c)
d 2 y(t ) 4 dy(t ) 2 dx(t )
5 s 5 s + + 3 y(t ) = + 4 x(t )
X(s) = = dt 2 dt dt
s2 s 2 (s + 1) (s 2)
Taking Laplace transform on both sides
5 s A B
= + (assuming zero initial conditions),
(s + 1) (s 2) s+1 s 2
s2Y(s) + 4sY(s) + 3Y(s) = 2sX(s) + 4X(s)
B=1 Y (s ) 2s + 4 2(s + 2)
A = –2 or, = 2 =
X( s ) s + 4s + 3 (s + 1) (s + 3)
x(t) = –2e–t u(t) – e2t u(–t)
Given that, x(t) = e–2t u(t)
ROC : –1 < < 2
1
Note : ROC of Laplace transform should include X(s) =
s+2
= 0 line.
2(s + 2)
Y(s) =
25. (c) (s + 1) (s + 3) (s + 2)
lim i(t ) = lim sI (s) 2 1 1
t s 0 = =
(s + 1) (s + 3) s + 1 s + 3
2 Taking inverse Laplace transform on both sides,
= lim s =2
s 0 s (1 + s ) y(t) = (e–t – e–3t) u(t)

26. (c) 30. (a)

f (t )
Unit step response,
s(t) = (1 – e– t)
So, unit impulse response is
t ds(t ) d t t
h(t) = = (1 e )= e
dt dt

31. (d)
L–1[F(s)] = sin ot
x(t) = e–2t u(t) + (t – 6)
f(t) = sin ot
–1 f( ) 1 1 6s
X(s) = +e
s+2
27. (d) 1
Final value theorem is applicable only when all H(s) =
s
the poles of system lies in left half of s-plane. Y(s) = X(s) H(s)
Q s = 1 is right s-plane pole. Therefore, final
1 1
value theorem cannot be applied, Y(s) = + e 6s
s (s + 2) s
1 1
Y(s) = 1 1 1 1 6s
s 1 s Y(s) = + e
2 s s+2 s
y(t) = (et – 1) u(t)
Taking inverse Laplace transform, we have
Q Unbounded.
y(t) = 0.5(1 – e–2t) u(t) + u(t – 6)
 GATE Previous Years Solved Paper 79

32. (b) Taking Laplace transform

s
2 (s + 1) e
F(s) = L[ f (t )] = 2 X(s) =
s + 4s + 7 s
Initial value, Y (s )
= H(s)
2(s + 1) X( s )
lim f (t ) = lim sF(s) = lim s
t 0 s s 2
s + 4s + 7 Y(s) = H(s) X(s)
1 e s e s
1 2 Y(s) = = 3
2s 1 + s2 s s
s
= lim
s 4 7 Taking the inverse Laplace transform,
s2 1 + + 2
s s (t 1)2
y(t) = u (t 1)
2 (1 + 0) 2
= =2
(1 + 0 + 0) 35. (b)
Final value, x(t) = u(t)
2(s + 1) Apply Laplace transform,
lim f (t ) = lim sF(s) = lim s 2
t s 0 s 0 s + 4s + 7 1
X(s) =
s
33. (d)
1
If, L[f(t)] = f (s) H(s) = (given)
s
dn Y(s) = H(s) X(s)
then, L[tn f(t)] = ( 1)n f (s )
dsn 1 1 1
× ==
In this problem: s s s2

1 Taking inverse Laplace transform

Given, L[f(t)] = = f (s) y(t) = tu(t)
s2 + s + 1
We need, 36. (c)
1
d For a causal and stable LTI system the ROC
L[tf(t)] = ( 1)n 1
f (s)
ds must be right sided and it must include the s = j
line ( = 0). For all the poles must lie within
d d 1
= f (s ) =
ds ds s 2 + s + 1 s = 1 then all poles will lie between –1 to +1.
We know that ROC does not include any pole,
1
= × (2s + 1) so in this case ROC will not be able to include
(s2 + s + 1)2
s = j line ( = 0). Thus the system will not be
2s + 1 stable. Hence this statement is not true.
=
(s + s + 1)2
2
37. (b)
x(t) = u(t) – u(t – 2)
34. (c)
h(t) = tu(t) 2s 2s
1 e 1 e
X(s) = =
Taking Laplace transform, s s s
1 Given that,
H(s) =
s2
d2 y dy
x(t) = u(t – 1) +5 + 6 y(t ) = x(t )
2 dt
dt
80 Electronics Engineering Signals & Systems

Taking Laplace transform on both the sides Y(s) = X(s) H(s)

dy 3(s 2)
y(0) = 0 and = 0 at t = 0 Y(s) =
dt (s + 3) (s + 2)
s2Y(s) + 5sY(s) + 6Y(s) = X(s) Using final value theorem, the steady-state value
2s of Y(s) is
1 e
Y(s) [s2 + 5s + 6] = y ( ) = lim sY (s) = 0
s
s 0
2s
1 e
Y(s) = 40. (a)
s (s 2 + 5s + 6)
1
1 e 2s Given, H(s) =
= (s + 1)
s (s + 2) (s + 3)
h(t) = e–t u(t)
At t = , h(t) = 0
38. (a)
h(t)
y(t ) + 5 y(t ) = u(t)
Taking the Laplace transform of the above
equation,
1 t
sY(s) – y(0) + 5Y(s) =
s
i.e. the system is stable as it converges to zero as
1 t .
(s + 5) Y(s) – 1 =
s • The system is causal as it can be seen from
s+1 the graph that, h(t) = 0 for t < 0.
(s + 5) Y(s) =
s
h (t + 1) e (t + 1) e t e 1
• = =
(s + 1) h( t ) e t e t
Y(s) =
s(s + 5)
= e–1 i.e. independent of t.
A B (s + 1) So, statements S1 and S2 are true.
+ =
s (s + 5) s(s + 5)
41. (d)
A = 0.2
1
4 Given, F(s) =
B= = 0.8 s +s+1 2
5
Laplace transform of g(t) = t f(t)
0.2 0.8
Y(s) = + dF(s ) d 1
s (s + 5) = =
ds ds s 2 + s + 1
y(t) = 0.2 + 0.8e–5t
(2s + 1)
=
39. Sol. (s + s + 1)2
2

Given, x(t) = –3e–2t

Laplace transform of 42. (b)
1
x( t ) X(s) H(s) = 2
(s + s 6)
3
where, X(s) = The system is said to be causal if the output at
s+2
any time depends only on present and/or past
s 2
Also, H(s) = values of input or h(t) = 0 for t > 0 i.e. all poles
(s + 3)
must lie on LHS of s-plane.
 GATE Previous Years Solved Paper 81

1 45. (c)
H(s) =
(s + 3) (s 2) Applying Laplace transform,
From the above expression it can be seen that 10
sX(s) – x(0) = 0.2 X(s)
s = 2 is a pole which lies on RHS of s-plane. So, s
H1(s) system that needs to be cascaded should 10 1
have one zero at 2. X(s) = +
s (s + 0.2) (s + 0.2)
1 Applying inverse Laplace transform, we get
Thus, H(s) H1(s) = (s 2)
(s + 3) (s 2) x = 50 – 49 e(–0.2)t
1
= causal system 46. (c)
(s + 3)
1, if a t b
43. Sol. f(t) = = u(t – a) –u(t – b)
0 , otherwise
d 2 y(t ) dy 2 e as
e bs
+ + y ( t ) = x (t )
dt 2 dt F(s) =
s
(s2 + s + 2) Y(s) = X(s)

Y (s) 1 47. (c)

H(s) = = 2 2
X (s) (s + s + ) ROC of a finite duration signal is entire s-plane.
Let another signal g(t) be defined as, 48. Sol.
2
G(s) = H (s ) + sH (s ) + H (s ) x(t) = e–4t u(t) + e4t u(–t)
s
16
2 Given, X(s) = 2 ; –4 < Re[s] < 4
s 16
G(s) = +s+ H (s )
s 16 2 2
= = +
2 (s + 4) (s 4) s + 4 s 4
+ s2 + s 1
= 2 2 x(t) = –2e–4t u(t) – 2e4t u(–t)
s (s + s + )
On comparison,
(s 2 + s + 2
) 1 = –2
= =
s(s 2 + s + 2
) s
So, number of poles of G(s) is 1. 49. Sol.
g(t) = e–t sin (2 t) u(t)
44. (b)
y (t) – y (t) – 6y(t) = x(t) Area = g(t ) dt
Y (s )
H(s) =
X( s )
1 1 = sin(2 t ) u(t ) e t dt = G(s) s = 0
= =
s2 s 6 (s 3) (s + 2)
1 1 2
G(s) =
5 5 (s + 1)2 + 4 2
H(s) = +
s 3 s+2
If the system is neither causal not stable, then, 2
Area = = 0.155
1 3t 1 (s + 1)2 + 4 2 s = 0
2t
h(t) = e u( t ) + e u( t )
5 5
ROC : < –2
82 Electronics Engineering Signals & Systems

50. (b) 52. (b)

T /2 Given, H(s) = es + e–s
1 st
L [f(t)] = sT
e dt H(ej ) = ej + e–j = 2 cos
1 e 0
j
x(t ) H(e ) y(t )
T /2
st
1 e
= sT 2
1 e s If, x1(t) = 2 cos t
0
3
1
= sT
(1 e sT /2 ) 2
s (1 e ) o =
3
1 1 e sT /2 2 1
= H (j o) = 2 cos =2 = 1
s (1 e sT /2 ) (1 + e sT /2 ) 3 2
1 1
= 2
s 1 + e sT /2 y1(t) = 2 cos t + 180°
3
51. (c) If, x2(t) = cos t
For 1st order system, o =
j
H (e ) = 2 cos( ) = –2
o
1
G(s) = ; H (s) = 1 y2(t) = 2 cos( t + 180°)
sT
Impulse response, 2
y(t) = 2 cos t+ 2 cos( t + )
R(s) = 1 3

G(s ) 2
Y(s) = R(s ) = , 2 =
1 + G(s) H (s) 1 3
T1 = 3, T2 = 2
1 1
= = e t /T for t 0 T0 = 6
1 + sT T
2
Unit step response, o = =
T0 3
1
R(s) = y(t) = 2 cos(2 t + ) – 2 cos(3
s ot + )
e j(2 ot + ) +e j(2 ot + )
e j(3 ot + ) j(3 ot + )
1 (1 + sT ) ( sT ) y(t) = e
Y(s) = =
s (1 + sT ) s (1 + sT ) y(t) = e
j(2 ot ) e j(2 ot )
+ e j(3 ot ) +e j(3 ot )

1 T 1 T C3 = 1
= =
s (1 + sT ) s 1
T s+ 53. (d)
T
• A BIBO stable system can have poles in right
y(t) = 1 – e–t/T for t 0
half of complex plane, if it is a non-causal
Ramp response,
system. So, statement-I is wrong.
1
R(s) = • A causal and BIBO stable system should
s2
have all poles in the left half of complex
1 1 T T plane. So, statement-II is correct.
Y(s) = = +
s 2 (1 + sT ) s2 s s+ 1
• So, option (d) is correct.
T
y(t) = t – T(1 – e–t/T) for t 0
 GATE Previous Years Solved Paper 83

54. Sol. 56. Sol.

1 Given that,
H(s) = Integrator
s 1
G(j ) =
sin t (1 + j ) ( + j )
x(t) = u(t )
t 1
G( j ) =
2
t ( + 1) ( 2 + 2 )
sin
So, y(t) = d
According to question,
0
1
G( j ) =3 =
1 sin 5 10
y(t)t = d
0 1 1
=
( 2
+ 1) ( 2
+ 2
) 5 10
sin
d =
2 1 1
0 =
10( a + 9) 2 5 10
1 1
So, y(t)t = × = = 0.5 2 + 9 = 25
2 2
2 = 16

55. (a, c) =4
3 s
Given, G(s) = 57. (0)
(s + 1) (s + 3)
Given that, x1(t) = e–2t u(t)
G(s) 3 s x2(t) = u(t) – u(t – 2)
Y(s) = =
s s(s + 1) (s + 3) y(t) = x1(t) x2(t)
Using partial fractions, we get, By applying Laplace transform,
Y(s) = X1(s) X2(s)
A B C
Y(s) = + + 2s
s (s + 1) ( s + 3) 1 (1 e )
=
A = 1, B = –2 and C = 1 (s + 1) s
1 2 1 By applying final value theorem,
So, Y(s) = +
s (s + 1) (s + 3)
y(t ) t = 1 e 2s
= lim sY (s) = lim =0
and y(t) = u(t) – 2e–t u(t) + e–3t u(t) s 0 s 0 s+1
Forced response,
yf (t) = u(t) option (a)
Some authors also consider the zero state
response (ZSR) as the forced response and
according to them the correct answer will be
option (c).
84 Electronics Engineering Signals & Systems

Answers
EE Laplace Transform

1. (d) 2. (c) 3. (c) 4. (d) 5. (d) 6. (c) 7. (d) 8. (a)

9. (d) 10. (d) 11. (c) 12. (b) 13. (d) 14. (d) 15. (b) 16. (c)

17. (b) 18. (b) 19. (a) 20. (b) 21. (d) 22. (a) 23. (0.707) 24. (b)

25. (b) 26. (a) 27. (d) 28. (a) 29. (a)

Solutions
EE Laplace Transform

1. (d) 2 1
H(s) = +
s + 3 (s + 3)2
Q F(s) =
s2 + 2 2s + 7
T.F. = H(s) =
f(t) = sin( t) u(t) (s + 3)2
which is a periodic signal so final value theorem
6. (c)
is not applicable on it. Final value of f(t) may be
any value between-1 and +1. Final value = lim sY (s)
s 0

2. (c)
7. (d)
L (t2 – 2t) u(t – 1) = L [(t – 1)2 u(t – 1) – u(t – 1)]
5
F(s) =
2e s e s 2
s (s + 3s + 2)
=
s3 s
By initial value theorem,
3. (c) lim f (t ) = lim sF(t )
t 0 s

r (t )
LTI system
c(t ) 5×s
= lim 2
=0
s s (s + 3s + 2)
Unit delay block LTI system
r (t ) r(t 1) C (t 1)
e- s
8. (a)
4. (d) 5s 2 + 23s + 6
F(s) =
f( t) s(s 2 + 2s + 2)
f(t) = u(t) – u(t – T) A F(s) has only left hand poles.
By final value theorem,
6
lim F(t ) = lim sF(s) = =3
t t 0 s 0 2
T
Ts Ts
1 e 1 e 10. (d)
L [u(t) – u(t – T)] = =
s s s
f( t) g (t )
5. (d)
1
1
t
3(t )
y(t) = (2 + t )e u( ) d
0 t t
1 3 5
y(t) = [(2 + t) e–3t] u(t)
 GATE Previous Years Solved Paper 85

Since, g(t) has width of 2 unit and f(t) has 1 unit 14. (d)
therefore we have to first expand f(t) by 2 unit
and for this we have to a scale f(t) by 1/2. d 2 y(t ) dy(t )
2
+2 + y (t ) = ( t )
dt dt
1
i.e., f1(t) = f t
2 s2Y(s) – sy(0) – y (0) + 2[sY(s) – y(0)]+Y(s) = 1
Now shift it by three unit to get g(t), Y(s) [s2 + 2s + 1] – (s × –2) –y (0)–(2 × –2) = 1
g(t) = f1(t – 3) Y(s) [s2 + 2s + 1] = –2s – 3
t 3 t 3 2s 3 2 1
g(t) = f =f Y(s) = =
2 2 2 s2 + 2s + 1 s + 1 (s + 1)2
y(t) = –2e–t – te–t
11. (c)
dy (t )
1, 3<t<5 = 2e–t – (–te–t + e–t) = e–t + te–t
g(t) = dt
0 , t < 3, t > 5
dy(t )
g(t) = u(t – 3) – u(t – 5) = 1+0=1
dt t = 0 +
L{g(t)} = L{u(t – 3)} – L{u(t – 5)}
1 3s 1 5s
= e e 15. (b)
s s
1 1 1
= (e 3s
e 5s
) Y(s) = U (s) = 2
s s s
3s y(t) = t u(t)
e 2s
= (1 e )
s 16. (c)
12. (b) All poles must lie within Z = 1.
F2(t) = L{f(t – )} = e–s F1(s)
s
17. (b)
e F1 ( s ) F1 (s ) s
G(s) = 2
=e 1
F1 (s ) Given, H(s) =
s ( s + 4)
G(t) = (t – )
r(t) = input = cos(3t) = cos t
13. (d) = 3 rad/sec
If, L[f(t)] = F(s) 1
H(j ) =
d n ( j ) ( j + 4)
then, L[tn f(t)] = ( 1)n F( s )
dsn 1 1 1
Now, H( j ) = = =
In this problem: 2
+4 2 3 25 15
1 (at = 3)
d d
Given, L[f(t)] = ( 1)1 1
F(s) = F(s )
ds ds 1
H(j ) = 90° tan
d 1 4
= 2
ds s + s + 1 1 3
= 90° tan
4
1
= × (2s + 1) = –126.86° (at = 3)
(s + s + 1)2
2

2s + 1
=
(s + s + 1)2
2
86 Electronics Engineering Signals & Systems

1 Q Hence, A = 1 × H ( j ) = 2
c(t) = cos(3t 126.86°)
15
1
1 = 1× = 0.707
= sin(3t 36.86°) ...(i) 2
15 and = +45°
or, c(t) = A sin(3t + ) ...(ii)
Comparing equations (i) and (ii), we have 21. (d)

1 The differential equation,

A=
15 dy(t ) 1
+ y(t ) = 3x(t)
dt 6
18. (b)
1
So, sY (s ) + Y (s ) = 3X(s)
t 3/2 6
Given that, f(t) = 2 F( s ) = s
3X(s)
By using property of differentiation in time, Y(s) =
1
df (t ) s+
sF(s ) 6
dt
3
2 1 1/2 X(s) =
t sF(s ) 1
2 s+
3
1 3 /2
s s
t 9 54 54
So, Y (s) = =
1 1 1 1 1
s 1 /2 s+ s+ s+ s+
t 3 6 6 3
So, y(t) = (54e–t/6 – 54e–t/3) u(t)
19. (a)
5 22. (a)
Laplace transform of sin 5t u(t )
s 2 + 25 The differential equation is,
5 5 y (t) + 2y (t) + y(t) = 0
e 2 t sin 5t u(t ) 2
= 2 2
So, (s Y(s) – sy(0) – y (0)) + 2[sY(s) – y(0)] + Y(s)
(s 2) + 25 s 4s + 29
=0
20. (b) sy(0) + y (0) + 2 y(0)
So, Y(s) =
Y (s ) s (s 2 + 2 s + 1)
=
R( s ) s+2 Given that, y (0) = 1, y(0) = 0
y(t) = A cos(2t + ), r(t) = cos2t 1
So, Y(s) =
s (s + 1)2
Q H(s) =
(s + 2) So, y(t) = te–t u(t)
j
H(j ) = ; H( j ) = 23. Sol.
j +2 2
+4
1
H(s) =
H(j ) = 90° tan 1 (s + 1)
2
Put, s=j
Q = 2 (Given)
1
2 2 1 H(j ) =
H( j ) = = = j +1
4+4 2 2 2
1
H ( j ) = 90° – tan–1(1) = 45° H( j ) =
2
+1
 GATE Previous Years Solved Paper 87

Q Input, x(t) = cos(t) 27. (d)

Here, = 1 rad/sec.
dy(t )
and x(t ) = 1 + 4 y(t ) = 6 r(t)
dt 2
Hence, steady state output, [s2 + 4] Y(s) = 6 R(s)
y(t) = H ( j ) = 1 cos(t + H ( j ) = 1 ) Y (s ) 6
= 2
1 R(s ) s +4
A = H( j ) =1
= = 0.707
2 Poles: s2 + 4 = 0
s = ±j2
24. (b)
Steady state value of y(t) 29. (a)
= lim sY (s) Pole locations: (2, –1) and (–2, 1)
s 0 Zero locations: (2, 1) and (–2, 1)
10s j
= lim 2
s 0 s(s + s + 100 2 )
1
10 1
= =
100 2 10 2

25. (b) –2 2
c1
H(s) = 2 (as a1 = b1 = 0)
a2 s + b2 s + c 2
–1
c1
=
(1 + s 1 ) (1 + s 2) Filter is all pass since for each pole, there is a
c1 mirror image zero.
Put s = 0, H(0) =
c2 System is unstable because for stability of causal
Put s = , H( ) = 0 system all poles should lie in the L.H.S. of
s-plane and here one pole is lying in the R.H.S.
At s = 0 At s = Type of filter
[s (2 + j ) [s ( 2 j )]
Constant 0 LPF H(s) =
[s (2 + j ) [s ( 2 + j )]
0 Constant HPF
[s (2 + j ) [s + (2 + j )]
0 0 BPF =
[s (2 j ) [s + ( 2 j )]
1 1 BSF
s2 (2 + j )2 s 2 (4 1 + 4 j )
which represents 2nd order low pass filter. = =
s 2 (2 j )2 s2 (4 1 4 j )
26. (a)
s2 3 4j
=
s+3 s + 1+ 2 s 2
3+ 4j
L 1
= L 1
s2 + 2s + 1 (s + 1)2 i.e. transfer function is complex.

1 1 2
= L +
s + 1 (s + 1)2
= e–t + 2te–t
 6 Z-Transform

ELECTRO NICS EN GINEERIN G Q.4 The z-transform of the time function

(GATE Previous Years Solved Papers)

(n k ) is
k=0
Q.1 Consider the system shown in the figure below.
z 1 z
The transfer function Y(z)/X(z) of the system is (a) (b)
z z 1
y (k ) z
x(k) ( z 1)2
(c) 2 (d)
( z 1) z
Z–1
[EC-1998 : 1 Mark]

Q.5 The z-transform F(z) of the function f(nT) = anT

a is
–b
z z
(a) T (b)
z a z + aT
1 + az 1 1 + bz 1
(a) (b)
1 + bz 1 1 + az 1 z z
(c) T (d) T
1 1 z a z+a
1 + az 1 bz
(c) (d) [EC-1999 : 1 Mark]
1 bz 1 1 + az 1
[EC-1988 : 2 Marks] Q.6 The z-transform of a signal is given by

Q.2 The Z-transform of the following real 1 z 1 (1 z 4 )

C ( z) =
exponential sequence: 4 (1 z 1 )2
x(nT) = an, nT 0 Its final value is
= 0 , nT < 0, a > 0
1
1 1 (a) (b) zero
(a) ; z >1 (b) ; z >a 4
1 1
1 z 1 az (c) 1.0 (d) infinity
1 [EC-1999 : 2 Marks]
(c) 1 for all z (d) 1
; z <a
1 az
Q.7 The region of convergence of the z-transform of
[EC-1990 : 2 Marks] a unit step function is
Q.3 A linear discrete-time system has the (a) z >1
characteristic equation, z3 – 0.81z = 0. The
(b) z <1
system
(c) (Real part of z) > 0
(a) is stable.
(d) (Real part of z) < 0
(b) is marginally stable.
[EC-2001 : 1 Mark]
(c) is unstable.
(d) stability cannot be assessed from the given Q.8 If the impulse response of a discrete-time system
information. is h[n] = –5n u[–n –1], then the system function
[EC-1992 : 2 Marks] H(z) is equal to
 GATE Previous Years Solved Paper 89

z Q.12 The region of convergence of z-transform of the

(a) and the system is stable.
z 5 n n
5 6
z sequence u(n) u ( n 1) must be
(b) and the system is stable. 6 5
z 5
z 5 5
(c) and the system is unstable. (a) z < (b) z >
z 5 6 6

z 5 6 6
(c) < z< (d) < z<
(d) and the system is unstable. 6 5 5
z 5
[EC-2005 : 1 Mark]
[EC-2002 : 2 Marks]
Q.13 If the region of convergence of x1[n] + x2[n] is
Q.9 A sequence x(n) with the z-transform
X(z) = z4 + z2 – 2z + 2 – 3z–4 is applied as an 1 2
< z < , then the region of convergence of
input to a linear, time-invariant system with the 3 3
impulse response h(n) = 2 (n – 3) where,
x1[n] – x2[n] includes
1, n=0 1 2
(n) = (a) < z <3 (b) < z <3
0 , otherwise 3 3
The output at n = 4 is 3 1 2
(c) < z <3 (d) < z<
(a) –6 (b) zero 2 3 3
(c) 2 (d) –4 [EC-2006 : 1 Mark]
[EC-2003 : 1 Mark] Q.14 The z-transform X[z] of a sequence x[n] is given

z 0.5
Q.10 The z-transform of a system is H ( z) = . If by X [ z] = . It is given that the region of
1
z 0.2 1 2z
convergence of X[z] includes the unit circle. The
the ROC is z < 0.2, then the impulse response
value of x[0] is
of the system is
(a) –0.5 (b) 0
(a) (0.2)n u[n]
(c) 0.25 (d) 0.5
(b) (0.2)n u[–n – 1]
[EC-2007 : 2 Marks]
(c) –(0.2)n u[n]
(d) –(0.2)n u[–n – 1] Statement for Linked Answer Questions (15 and 16):
[EC-2004 : 1 Mark] In the following network, the switch is closed at t = 0–
and the sampling starts from t = 0. The sampling
Q.11 A causal LTI system is described by the frequency is 10 Hz.
difference equation,
S 10 µF
2y[n] = y[n – 2] – 2x[n] + x[n – 1]
x(n) X(z )
The system is stable only if 5V 200 k
Sampler
z-transform
(fs = 10 Hz)
(a) = 2, <2

(b) > 2, >2

Q.15 The sample x(n) (n = 0, 1, 2,....) are given by
(c) < 2, any value of
(a) 5(1 – e–0.05n) (b) 5 e–0.05n
(d) < 2, any value of
(c) 5(1 – e–5n) (d) 5 e–5n
[EC-2004 : 2 Marks] [EC-2008 : 2 Marks]
 90 Electronics Engineering Signals & Systems

Q.16 The expression and the region of convergence Consider the following statements:
of the z-transform of the sampled signal are S1 : The system is stable and causal for
5z 5 1
(a) 5
, z <e ROC : z >
z e 2
5z 0.05 S2 : The system is stable but not causal for
(b) 0.05
, z <e
z e
1
5z ROC : z <
0.05 4
(c) 0.05
, z >e
z e S3 : The system is neither stable nor causal for
5z 5
(d) , z >e 1 1
z e 5 ROC : < z<
4 2
[EC-2008 : 2 Marks]
Which one of the following statement is valid?
Q.17 The ROC of z-transform of the discrete-tme (a) Both S1 and S2 are true.

n n
(b) Both S2 and S3 are true.
1 1
sequence, x(n) = u(n) u ( n 1) is (c) Both S1 and S3 are true.
3 2
(d) S1, S2 and S3 are all true.
1 1 1 [EC-2010 : 2 Marks]
(a) < z< (b) z >
3 2 2
Q.21 Two systems H1(z) and H2(z) are connected in
1 cascade as shown below. The overall output
(c) z < (d) 2 < z < 3
3 y(n) is the same as the input x(n) with a one unit
[EC-2009 : 1 Mark] delay. The transfer function of the second system
H2(z) is
Q.18 Consider the z-transform, X(z) = 5z2 + 4z–1 + 3,
1
0 < z < . The inverse z-transform x[n] is x(n) H 1 ( z) =
(1 0.4z )
H2(z) y(n)
1
(1 0.6z )
(a) 5 [n + 2] + 3 [n] + 4 [n – 1]
(b) 5 [n – 2] + 3 [n] + 4 [n + 1]
(1 0.6 z 1 ) z 1 (1 0.6 z 1 )
(c) 5u[n + 2] + 3u[n] + 4u[n – 1] (a) (b)
z 1 (1 0.4 z 1 ) (1 0.4 z 1 )
(d) 5u[n – 2] + 3u[n] + 4u[n + 1]
[EC-2010 : 1 Mark] z 1 (1 0.4 z 1 ) (1 0.4 z 1 )
(c) (d)
Q.19 Two discrete-time systems with impulse (1 0.6 z 1 ) z 1 (1 0.6 z 1 )
responses h1[n] = [n – 1] and h2[n] = [n – 2] are [EC-2011 : 2 Marks]
connected cascade. The overall impulse
response of the cascaded system is n
Q.22 If x [n] = (1/3) (1/2)n u[ n], then the region
(a) [n – 1] + [n – 2] (b) [n – 4]
of convergence (ROC) of its z-transform in the
(c) [n – 3] (d) [n – 1] [n – 2]
z-plane will be
[EC-2010 : 1 Mark]
1 1 1
(a) < z <3 (b) < z<
Q.20 The transfer function of a discrete-time LTI 3 3 2
system is given by
1 1
(c) < z <3 (d) < z
3 1 2 3
2 z
H ( z) = 4 [EC-2012 : 1 Mark]
3 1 1 2
1 z + z
4 8
 GATE Previous Years Solved Paper 91

Q.27 Let H1(z) = (1 – pz–1)–1, H2(z) = (1 – qz–1)–1,

n n
1 1 H(z) = H1(z) + rH2(z). The quantities p, q, r are
Q.23 Let x [n] = u(n) u ( n 1). The
9 3
1 1
region of convergence (ROC) of the z-transform real numbers. Consider p = ,q= , r < 1.
2 4
of x[n] is
If the zero of H(z) lies on the unit circle, then
1 1 r = ________ .
(a) z > (b) z <
9 3
[EC-2014 : 2 Marks]
1 1
(c) > z > (d) does not exist Q.28 The z-transform of the sequence x[n] is given by
3 9
[EC-2014 : 2 Marks] 1
X ( z) = , with the region of
Q.24 Let x[n] = x[–n], X(z) be the z-transform of x[n]. If (1 2 z 1 )2
0.5 + j0.25 is a zero of X(z), which one of the
convergence z > 2. Then, x[2] is ________ .
following must also be a zero of X(z).
1 [EC-2014 : 2 Marks]
(a) 0.5 – j0.25 (b)
(0.5 + j 0.25) Q.29 For the discrete-time system shown in the figure,
1 the poles of the system transfer function are
(c) (d) 2 + j4
(0.5 j 0.25) located at
[EC-2014 : 1 Mark] x[ n] y[n]

Q.25 The input-output relationship of a causal stable

LTI system is given as, 1 5
6 6
y[n] = y[n – 1] + x[n]
If the impulse response h[n] of this system –1 –1
Z Z

satisfies the condition h[n] = 2, the 1

(a) 2, 3 (b) ,3
n=0 2
1 1 1
relationship between and is (c) , (d) 2,
2 3 3
(a) =1 (b) = 1+ [EC-2015 : 2 Marks]
2 2
(c) =2 (d) = –2 Q.30 The pole-zero diagram of a causal and stable
[EC-2014 : 2 Marks] discrete-time system is shown in the figure. The
zero at the origin has multiplicity 4. The impulse
1 response of the system is h[n]. If h[0] = 1, we can
(z b)
Q.26 For an all-pass system H ( z) = , where conclude
(1 az 1 )
Im(z)

j
H(e ) = 1, for all . If Re(a) 0, Im(a) 0,
0.5
then b equals
(a) a (b) a Re(z)
–0.5 0.5
1 1
(c) (d) –0.5
a a
[EC-2014 : 1 Mark]
 92 Electronics Engineering Signals & Systems

(a) h[n] is real for all n. The condition on the filter coefficient that results
(b) h[n] is purely imaginary for all n. in a null at zero frequency is
(c) h[n] is real for only even n. (a) 1 = 2 = 0; 0 =– 3

(d) h[n] is purely imaginary for only odd n. (b) 1= 2 = 1; 0=– 3

[EC-2015 : 2 Marks] (c) 0= 3 = 0; 1= 2

(d) 1= 2 = 0; 0= 3
Q.31 Two causal discrete-time signals x[n] and y[n]
[EC-2015 : 1 Mark]
n
are related as y[n] = x [m]. If the Q.34 Suppose x[n] is an absolutely summable
m=0 discrete-time signal. Its z-transform is a rational
function with two poles and two zeroes. The
2 poles are at z = ±2j. Which one of the following
z-transform of y[n] is , the value of x[2]
z ( z 1)2 statements is true for the signal x[n]?
(a) It is a finite duration signal.
is ________ .
(b) It is a causal signal.
[EC-2015 : 1 Mark]
(c) It is a non-causal signal.
Q.32 A realization of a stable discrete-time system is (d) It is a periodic signal.
shown in figure. If the system is excited by a
[EC-2015 : 2 Marks]
unit step sequence input x[n], the response y[n] is
Q.35 Consider the sequence x[n] = an u[n] + bn u[n],
x[n] where u[n] denotes the unit-step sequence and

–1
0 < a < b < 1. The region of convergence (ROC)
Z
of the z-transform of x[n] is
1 –5/3
y[n] (a) z > a (b) z > b

–1
(c) z < a (d) a < z< b
Z
[EC-2016 : 1 Mark]

–2/9 5/3 Q.36 The ROC (region of convergence) of the

z-transform of a discrete-time signal is
1 n
2 n represented by the shaded region in the z-plane.
(a) 4 u [ n] 5 u [ n] If the signal,
3 3
n n x[n] = (2.0) n , <n<+
2 1
(b) 5 u [ n] 3 u [n] then the ROC of its z-transform is represented
3 3
by
n n
1 2
(c) 5 u [ n] 5 u [ n]
3 3 Img
Unit circle
n n z-plane
2 1
(d) 5 u [ n] 5 u [ n]
3 3
[EC-2015 : 2 Marks] (a) Re
0.5 2

Q.33 Consider a four point moving average filter

3
defined by the equation y[n] = i=0 i x [n i ].
 GATE Previous Years Solved Paper 93

Img Q.39 Let H(z) be the z-transform of a real-valued

Unit circle z-plane 1

discrete-time signal h[n]. If P( z ) = H ( z) H
z

(b) Re 1 1
0.5 2 has a zero at z = + j , and P(z) has a total of
2 2
four zeros, which one of the following plots
represents all the zeros correctly?

Img Img
Unit circle z-plane 2
z-plane
z =1
(c) Re 0.5
0.5 2
(a) Re
–2 –0.5 0.5 2
–0.5

Img
Unit circle z-plane –2

(d) Re Img
0.5 2
2

(ROC does not exist) z-plane

z =1
[EC-2016 : 2 Marks] 0.5

Q.37 A discrete-time signal x[n] = [n – 3] + 2 [n – 5] (b) Re

–2 0.5 2
has z-transform X(z). If Y(z) = X(–z) is the
–0.5
z-transform of another signal y[n], then
(a) y[n] = x[n] (b) y[n] = x[–n]
–2
(c) y[n] = –x[n] (d) y[n] = –x[–n]
[EC-2016 : 1 Mark]

Q.38 A discrete-time all pass system has two of its Img

poles at 0.25 0° and 2 30°. Which one of the 2

following statements about the system is true? z-plane
z =1
(a) It has two more poles at 0.5 30° and 4 0°. 0.5
(b) It is stable only when the impulse response
(c) Re
is two-sided. –2 –0.5 0.5 2
(c) It has constant phase response over all
frequenicies.
(d) It has constant phase response over the
–2
entire z-plane.
[EC-2018 : 1 Mark]
 94 Electronics Engineering Signals & Systems

Img
2 Img
z-plane
1 1
3rd order pole
z =1
0.5

(d) Re –1 1
–2 0.5 1 2
(d) Re

–0.5

–1
–1
–2

[EC-2019 : 1 Mark]
[EC-2020 : 1 Mark]
Q.40 Which one of the following pole-zero
corresponds to the transfer function of an LTI Q.41 The transfer function of a stable discrete-time
system characterized by the input-output K( z )
LTI system is H ( z) = , where K and
difference equation given below? z + 0.5
3
are real numbers. The value of (rounded off to
y[n] = ( 1)k x [n k ]
k=0 one decimal place) with > 1, for which the
Img magnitude response of the system is constant
1 over all frequencies, is ________ .
4th order pole
[EC-2020 : 2 Marks]

–1 1
(a) Re ELECTRICAL EN GINEERIN G
(GATE Previous Years Solved Papers)

–1 Q.1 If u(t) is the unit step and (t) is the unit impulse

1
Img function, the inverse z-transform of R( z ) =
z+1
1
3rd order pole
for k > 0 is
(a) (–1)k (k) (b) (k) – (–1)k u(k)
–1 1
(b) Re (c) (–1)k u(k) (d) u(k) – (–1)k (k)
[EE-2005 : 2 Marks]

Q.2 A discrete real all pass system has a pole at

–1
z = 2 30°, it therefore,
(a) also has a pole at 0.5 30°
Img
1 (b) has a constant phase response over the
3rd order pole
z-plane: arg |H(z)| = constant
(c) is stable only if it anti-causal
–1 1
(c) Re (d) has a constant phase response over the
unit circle: arg |H(e j )| = constant
[EE-2006 : 1 Mark]
–1
 GATE Previous Years Solved Paper 95

Q.3 The discrete-time signal Linked Answer Questions (6 and 7):

3n 2 n Q.6 A signal is processed by a causal filter with

x[n ] X ( z) = n=0 2 + n
z transfer function G(s). For a distortion free output
signal waveform, G(s) must
where denotes a transform-pair relationship,
is orthogonal to the signal. (a) provide zero phase shift for all frequency.
(b) provide constant phase shift for all
n
2 n frequency.
(a) y1 [n] Y1 ( z) = n=0
z
3 (c) provide linear phase shift that is
proportional to frequency.
(b) y2 [n] Y2 ( z) = n=0
(5n n) z (2 n + 1)
(d) provide a phase shift that is inversely
n n proportional to frequency.
(c) y3 [n] Y3 ( z) = n=
2 z
[EE-2007 : 2 Marks]
(d) y4[n] Y4(z) = 2z–4 + 3z–2 + 1
Q.7 G(z) = z–1 + z–3 is a digital band stop filter
[EE-2006 : 2 Marks]
with a phase characteristics same as that of the
Q.4 Consider the discrete-time system shown in the above question if
figure where the impulse response of G(z) is (a) = (b) =–
g(0) = 0, g(1) = g(2) = 1, g(3) = g(4) = .... = 0. (c) = (1/3) (d) = (–1/3)

+
G(z ) [EE-2007 : 2 Marks]
+ Q.8 H(z) is a transfer function of a real system when
a signal x[n] = (1 + j)n is the input to such a
K system, the output is zero.
Further, the region of convergence (ROC) of
The system is stable for range of values of K

1 1 1
(a) 1, (b) (–1, 1) 1 z H ( z) is the entire z-plane
2 2

(except z = 0). It can then be inferred that H(z)

1 1
(c) ,1 (d) ,2 can have a minimum of
2 2
(a) one pole and one zero
[EE-2007 : 2 Marks]
(b) one pole and two zeroes
Q.5 X(z) = 1 – 3z–1, Y(z) = 1 + 2z2 are Z-transforms of (c) two poles and one zero
two signals x[n], y[n] respectively. A linear time (d) two poles and two zeroes
invariant system has the impulse response h[n]
[EE-2008 : 2 Marks]
defined by these two signals as h[n] = x[n – 1]
y[n] where denotes discrete time convolution.
z
Then the output of the system for the input Q.9 Given, X ( z) = with z > a , the residue
[n –1]. ( z a )2
of X(z) zn – 1 at z = a for n 0 will be
(a) has Z-transform z–1 X(z) Y(z)
(a) an – 1 (b) an
(b) equals [n – 2] – 3 [n – 3] + 2 [n – 4] – 6 [n – 5]
(c) nan (d) nan – 1
(c) has Z-transform 1 – 3z–1 + 2z–2 – 6z–3
[EE-2008 : 2 Marks]
(d) does not satisfy any of the above three
[EE-2007 : 2 Marks] Q.10 The Z-transform of a signal x[n] is given by
4z–3 + 3z–1 + 2 – 6z2 + 2z3
 96 Electronics Engineering Signals & Systems

It is applied to a system, with a transfer function N

Y ( z) 1 1 z
H(z) = 3z–1 – 2. Let the output be y(n). Which of = 1
X ( z) N 1 z
the following is true?
(a) y(n) is non-causal with finite support. where, N represents the number of samples per
(b) y(n) is causal with infinite support. cycle. The output y(n) of the system under steady-
(c) y(n) = 0, |n| > 3 state is
(a) 0 (b) 1
(d) Re[Y ( z)] = Re[Y ( z)] ;
z=ej z=e j (c) 2 (d) 5
Im[Y ( z)] = Im[Y ( z)] ;– < [EE-2014 : 2 Marks]
z=ej z=e j

[EE-2009 : 2 Marks] Q.15 Consider a discrete time signal given by:

x[n] = (–0.25)n u[n] + (0.5)n u[–n – 1]
n The region of convergence of its Z-transform
1 1
Q.11 If x[n] = u [n], the region of
3 2 would be
(a) the region inside the circle of radius 0.5 and
convergence (ROC) of its Z-transform in the
centered at origin.
z-plane will be
(b) the region outside the circle of radius 0.25
1 1 1
(a) < z <3 (b) < z< and centered at origin.
3 3 2
(c) the annular region between the two circles,
1 1
(c) < z <3 (d) < z <3 both centered at origin and having radii
2 3
0.25 and 0.5.
[EE-2012 : 1 Mark]
(d) the entire z-plane.
1 [EE-2015 : 2 Marks]
Q.12 Let, X ( z ) = be the Z-transform of a
3
1 z Q.16 The Z-transform of a sequence x[n] is given as
causal signal x[n]. Then, the values of x[2] and 4 3
X ( z) = 2 z + 4 +
x[3] are z z2
(a) 0 and 0 (b) 0 and 1 If y[n] is the first difference of x[n], then Y[z] is
(c) 1 and 0 (d) 1 and 1 given by
[EE-2014 : 2 Marks] 8 7 3
(a) 2 z + 2 +
Q.13 A discrete system is represented by the difference z z2 z3
equation, 6 1 3
(b) 2z + 2 +
X1 ( k + 1) a a 1 X1 ( k ) z z2 z3
=
X 2 ( k + 1) a+1 1 X2 ( k ) 8 7 3
(c) 2z 2 + 2
+
It has initial conditions X1(0) = 1, X2(0) = 0. The z z z3
pole locations of the system for a = 1, are 8 1 3
(d) 4 z 2 2
+
(a) 1 + j0 (b) –1 – j0 z z z3
(c) –1 + j0 (d) 0 + j1 [EE-2015 : 2 Marks]
[EE-2014 : 2 Marks]

Q.17 Let, S = n where < 1. The value of a

Q.14 An input signal x(t) = 2 + 5 sin(100 t) is sampled n
with a sampling frequency of 400 Hz and n=0
in the range of 0 < < 1, such that S = 2a is ___.
applied to the system whose transfer function
is represented by [EE-2016 : 2 Marks]
 GATE Previous Years Solved Paper 97

Q.18 Consider a causal and stable LTI system with Q.20 A cascade system having the impulse responses
rational transfer function H(z), whose
h1 (n) = {1 , 1} and h2 (n) = {1 , 1} is shown in
corresponding impulse response begins at n = 0.

5 the figure below, where symbol denotes the time

Furthermore, H (1) = . The poles of H(z) are:
4 origin.

1 2 k 1) x(n) h1(n) h2(n) y(n)

pk = exp j for k = 1, 2, 3, 4
2 4
The input sequence x(n) for which the cascade
The zeros of H(z) are all at z = 0. Let g[n] = jn h[n]. system produces an output sequence
The value of g[8] equals _____.
y(n) = {1, 2, 1, 1, 2, 1} is
(Given the answer upto 3 decimal places.)
[EE-2017 : 2 Marks]
(a) x(n ) = {1, 2, 1, 1} (b) x(n ) = {1, 1, 2, 2}
Q.19 The pole zero plots of three discrete-time systems
(c) x(n) = {1, 1, 1, 1} (d) x(n ) = {1, 2, 2, 1}
P, Q and R on the z-plane are shown below.
[EE-2017 : 2 Marks]
Im(z) Im(z)

2 poles P Q n
1
0.5 Q.21 Consider a signal x[n] = 1[n], where
2
Re(z) Re(z)
–0.5 1[n] = 0 if n < 0 and 1[n] = 1 if n 0. The
Unit circles Unit circles
z k
(i) (ii) Z-transform of x[n – k], k > 0 is with
1 1
1 z
2
Im(z) region of convergence being
R 1 1
(a) z< (b) z >
2 2
(c) z <2 (d) z >2
Re(z)
[EE-2020 : 1 Mark]
Unit circles
Q.22 Which of the following options is true for a linear
(iii) time-invariant discrete time system that obeys
the difference equation:
Which one of the following is true about the
y[n] – ay[n – 1] = b0 x[n] – b1x[n – 1]
frequency selectivity of these systems?
(a) Where x[n] = 0, n < 0 the function y[n], n > 0
(a) All three are high-pass filters.
is solely determined by the function x[n].
(b) All three are band-pass filters.
(b) The system is necessarily causal.
(c) all three are low-pass filters. (c) y[n] is unaffected by the values of x[n – k],
(d) P is a low-pass filter, Q is a band-pass filter k > 2.
and R is a high-pass filter. (d) The system impulse response is non-zero
[EE-2017 : 1 Mark] at infinitely many instants.
[EE-2020 : 2 Marks]
98 Electronics Engineering Signals & Systems

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Z-Transform

1. (a) 2. (b) 3. (a) 4. (b) 5. (a) 6. (c) 7. (a) 8. (b)

9. (b) 10. (d) 11. (c) 12. (c) 13. (d) 14. (b) 15. (b) 16. (c)

17. (a) 18. (a) 19. (c) 20. (c) 21. (b) 22. (c) 23. (c) 24. (b)

25. (a) 26. (b) 27. (–0.5) 28. (12) 29. (c) 30. (a) 31. (0) 32. (c)

33. (a) 34. (c) 35. (b) 36. (d) 37. (c) 38. (b) 39. (d) 40. (c)

41. (–2)

Solutions
EC Z-Transform

1. (a) 2. (b)
Given that,
+ y1(z) +
x(z) y (z ) x(n) = an u(n) a>0
+ +
Z.T. z
x( n ) X(z) = ; z > a
Z–1 z a
1
X(z) = 1
; z >a
–b a 1 az

y1(z) = x(z) – bz–1 y1(z) 3. (a)

y1(z)[1 + bz–1] = x(z) Given that:
x( z ) Characteristic equation,
y1(z) = z3 – 0.81z = 0
(1 + bz 1 )
z(z2 – 0.81) = 0
y(z) = y1(z) + az–1 y1(z)
z(z – 0.9) (z + 0.9) = 0
y(z) = y1(z) [1 + az–1]
So, poles are z = 0, 0.9 and –0.9
x( z) As all the three poles are inside the unit circle,
y(z) = × (1 + az 1 )
(1 + bz 1 ) so the system is stable.
Transfer function,
4. (b)
1
y( z) 1 + az
H(z) = =
x( z) 1 + bz 1 Given that, x(n) = (n k )
k=0
 GATE Previous Years Solved Paper 99

x(n) = (n) + (n – 1) + (n – 2) +... 8. (b)

x(n) = u(n) h(n) = –5n u[–n – 1]
X(z) = Z.T. [u(n)]
n
z H(z) = h(n) z
X(z) = n=
z 1
1
5. (a) 5n z n
H(z) = Let, n = –k
Z-transform is given by n=

n z z
z[f(nT)] = f (nT ) z = = z<5
n= 5 z z 5
The system is stable because ROC includes unit
= a nT
z n
=
T 1 n
(a z ) circle.
n= n=
9. (b)
1 z
F(z) = = Y(z) = H(z) X(z)
1 aT z 1
z aT
H(z) = 2z –3
6. (c) Y(z) = 2z–3 (z4 + z2 – 2z + 2 – 3z–4)
= 2(z + z–1 – 2z–2 + 2z–3 – 3z–7)
Final value theorem,
Taking inverse of z-transform,
lim x[ N ] = lim (1 z 1 ) X( z) y(n) = 2[ (n + 1) + (n – 1) – 2 (n – 2)
N z 1
+ 2 (n – 3) – 3 (n – 7)]
1 z 1 (1 z 4 )
lim (1 z 1 ) At n = 4,
z 1 4 (1 z 1 )2 y(4) = 0
1 z 1 (1 z 4 )
= lim 10. (d)
z 1 4 (1 z 1 )
z z
H(z) = =
1 ( z2 1) ( z2 + 1) z 0.2 z(1 0.2 z 1 )
= lim
z 1 4 z 4 ( z 1)
1
2 H(z) = 1
1 ( z + 1) ( z 1) ( z + 1) 1 0.2z
= lim
z 1 4 z 4 ( z 1) ROC : z < 0.2
1 Comparing with,
= ×4 =1
4 1
an u( n 1) ; z <a
7. (a) 1 az 1
h(n) = u(n) we get, h(n) = –(0.2)n u(–n – 1)

n 11. (c)
H(z) = 1 z
n=0 2y[n] = y[n – 2] – 2x[n] + x[n – 1]
For the convergence of H(z), Taking z-transform,
2Y(z) = Y(z)z–2 – 2X(z) + X(z)z–1
( z 1 )n < Y(z) [2 – z–2] = X(z) [ z–1 – 2]
n=0
1
Y ( z) z 2
ROC is the range of values of z for which, = 2
X( z ) 2 z
1
z < 1 or z > 1 For system to be stable, can be of any value,
100 Electronics Engineering Signals & Systems

2 – z–2 = 0 16. (c)

2z2 – = 0
0.05n n
X(z) = 5e z
z= <1 n=0
2
<2 = 5 (e 0.05
z 1 )n
For system to be stable all poles should be inside n=0
unity circle. 0.05 1
For e z or z > e 0.05
12. (c)
5z
n n X(z) = 0.05
5 6 z e
z-transform of u(n) u( n 1)
6 5
17. (a)
n n
1 1
1 1 x(n) = u(n) u( n 1)
= + 1 3 2
5 1 6 1
1 z 1 z n
6 5 1
u(n) is right sided signal so ROC will be
(ROC) (ROC) 3

5 6 1
= z > = z < z > ...(i)
6 5 3
5 6
< z< 1 n
6 5 u( n 1) is left sided signal so ROC
2
13. (d) will be
ROC remains the same for addition and 1
z < ...(ii)
subtraction in z-domain. 2
From equation (i) and (ii) we see that ROC of the
14. (b) function will be
0.5 1 1
X(z) = 1 < z <
1 2z 3 2
Q ROC includes unit circle 18. (a)
Left handed system
z
x(n) = –(0.5) (2)n u(–n – 1) [ n + n0 ] zn0
x(0) = 0 X(z) = 5z2 + 4z–1 + 3; 0< z <

15. (b) x(n) = 5 (n + 2) + 4 (n – 1) + 3 (n)

19. (c)
200 × 10 3 5
VR(s) = z
1 s h1[n] = [n 1] H 1 ( z) = z 1
200 × 10 3 +
10 × 10 6 s z
h2[n] = [ n 2] H 2 ( z) = z 2
5 6
5 × 2 × 10 × 10 × 10
= Overall impulse response in z-domain,
2 × 10 5 × 10 5 s + 1
H(z) = H1(z) H2(z)
10 5
= = = z–1 z–2 = z–3
2s + 1 s + 0.5
Overall impulse response in discrete-time
Therefore samples,
domain, h[n] = (n – 3)
x(n) = 5e–0.5n/10 = 5e–0.05n
 GATE Previous Years Solved Paper 101

20. (c) 22. (c)

1. A discrete-time LTI system is causal if and n n
1 1
only if the ROC of its system function is the x[n] = u [ n]
3 2
exterior of circle, including infinity.
n n n
2. A discrete-time LTI system is stable if and 1 1 1
x[n] = u [n] + u [ n 1] u[n]
3 3 2
only if the ROC of its system function
includes the unit circle, 1 n 1 1
u [n] ROC : z >
3 1 1; 3
1 1 1 1 1 z
1 z + 1 z 3
4 2
H(z) =
1 1 1 1 1 n
1
1 z 1 z u [ n 1] ; ROC : z < 3
4 2 3 1
1 3z
1 1
or, H(z) = + 1 n 1 1
1 1 u [n] ROC : z >
1 z 1
1 z 1
2 1 1 ; 2
4 2 1 z
2
1 So overall ROC will be intersection of there ROCs
For ROC : z > , the system is stable and
i.e.,
2
causal. 1
< z <3
2
1
For ROC : z > , ROC does not include 23. (c)
4
unit circle. So, system is not stable. 1 n 1 n
x(n) = u(n) u( n 1)
1 1 9 3
For ROC : < z < , ROC does not x1 ( n ) x2 ( n )
4 2
n
include unit circle. So, system is not stable. 1
x1(n) = u(n) Right sided signal
9
1
Also ROC is not the exterior of z = . So, it
1
2
X1(z) =
1
is not causal. 1 z 1
9
21. (b)
1 1
y[n] = x[n – 1] ROC : z or z >
9 9
Taking z-transform of both sides,
n
Y(z) = z–1 X(z) 1
x2(n) = u ( n 1) Right sided signal
Y ( z) 3
= z –1
X( z ) 1 1
X2(z) = ; z
For cascaded system, 1 3
1 z 1
H(z) = H1(z) H2(z) 3
(1 0.4 z 1 ) Now, when both the transforms are added using
z –1 = H 2 ( z)
(1 0.6 z ) 1 linearity property.
1 1
z 1 (1 0.6 z 1 ) ROC must be between > z > .
H2(z) = 3 9
(1 0.4 z 1 )
102 Electronics Engineering Signals & Systems

24. (b) 1 1 1
= j a
=
Given, x[n] = x[–n] b ae a
As we know that, or, b=a
z
x[ n] X[ z ]
27. Sol.
z 1
x[ n] X[ z ] 1
Here, H1(z) =
So, when (0.5 + j0.25) is a zero fo X(z). 1 pz 1
Now, x[ n] X( z 1 ) 1
H2(z) =
1 1 qz 1
Zero of X ( z) =
(0.5 + j0.25)
1 1
Also, p = ; q=
25. (a)
2 4
H(z) = H1(z) + rH2(z)
Input-output relation is given as,
y[n] = y(n – 1) + x[n] 1 r
= 1
+
Taking z-transform, 1 pz 1 qz 1
Y(z) = z–1 Y(z) + X(z) 1 r
Y(z) [1 – z–1] = X(z) = +
1 1 1 1
1 z 1+ z
Y ( z) 2 4
= H ( z) = 1
X( z ) 1 z
1 1 1 1
h[n] = an u[n] 1+ z +r 1 z
4 2
H(z) = ...(i)
1 1 1
Alos, h[n] = 2 n
u[n] = 2 1 z 1+ z 1
2 4
n=0 n=0
zero of H(z) (1 + r) – (q + pr) z–1 = 0 ...(ii)
=2 =2–2 q + pr
1 or, z=
2 = 2– 1+r
Since, the zero of H(z) lies on the unit circle.
= 1
2 Therefore, z = 1 or z = ±1
Taking, z = 1, we get
26. (b)
q + pr
For the given system, z= =1 ...(iii)
1+r
1
(z b)
H(z) = 1 ...(i) 1
(1 az ) By solving above expression with p = and
2
1
The pole lie at z = a and the zero lie at z = . 1
b q= , we get
Also ‘a’ is complex in nature. 4

j a r = –2.5
Thus, a= ae
But, r < 1 or –1 < r < 1
For an all pass system where H ( e j ) = 1, for Taking, z = –1 in equation (iii), we get
r = –0.5
all , if the pole lies at ‘a’ then the zero must be
present at 1/a .
 GATE Previous Years Solved Paper 103

28. Sol. 2 z
Given, Y(z) = 2
= X ( z)
1 z( z 1) z 1
Given, X(z) = 1
; z >2
(1 2 z )
2 z 2 z
X(z) = (1 – 2z–1)–2 X(z) = = 2z 3
z 1 z 1
X(z) = 1 + 4z–1 + 12z–2 +...
x[n] = 2u[n – 3]
(using binomial expansion)
Therefore, x[2] = 0
Taking inverse z-transform, we get
x[n] = {1, 4, 12, ... 32. (c)

x[2] = 12 5
Y ( z) [1 z]
3
29. (c) =
X( z ) 2
z2 z +
The difference equation of the system, 9
1 5 (1 z)
x( n ) y(n 2) + y(n 1) = y(n) 5
6 6 Y(z) =
3 1 2
z z
Y ( z) 1 3 3
H(z) = =
X ( z) 1 + 1 z 2 5 z 1
6 6 5z 5z
Y(z) =
1 2
1 1 z z
Poles are at, z= , 3 3
2 3
n n
30. (a) 1 2
y(n) = 5 u(n) 5 u(n)
3 3
z4
H(z) =
1
2
1 1
2
1 33. (a)
z + z+ +
2 4 2 4 3
y(n) = i x(n i )
1 1 1 1 i=0
Poles = ± j , ±j
2 2 2 2 = 0x[n]
+ 1x[n – 1] + 2x[n – 2]
z4 +3 x[n – 3]
=
1 1 Y(z) = 0X(z) + 1X(z)z–1 + 2X(z)z–2
z2 z+ z2 + z +
2 2 + 2X(z)z–2 + 3X(z)z–3
z4 H(z) = 0 + 1z–1 + 2z–2 + 3z–3
= 1 2 Null at zero frequency
z 4 + z2 + z
4 j
z= e
=0
z4 1 4
H(z) = =1 z 0 = 0+ 1+ 2+ 3
1 4
z4 +
4 Option (a) satisfies the condition.
So answer is (a).
34. (c)
31. Sol. Poles are at z = ±2j
n Since x[n] is absolutely summable hence, DTFT
y[n] = x[m] = x[n] u[n] exists which implies ROC must include unit
m=0 circle.
z x(t) is a non-causal signal.
Y(z) = X( z)
z 1
104 Electronics Engineering Signals & Systems

35. (b) 39. (d)

Given, x[n] = an u[n] + bnu[n] 1
P(z) = H ( z) H
Also given, 0 < a < b <1 z

ROC = ( z > a ) and ( z > b ) (i) h(n) is real. So, p(n) will be also real.
(ii) P(z) = P(z–1)
ROC = z > b
From (i) : If z1 is a zero of P(z), then z1 will be
36. (d)
also a zero of P(z).
n
Given, x[n] = (2.0) , <n<+ From (ii) : If z1 is a zero of P(z), then 1/z1 will be
n also a zero of P(z).
1
= 2 n u [ n] + u [ n 1] So, the 4 zeros are,
2
1 1
1 z1 = + j
ROC : z > 2 and z < 2 2
2
1 1
No ROC z 2 = z1 = j
2 2
37. (c) 1 1
z3 = = =1 j
Given, x[n] = [n – 3] + 2 [n – 5] z1 1 1
+ j
2 2
X(z) = z–3+ 2z–5
X(–z) = (–z)–3 + 2(–z)–5 1
Y(z) = X(–z) = –z–3 – 2z–5 z4 = = z3 = 1 + j
z1
= –[z–3 + 2z–5]
So, option (d) is correct.
y[n] = –x[n]
40. (c)
38. (b)
3
y(n) = ( 1)k x (n k )
Im
k=0
2 = x(n) – x(n – 1) + x(n – 2) – x(n – 3)
1 Y(z) = X(z) – z–1 X(z) + z–2 X(z) – z–3 X(z)
Y ( z) 1 2 3
H(z) = =1 z +z z
30° X( z)
Re
0.25
z3 z2 + z 1 ( z 1) ( z 2 + 1)
= =
z3 z3
Unit circle Pole zero plot :

Img
1
The ROC should include unit circle to make the 3rd order pole

system stable. From the given pole pattern it is

clear that, to make the system stable, the ROC –1 1
Re
should be finite circular strip or finite annular
strip and hence the impulse response of the
system should be also two sided.
–1
 GATE Previous Years Solved Paper 105

41. Sol. By given transfer function,

System is all-pass filter. Zero =
For digital all-pass filter, condition is Using condition (i),

1 1
Zero = ...(i) = = 2
Pole 0.5

Answers
EE Z-Transform

1. (b) 2. (c) 3. (b) 4. (a) 5. (b) 6. (c) 7. (a) 8. (d)

9. (d) 10. (a) 11. (c) 12. (b) 13. (a) 14. (c) 15. (c) 16. (a)

17. (0.29) 18. (0.097) 19. (b) 20. (d) 21. (b) 22. (d)

Solutions
EE Z-Transform

1. (b)
(2 n + 1)
y( n) Y2 ( z) = (5n n) z
z 1 n=0
F(z) = 1 =1 1
z+1 1+ z In Y2(z), powers of ‘z’ are odd. Therefore samples
= (k) – (–1)k u(k) in y2(n) are available only at odd instants of
time.
2. (c)
For causal system, if all the poles are inside the, Hence, x(n) y2 (n) = 0
n=

2 Thus, x(n) is orthogonal to y2(n).

r 2 30°
4. (a)
Given, g(1) = g(2) = 1, otherwise 0
i.e., g[n] = [n – 1] + [n – 2]
Therefore, G(z) = z–1 + z–2
Therefore, overall transfer function of closed-
Unit circle then system is stable, and converse loop system,
in true for anti-causal system. G( z)
T(z) =
3. (b) 1 k G( z)
G( z)
3n 2n T(z) =
x(n) X( z) = z 1 k G( z)
n=0 2+n
z+1
In X(z), power of ‘z’ are even. Therefore, samples or, T(z) = 2
z k( z + 1)
in x(n) are available at even instants of time. By
So, the system will be stable if it’s outer most
observing all the options.
pole will lie inside the unit circle.
106 Electronics Engineering Signals & Systems

Location of poles, For linear phase FIR filter, g(n) should be either
even symmetric or odd symmetric about virtual
+ k ± k2 + 4 k
z= y-axis.
2
To, satisfy the above condition.
k + k2 + 4 k Case (i): =–
<1
2 (odd symmetric about virtual y-axis)
But in this case,
k + k2 + 4 k < 2 G(1) = G(–1) = 0
i.e. at low frequency G(z) = 0
k 2 + 4 k < (2 – k)
and at high frequency G(z) = 0
k2 + 4 k < (2 – k)
Thus in this case filter will be band-pass.
k2 + 4 k < 4 + k2 – 4 k
8k < 4 Case (ii): =
(even symmetric about virtual y-axis)
1
k< In this case:
2
At low frequency (z = 1) G(1) = 2
This condition is satisfied in only option (a).
At low frequency (z = –1) G(–1) = –2
5. (b) Filter will be either all pass or band stop.
X(z) = (1 – 3z–1) Now, G(e j ) = (e–j + e–j3 )
Y(z) = (1 + 2z–1) = e–j2 (ej + e–j )
h[n] = x(n – 1) y[n] = 2 e–j 2 cos
H[z] = z–1 X(z) Y(z)
9. (d)
H[z] = z–1 (1 – 3z–1) (1 + 2z–2)
H[z] = z–1 (1 + 2z–2 – 3z–1 – 6z–3) z
x(z) =
H[z] = (z–1 – 3z–2 + 2z–3 – 6z–4) ( z a)2
when input, zn
I(n) = [n – 1] zn – 1 X(z) =
( z a)2
then, I[z] = z –1
Since, z = a is a pole of second order therefore
Therefore output,
residue at z = a,
P(n) = h[n] I[n]
P(z) = H(z) I(z) 1 d zn
= ( z a)2
P(z) = (z–1 – 3z–2 + 2z–3 – 6z–4) × z– 1 dz ( z a )2
at z = a
1
= [nzn – 1]z = a = nan – 1
P(z) = z–2 – 3z–3 + 2z–4 – 6z–5
P(n) = [n – 2] – 3 [n – 3] + 2 [n – 10. (a)
4] y[n] = x[n] H[n]
– 6 [n – 5] y[z] = x[z] H[z]
y[z] = (4z–3 + 3z–1 + 2 – 6z2 + 2z3)
7. (a)
(3z–1 – 2)
G(z) = z–1 + z–3
= 12z–4 + 9z–2 + 6z–1 – 18z + 6z2
For, low frequency (z = 1) G(1) = + ...(i)
–8z–3 – 6z–1 – 4 + 12z2 – 4z3
For, low frequency (z = 1) G(–1) = –( + )
= –4z3 + 18z2 – 18z + 12z–4 + 9z–2
...(ii)
– 4 – 8z–3
For now, g(n) = {0, , 0, } y[n] 0 for n < 0
Therefore it is non-causal with finite support.
 GATE Previous Years Solved Paper 107

11. (c) Applying z-transform on equation (i),

n
zX1(z) – zX1(0) = X1(z)
1
Let, x1[n] = X1(z) [z – 1] = zX1(0)
3
z
n X1(z) = [Given, X2(0) = 0]
1 z 1
and x2[n] = u [ n]
2 ...(iii)

1 n
1 n X 2 ( z) 2
x1[n] = u [ n] + u [ n 1] =
3 3 X1 ( z) z 1
Thus, transfer function,
1 n z 1 1
u [n] ; ROC : z > 2
3 1 1 3 H(z) =
1 z z 1
3
n
Therefore, pole location is z = 1.
1 z 1
u [ n 1] 1
; ROC : z < 3
3 1 1 14. (c)
1 z
3
LTI
x(t) Sampler y(n)
1 n z 1 1 x(n) system
and x2 [n] = u [ n]
1 1
; ROC : z >
2 1 z 2 H(z )
2
f s = 400 Hz
1
ROC is < z < 3. x(t) = 2 + 5 sin(100 t)
2
n n
t xTs = =
12. (b) f s 400
1 n
X(z) = 3
x(n) = 2 + 5sin 100
1 z 400
From z-transform definition,
= 2 + 5 sin n
4
n
X(z) = x[n] z
n= Here, 0 =
4
= x(0) + x(1) z–1 + x(2) z–2 + x(3) x–3... 2 2
Therefore, N = time period of x(n) = = =8
1 /4
X(z) = = 1 + 0 z 1 + 0 z 2 + 1 z 3 ... 0
1 z 3 N
1 1 z
By comparison x(2) = 0 and x(3) = 1. Now, H(z) = 1
N 1 z
13. (a) Applying, z = ej
Given that,
1 1 e j N
X1 (K + 1) a a 1 X1 ( K ) H(ej ) =
N 1 e j
X 2 (K + 1) = a+1 a X 2 (K )
For sinusoidal part of x(n):
With initial conditions,
X1(0) = 1; X2(0) = 0 1 1 e j 0N
H(e j 0 ) =
For a = 1 we can write, N 1 e j 0

X1 (K + 1) 1 0 X1 (K )
= Note: 0N = ×8 = 2
X 2 (K + 1) 2 1 X 2 (K ) 4
X1(K + 1) = X1(K) ...(i) j2 N
1 1 e
= =0
X2(K + 1) = 2X1(K) + X2(K) ...(ii) N 1 e j /40
108 Electronics Engineering Signals & Systems

Therefore for sinusoidal part of input, system

n
output is zero. For dc part of input, Since, n = 2 =
n=0 (1 )2
y(n) = Output
= H(ej0) × Input dc value 1
So, 2 = = 0.29
(1 )2
H(e j ) =1
=0
18. Sol.
= 1×2=2
Pole location of H(z) are given as,
Thus, steady-state output = 2
1 j(2 K 1) /4
z= e ; K = 1, 2, 3, 4
15. (c) 2
x[n] = (–0.25)n u(n) + (0.5)n u(–n – 1) 1 j /4 1
z1 = e = (1 + j )
Signal x[n] is sum of two signals, one is right 2 2
sided [(–0.25)n u(n)] and other is left sided 1 j3 /4 1
z2 = e = [ 1 + j]
[(0.5)n u(–n – 1)]. The right sided signal will have 2 2
pole at location with magnitude 0.25. So, ROC
1 j5 /4 1
z3 = e = [ 1 j]
is z > 0.25. The left sided signal will have pole 2 2
at location with magnitude 0.5. So, ROC is 1 j7 /4 1
z4 = e = [1 j ]
z < 0.5. So, ROC of X(z) (Z-transform of x(n) 2 2
Now,
will be) 0.25 < z < 0.5.
K z4
H(z) = ...(i)
16. (a) ( z z1 ) ( z z2 ) ( z z3 ) ( z z4 )
y(n) is first difference of x(n). [As h(n) is causal and it starts from n = 0, so
So, y(n) = x(n) – x(n – 1) numerator will have same order as denominator
is having].
So, Y(z) = X(z) – Z–1 X(z)
By solving equation (i),
Y(z) = (2z + 4 – 4z–1 + 3z–2)
–(2 + 4z–1 – 4z–2 + 3z–3) Kz 4
H(z) = ...(ii)
1
Y(z) = 2z + 2 – 8z–1 + 7z–2 – 3z–3 z4 +
4
17. Sol. 5
Given that, H(1) =
The Z-transform of 4
From equation (ii),
n 1
u(n) K 5 K
(1 Z 1) H(1) = =
1 4 5
1+
Z 1 4 4
n
and n u(n )
(1 Z 1 )2 25
K =
16
1
Z n n 25 4
so, = n Z z
(1 Z 1 )2 n=0 16
H(z) = 1
If we put Z = 1 in above equation we get, z4 +
4
n n By using division rule,
2 =
(1 ) n=0 25 1 4 1 8
H(z) = 1 z + z + ...
16 4 16
 GATE Previous Years Solved Paper 109

1 25 Now, h(n) = overall system impulse response

Thus, h(8) = × = 0.097
16 16 = h1(n) h2(n)
It is given that, = {1, 1} {1, 1}
g(n) = jn h(n)
2
g(8) = j8 h(8) = h(8) = {1, 0, 1} H ( z) = 1 z
g(8) = h(8) = 0.097
At we know,
19. (b) Y ( z) Y ( z)
H(z) = X( z) =
For Fig. (i): X( z) H ( z)

K ( z2 1) 1 + 2z 1 + z 2 z 2 2z 4 z 5
H(z) = =
z2 1 z 2
At low frequency, i.e. z = 1 = 1 + 2z–1 + 2z–2 + z–3
H(1) = 0 x(n) = {1, 2, 2, 1}

At high frequency, i.e. z = –1 21. (b)

H ( 1) = 0 n
1 1
Hence filter type is BPF. x(n) = u(n) ; ROC of x(n) : z >
2 2
For Fig. (ii):
z k
2 x( n k ) X( z) =
K(z 1) 1 1
H(z) = 2
1 z
z + 0.25 2
At low frequency, i.e. z = 1 1
ROC of x(n – k) : z >
H(1) = 0 2
At high frequency, i.e. z = –1 1
For x(n – k) ROC will be z > .
2
H ( 1) = 0
Hence filter type is BPF. 22. (d)
y(n) – ay(n – 1) = b0 x(n) – b1 x(n – 2)
For Fig. (iii):
By applying ZT,
z2 1 Y(z) – az–1 Y(z) = b0 X(z) – b1 z–1 X(z)
H(z) = 2
z +1
Y ( z) b0 b1 z 1
At low frequency, i.e. z = 1 H(z) = =
X( z) 1 az 1
H(1) = 0
By taking right sided inverse ZT,
At high frequency, i.e. z = –1 h(n) = b0an u(n) – b1an – 1 u(n – 1)
H ( 1) = 0 By taking left sided inverse ZT,
h(n) = –b0an u(–n – 1) + b1an – 1 u(–n)
Hence filter type is BPF.
Thus, system is not necessarily causal.
20. (d) The impulse response is non-zero at infinitely
many instants.
x(n) h1(n) h2(n) y(n)

h(n)
 7 DTFS, DTFT and DFT

ELECTRO NICS EN GINEERIN G (a) E-3, F-2, G-4, H-1

(GATE Previous Years Solved Papers) (b) E-1, F-3, G-2, H-4
(c) E-1, F-2, G-3, H-4
Q.1 The impulse response h[n] of a linear time (d) E-2, F-1, G-4, H-3
invariant system is given as, [EC-2005 : 2 Marks]
2 2, n = 1, 1
Statement of Linked Answer Questions (4 and 5):
h[n ] = 4 2, n = 2, 2 A sequence x(n) has non-zero values as shown in the
0, otherwise figure.
If the input to the above system is the sequence x(n)
e j n/4, then the output is
2
(a) 4 2 e j n /4 (b) 4 2 e j n /4 1 1

(c) 4e j n/4 (d) –4e j n/4 1/2 1/2

[EC-2004 : 2 Marks]
n
n –2 –1 0 1 2
1 2
Q.2 Let x (n) = u (n), y(n) = x (n) and Y(ej ) be
2
Q.4 The sequence,
the Fourier transform of y(n). Then Y(ejo) is
n
(a) 1/4 (b) 2 x 1 ; for n even
y(n ) = 2
(c) 4 (d) 4/3 0; for n odd
[EC-2005 : 1 Mark]
will be
Q.3 Match the following and choose the correct
combination:
2
Group-I
1 1
E. Continuous and aperiodic signal
F. Continuous and periodic signal 1/2 1/2
(a)
G. Discrete and aperiodic signal
n
H. Discrete and periodic signal –2 0 2 4 6
Group-II
1. Fourier representation is continuous and
aperiodic. 2

2. Fourier representation is discrete and 1 1

aperiodic. 1/2 1/2

(b)
3. Fourier representation is continuous and
periodic. n
–3 –1 1 3 5
4. Fourier representation is discrete and
periodic.
 GATE Previous Years Solved Paper 111

Q.8 {x(n)} is a real-valued periodic sequence with a

period N. x(n) and X(k) form N-point Discrete
2
Fourier Transform (DFT) pairs. The DFT Y(k) of
1 1
N 1
1/2 1/2 1
(c) the sequence, y (n) = x(r ) x(n + r ) is
N r =0
n
–6 –4 –2 0 2
2
(a) X( k )

N 1
2 1
(b) X(r ) X( k + r )
1 1 N r =0

1/2 1/2 N 1
(d) 1
(c) X(r ) X( k + r )
N r =0
n
–5 –3 –1 1 3
(d) 0 [EC-2008 : 2 Marks]
[EC-2005 : 2 Marks]
Q.9 A system with transfer function H(z) has
Q.5 The Fourier transform of y(2n) will be impulse response h(n) defined as h(2) = 1,
(a) e–j2 [cos4 + 2 cos2 + 2] h(3) = –1 and h(k) = 0 otherwise. Consider the
(b) [cos2 + 2 cos + 2] following statements:
(c) e–j [cos2 + 2 cos + 2] S1 : H(z) is low-pass filter.
(d) e–j /2 [cos2 + 2 cos + 2] S2 : H(z) is an FIR filter.
[EC-2005 : 2 Marks] Which of the following is correct?
(a) Only S2 is true.
Q.6 A signal x(n) = sin( 0n + ) is the input to a linear
(b) Both S1 and S2 are false.
time-invariant system having a frequency
(c) Both S1 and S2 are true, and S2 is a reason
response H(e j ). If the output of the system
for S1.
Ax(n – n0), then the most general form of H(e j )
will be (d) Both S1 and S2 are true, and S2 is not a reason
for S1.
(a) –n0 0+ for any arbitrary real .
[EC-2009 : 2 Marks]
(b) –n0 0+2 k for any arbitrary integer k.
(c) n0 0+2 k for any arbitrary integer k. Q.10 The 4-point Discrete Fourier Transform (DFT)
(d) –n0 0
of a discrete time sequence {1, 0, 2, 3} is
[EC-2005 : 2 Marks] (a) [0, –2 + 2j, 2, –2 – 2j]
(b) [2, 2 + 2j, 6, 2 – 2j]
Q.7 A 5 point sequence x[n] is given as, x[–3] = 1,
(c) [6, 1 – 3j, 2, 1 + 3j]
x[–2] = 1, x[–1] = 0, x[0] = 5, x[1] = 1. Let X(ej )
denote the discrete-time Fourier transform of (d) [6, –1 + 3j, 0, –1 – 3j]
[EC-2009 : 2 Marks]

x[n]. The value of X( e j ) d is Q.11 For an N-point FFT algorithm with N = 2m, which
one of the following statements is true?

(a) 5 (b) 10 (a) It is not possible to construct a signal flow

graph with both input and output in normal
(c) 16 (d) 5 + j10
order.
[EC-2007 : 2 Marks]
 112 Electronics Engineering Signals & Systems

(b) The number of butterflies in the mth state is Q.16 The sequence x[n] = 0.5n u[n] where u[n] is the
N/m. unit step sequence, is convolved with itself to
(c) In place computation requires storage of
only 2 N node data. +
obtain y[n]. Then y[n] is ______ .
(d) Computation of a butterfly requires only one n=
complex multiplication.
[EC-2014 : 1 Mark]
[EC-2010 : 1 Mark]
Q.17 A Fourier transform pair is given by
Q.12 The first six points of the 8-point DFT of a real
Ae + j 6
valued sequence are 5, 1 –j3, 0, 3, –j4, 0 and n f
2 F.T.
u [n + 3]
3 – j4. The last two points of the DFT are 3 2 j2 f
1 e
respectively. 3
(a) 0, 1 – j3 (b) 0, 1 + j3
when u[n] denotes the unit step sequence. The
(c) 1 + j3, 5 (d) 1 – j3, 5
value of A is _______ .
[EC-2011 : 2 Marks]
[EC-2014 : 1 Mark]
Q.13 The DFT of a vector [a b c d] is the vector [ ].
Q.18 The N-point DFT of a sequence x[n], 0 n N – 1
Consider the product:
is given by
a b c d
N 1 2
d a b c 1 j nK
[p q r s] = [ a b c d ] X[K ] = x [n] e N , 0 K N 1
c d a b N n=0
b c d a
Denote this relation as X = DFT(x). For N = 4,
The DFT of the vector [p q r s] is a scaled version
which one of the following sequence satisfies
of
DFT (DFT (x)) = x.
(a) [ 2 2 2 2]
(a) x = [1 2 3 4] (b) x = [1 2 3 2]
(b) (c) x = [1 3 2 2] (d) x = [1 2 2 3]
(c) [ + + + + ] [EC-2014 : 2 Marks]
(d) [ ] [EC-2013 : 2 Marks] Q.19 Two sequence [a, b, c] and [A, B, C] are related
Q.14 Consider a discrete time periodic signal as,

A 1 1 1 a
n
x [n] = sin . Let ak be the complex Fourier 1 2
5 B 1 W3 W3 2 b j
where, W3 = e 3
C 1 W3 2 W3 4 c
series coefficients of x[n]. The coefficients {ak}
are non-zero when k = Bm ± 1, where m is any If another sequence [p, q, r] is derived as,
integer. The value of B is ______ .
[EC-2014 : 2 Marks] p 1 1 1 1 0 0 A /3
q = 1 W31 W32 0 W32 0 B /3
Q.15 An FIR system is described by the system
c 1 W32 W34 0 0 W34 C /3
function:
7 1 3 2 Tthen the relationship between the sequences
H ( z) = 1 + z + z
2 2 [p, q, r] and [a, b, c] is
The system is (a) [p, q, r] = [b, a, c] (b) [p, q, r] = [b, c, a]
(a) maximum phase (b) minimum phase (c) [p, q, r] = [c, a, b] (d) [p, q, r] = [c, b, a]
(c) mixed phase (d) zero phase [EC-2015 : 2 Marks]
[EC-2014 : 1 Mark]
 GATE Previous Years Solved Paper 113

Q.20 Consider two real sequences with time-origin Q.25 A continuous-time speech signal xa(t) is sampled
marked by the bold value: at a rate of 8 kHz and the samples are
x1[n] = {1, 2, 3, 0}, x2[n] = {1, 3, 2, 1} subsequently grouped in blocks, each of size N.
Let X1(k) and X2(k) be 4-point DFTs of x1[n] and The DFT of each block to be computed in real
x2[n], respectively. time using the radix-2 decimation-in-frequency
Another sequence x3[n] is derived by taking FFT algorithm. If the processor performs all
4-point inverse DFT of X3(k) = X1(k) X2(k). The operations sequentially, and takes 20 µs for
value of x3[2] is _______ . computing each complex multiplication
(including multiplications by 1 and –1) and the
[EC-2015 : 2 Marks]
time required for addition/subtraction is
n negligible, then the maximum value of N is
1
Q.21 The value of n is ________ . _________ .
n=0 2
[EC-2016 : 2 Marks]
[EC-2015 : 1 Mark]
Q.26 Let h[n] be the impulse response of a discrete-
time linear time invariant (LTI) filter. The
n
Q.22 Let x [n] = 1 + cos be a periodic signal impulse response is given by
8
1 1 1
with period 16. Its DFS coefficients are defined h [0] = ; h [1] = ; h [2] =
3 3 3
15 and h[n] = 0 for n < 0 and n > 2
1
by ak = x [n] exp j kn for all k. The
16 n=0 8 Let H( ) be the Discrete-Time Fourier Transform
(DTFT) of h[n], where is the normalized
value of the coefficient a31 is ______ . angular frequency in radians. Given that,
[EC-2015 : 2 Marks] H( o) = 0 and 0 < o < , the value of o
(in radians) is equal to ________ .
Q.23 Consider the signal:
[EC-2017 : 2 Marks]
x[n] = 6 [n + 2] + 3 [n + 1] + 8 [n] + 7 [n – 1]
+ 4 [n – 2] Q.27 An LTI system with unit sample response
If X(e j ) is the discrete-time Fourier transform of h[n] = 5 [n] – 7 [n – 1] + 7 [n – 3] – 5 [n – 4] is a

1 (a) low pass filter

x[n], then X ( e j ) sin 2 (2 ) d is equal to (b) high pass filter
(c) band pass filter
_______ .
(d) band stop filter
[EC-2016 : 2 Marks]
[EC-2017 : 1 Mark]
Q.24 The Discrete Fourier Transform (DFT) of the
4-point sequence: Q.28 Let X[k] = k + 1, 0 k 7 be 8-point DFT of a

x[n] = {x[0], x[1], x[2], x[3]} = {3, 2, 3, 4} is N 1

j 2 nk / N
sequence x[n], where X [ k ] = x [ n] e .
X[k] = {X[0], X[1], X[2], X[3]} = {12, 2j, 0, –2j} n=0
If X1[k] is the DFT of the 12-point sequence
x1[n] = {3, 0, 0, 2, 0, 0, 3, 0, 0, 4, 0, 0}, the value of The value (correct to two decimal places) of

3
X1 [8]
x [2n] is _________ .
X1 [11] is ________ . n=0

[EC-2016 : 2 Marks] [EC-2018 : 2 Marks]

 114 Electronics Engineering Signals & Systems

Q.29 It is desired to find a three-tap causal filter 2

(a) a1 = 1, a2 = W6 , a3 = W6
which gives zero signal as an output to an input
2
of the form (b) a1 = 1, a2 = W6 , a3 = W6

j n j n 2
x [n] = c1 exp + c2 exp (c) a1 = 1, a2 = W6 , a3 = W6
2 2
2
(d) a1 = 1, a2 = W6 , a3 = W6
where c1 and c2 are arbitrary real numbers. The
[EC-2019 : 2 Marks]
desired three-tap filter is given by
h[0] = 1, h[1] = a, h[2] = b Q.31 A finite duration discrete-time signal x[n] is
and h[n] = 0 for n < 0 or n > 2 obtained by sampling the continuous-time
What are the values of the filter taps a and b if signal x(t) = cos(200 t) at sampling instants
the output y[n] = 0 for all n, where x[n] is as t = n/400, n = 0, 1, ...., 7. The 8-point Discrete
given above? Fourier Transform (DFT) of x[n] is defined as,

7 nkn
n=0 j ,
x[n] y[n] = 0 X[ k ] = x [n] e 4 k = 0, 1, ..., 7
h[n] = {1, a, b} n=0

(a) a = 0, b = –1 (b) a = 1, b = 1 Which one of the following statement is true?

(c) a = –1, b = 1 (d) a = 0, b = 1 (a) Only X[2] and X[6] are non-zero.
[EC-2019 : 2 Marks] (b) Only X[3] and X[5] are non-zero.
(c) All X[k] are non-zero.
Q.30 Consider a six-point decimation-in-time Fast
(d) Only X[4] is non-zero.
Fourier Transform (FFT) algorithm, for which
[EC-2020 : 2 Marks]
the signal-flow graph corresponding of X[1] is
Q.32 Consider two 16 point sequences x[n] and h[n].
j2
shown the figure. Let W6 = exp . In the Let the linear convolution of x[n] and h[n] be
6
denoted by y[n], while z[n] denotes the 16 point
figure, what should be the values of the Inverse Discrete Fourier Transform (IDFT) of the
coefficients a1, a2, a3 in terms of W6 so that X[1] is product of the 16 point DFTs of x[n] and h[n].
obtained correctly? The value(s) of ‘k’ for which z[k] = y[k] is/are
(a) k = 15 (b) k = 0, 1, 2, ...., 15
x[0] X[0] (c) k = 0 (d) k = 0 and k = 15
[EC-2021 : 1 Mark]
a1
x[3] X[1]
–1 Q.33 Consider the signals, x[n] = 2n – 1 u[–n + 2] and
y[n] = 2–n + 2 u[n + 1), where u[n] is the unit step
x[1] a2 X[2]
sequence. Let X(ej ) and Y(ej ) be the discrete
time Fourier transform of x[n] and y[n],
x[4] a3 X[3] respectively. The value of the integral,
–1
2
1
x[2] X[4] X( e j )Y ( e j
)d
2
0

(Rounded off to one decimal place) is ______ .

x[5] X[5]
–1 [EC-2021 : 2 Marks]
 GATE Previous Years Solved Paper 115

Q.34 For a vector x = [ x[0], x[1],..., x[7]], the 8-port If, x = [1, 0, 0, 0, 2, 0, 0, 0]
discrete Fourier transform (DFT) is denoted by and y = DFT(DFT ( x ))
X = DFT ( x ) = [ X[0], X[1],..., X[7]] then the value of y[0] is ______ (Rounded off to
one decimal place).
7
2
where, X[ k ] = x[n] exp j nk [EC-2022]
n=0 8

Here, j= 1

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC DTFS, DTFT and DFT

1. (d) 2. (d) 3. (c) 4. (a) 5. (c) 6. (b) 7. (b) 8. (a)

9. (a) 10. (d) 11. (d) 12. (b) 13. (a) 14. (10) 15. (c) 16. (4)

17. (3.375) 18. (b) 19. (c) 20. (11) 21. (2) 22. (0.5) 23. (8) 24. (6)

25. (4096) 26. (2.10) 27. (c) 28. (3) 29. (d) 30. (d) 31. (a) 32. (a)

33. (8) 34. (8)

Solutions
EC DTFS, DTFT and DFT

1. (d) x(n) (n – a) x(n – a)

h(n) = 4 2 (n + 2) 2 2 (n + 1) y(n) = 4 2 [ e j /4( n+ 2) + e j /4( n 2) ]

/4( n + 1)
2 2 (n 1) + 4 2 (n 2) 2 2 [e j + ej /4( n 2)
]

x(n) = ej n/4 jn /4
= e [4 2 ( e j /2 + e j /2 )
y(n) = x(n) h(n)
= 4 2 (n + 2) e jn /4 2 2 (e j /4
+e j /4
)]

+ 4 2 (n 2) e jn /4
= e
jn /4
[0 2 2 × 2 cos /4]

2 2 [ ( n + 1) e jn /4 y(n) = –4e jn /4

+ (n 1) e jn /4
]
116 Electronics Engineering Signals & Systems

2. (d) j 1 2j 1
= e e + ej + 2 + e j
+ e 2j
n 2 2
1
x(n) = u(n)
2 j e2 j + e 2 j
= e + ej + e j + 2
2n 2
1
y(n) = x 2 (n) = u2 ( n )
2 f(n) = e–j [cos2 + 2cos + 2]
2 n
1 6. (b)
= u(n)
2 • sin( on + ) = x(n) X(ej )
n • y(n) = Ax(n – no)
1
y(n) = u(n)
4 Y(ej ) = Ae jno X ( e j )

1 Y(e j ) jno
Y(z) = • H(ej ) = = Ae
1 1 X( e j
)
1 z
4
• H(ej ) = –no o+2 k (at = o)
Put, z = e j
1 7. (b)
Y(e j ) =
1 j X(ej ) = e3j + e2j + 0 + 5 + e–j
1 e
4
1 4 Q e aj d =0 if a 0
Y(ejo) = =
1 3
1
4
e3 j e2 j e j
X( e j ) d = + +5 +
4. (a) 3j 2j j

n = 5 + 5 = 10
y(n) = x 1 , n even
2
9. (a)
= 0 for n odd
h(2) = 1
n = 0, y(n) = x(–1) = 1
h(3) = –1
n = 2, y(n) = x(0) = 2
h(k) = 0 otherwise
n = 4, y(n) = x(1) = 1
n = 6, y(n) = x(2) = 1/2 h(k)

5. (c) 1

y(2n) = x(n – 1) seeing in graph k

2
f(n) = y(2n) –1

1
= (n + 1) + (n) + 2 (n 1) It is finite impulse response. It is not low-pass
2
filter.
1
+ (n 2) + (n 3)
2 10. (d)
Taking z-transform, 4-point DFT of sequence [1, 0, 2, 3] is given as,
1 1
F(z) = z + 1 + 2z 1 + z 2 + z 3 1 1 1 1 1 1+2 +3 6
2 2
1 j 1 j 0 1 2 + 3j 1+ 3j
z = ej = =
1 1 1 1 2 1+2 3 0
1 j j 2j 1 3j 1 j 1 j 3 1 2 3j 1 3j
F(e j ) = e + 1 + 2e +e + e
2 2
 GATE Previous Years Solved Paper 117

11. (d) 15. (c)

For a N-point FFT algorithm with N = 2 m, Minimum phase all zeros are inside the unit
computation of a butterfly requires only one circle.
complex multiplication and two complex • For maximum phase, all the zeroes are
additions. outside the unit circle.

12. (b) 7 1 3 2
• H ( z) = 1 + z + z =0
x[n] real 2 2
X[k] conjugate symmetric 2z2 + 6z + z + 3 = 0
X[k] X [N – k] 1
z1 = (inside the unit circle)
N = 8 2
X[k] = X [8 – k] z2 = –3 (outside the unit circle)
X[6] = X [8 – 6] = X [2] = 0 Mixed phase
X[7] = X [8 – 7] = X [1] = 1 + j3
16. Sol.
13. (a) Given, x[n] = 0.5n u[n]
DFT [p q r s] = [a b c d] [a b c d] Z-transform of x(n) is
So, DFT [p q r s] = [ 2 2 2 2]
1
X(z) =
14. Sol. (1 0.5 z 1 )
From discrete Fourier series definition, and Z-transform of y(n) = x(n) x(n) is
N 1 2
jk
N
n 1
x[n] = ak e Y(z) =
k=0 (1 0.5 z 1 )2

2 2
j n j n
= a 1e N + a0 + a1 e N ...(i) y( n) = Y ( z ) z = 1 = 4
n=
1 j n 1 j 5n
x[n] = e 5 + e
2j 2j 17. Sol.
Also given, n
2 Ae j 6 f
u(n + 3)
n 3 2
x[n] = sin 1 e j2 f
5 3
2 n
where, N = 10 = N = 10 2
N 5 Let, x[n] = u(n + 3)
3
By comparison,
Discrete Fourier transform is given as,
1 1
a1 = and a9 = a 1 =
2j 2j j n
X(e j ) = x[n] e
As the TFS coefficient ak is also periodic with n=

period N = 10, n
2 j n
a 1 = a11 = a21 .... = e
n= 3 3
a –1 = a9 = a19 ....
Given that, ak is non-zero for k = Bm ± 1. Put, n+3 = m
To satisfy this condition B must be ‘10’ where
‘m’ is any integer.
118 Electronics Engineering Signals & Systems

m 3 2
2 j ( m 3) j
So, X(e j ) = e where, W31 = e 3
m=0 3
4
j
3 m
2 2 j m W32 = e 3
= e ej 3
3 m=0 3 2
j
3 m
W3 1 = e 3
3 2 j
= e ej 3
j
4
2 3 2 3
m=0 W3 = e
j2 f 3 j6 f 2
3.375 e 3.375 e j
= = W3 4 = W3 1 = e 3
2 j2 f 2 j2 f
1 e 1 e
3 3
p c
So, A = 3.375
q = a
18. (b) r b
DFT (DFT (x)) = x
Let, x = [1 2 3 2] 20. Sol.

1 1 1 1 1 x1[n] = {1, 2, 3, 0}
x2[n] = {1, 3, 2, 1}
1 1 j 1 j 2
DFT (x) = X(k)3 = X(k)1 X(k)2
4 1 1 1 1 3
x3[n] = circular convolution of
1 j 1 j 4
x1[n] and x2[n]
8 4 = x1[n] x2[n]
1 2 1
= = x3 (0) 1 0 3 2 1 9
2 0 0
x3 (1) 2 1 0 3 3 8
2 1 = =
x3 (2) 3 2 1 0 2 11
1 1 1 1 4 x3 (3) 0 3 2 1 1 14
1 1 j 1 j 1
DFT (DFT (x)) = x3[2] = 11
4 1 1 1 1 0
1 j 1 j 1 21. Sol.

2 1 n n
1 1
= n u(n)
1 4 2 2 n= 2
= = n=0
2 6 3
4 2 n
1
= Z T n u(n)
2
z=1
19. (c)

A a+b+c
1 1 z
z
B = a + bW3 1 + cW3 2 2 2
= = 2
1 1
C a + bW3 2 + cW3 1 1 z z
1
2 2 z=1
a+b+c
W32 W31 3
p 1
a + bW3 1 + cW3 2 1/2
q = 1 W31 W32 W32 W31 = =2
3 1/4
r 1 W32 W32 W31 W31
a + bW3 2 + cW3 1
3
 GATE Previous Years Solved Paper 119

22. Sol. 24. Sol.

n From the given question,
x [n] = 1 + cos
8 n
x1[n] = x
n n 3
j j
e 8 e 8
= 1+ + (N = 16) X1[k] = {12, 2j, 0, –2j, 12, 2j, 0, –2j}
2 2 X1[8] = 12 ; X1(11) = –2j
2 2
j n j n X1 (8) 12
e 16 e 16 = =6
= 1+ + X1 (11) 2j
2 2
1 1 25. Sol.
a1 = and also a31 = a15 = a 1 =
2 2
f s = 8000 samples/sec
23. Sol. 1 1
Ts = = sec.
From the definition of DTFT, f s 8000

j n
Time for each multiplication = Tm = 20 µsec
X(ej ) = x[n] e Tm = 20 × 10–6 sec
n=
Suppose block is size N, then time taken to
1
x[n] = X( e j ) e j n
d 1
2 generate each block = N × Ts = N × .
8000
1 Number of multiplication that can be performed
x[0] = X( e j ) d by processor in the time taken for each block =
2
N × Ts
1 j j
.
X( e ) Y(e )d = [x(n) y(n)]n = 0 Tm
2
Number of multiplication required to compute
Y(ej ) = sin2 (2 )
N
1 cos 4 DFT by Radix – 2 FFT algorithm = log 2 N .
= 2
2
Number of multiplication required by FFT
1 1 4j 1 4j
= e e Number of multiplication that can be performed
2 4 4
by the processor in the time taken in each block.
1 1 1
y[n] = [ n] [n + 4] [n 4] [Reason : Real time]
2 2 4
N N × Ts
log 2 N
1 1 1 2 Tm
y[n] = , 0, 0, 0, , 0, 0, 0,
4 2 4 1
2×
log2 N 8000 = 12.5
1 20 × 10 6
y[0] =
2 N 212.5 N 212 = 4096
x[n] = {6, 3, 8, 7, 4} ; x[0] = 8
26. Sol.
1 1 1 1
X( e j ) Y ( e j ) d = 2 x[0] y[0] Since, h[n] = [ n] + [n 1] + [n 2]
3 3 3
n=
1 j
1 So, H(e j ) = e [1 + 2 cos ]
= 2 ×8× = 8 3
2
H(e j o) = 0 ; for (1 + 2 cos o) = 0
120 Electronics Engineering Signals & Systems

1 To get y(n) = 0,
or, cos o =
2 = H(e j )
H(e j o) =0
2 = /2
or, o = = 2.10 radians
3 j j2
1 + ae + be 2 = 0
2
27. (c) 1 – ja – b = 0
h[n] = 5 [n] – 7 [n – 1] + 7 [n – 3] – 5 [n – 4] From the given options, a = 0 and b = 1.
Now, H(e j ) = 5 – 7e–j + 7e–3j – 5e–4j
Now for = 0, 30. (d)
H(e jo) = 5 – 7 + 7 – 5 = 0 N 1 2
j kn
and for = , X(k) = x(n) e N
H(e j ) = 5 – 7(–1) + 7(–1) – 5(1) n=0
= 5+7–7–5=0 5
System is attenuating low and high frequencies X(1) = x(n) W6n
whereas passing the mid frequencies. So, its a n=0

BPF. 2 3
= x(0) + x(1) W6 + x(2) W6 + x(3) W6
28. Sol. + x(4) W64 + x(5) W65 ...(i)
X(k) = {1, 2, 3, 4, 5, 6, 7, 8} From the given flow graph,
7 2 X(k) = [x(0) – x(3)] a1 + [x(1) – x(4)] a2
j kn
X(k) = x[n] e 8
+ [x(2) – x(5)] a3
n=0
By comparing equations (i) and (ii), we get,
7 j kn 2
a 1 = 1, a2 = W6 , a3 = W6
= x[n] e 4
n=0
31. (a)
3
x[2n] = x[n] x(t) = cos200 t
n=0 n = 0, 2, 4, 6 n
t=
400
1 7
= ( x[n] + ( 1)n x[n]) n
2 n=0 x(n) = cos 200 = cos n
400 2
7
n = 0, 1, ... , 7
x[n] = X(0) = 1
n=0 3
= cos 0, cos , cos , cos ;
2 2
7 7 j 4n
n x[n] e 4 = X(4) = 5 5 7
( 1) x[n] = cos 2 , cos , cos 3 , cos
n=0 n=0 2 2
3 DFT
1 = {1, 0, –1, 0, 1, 0, –1, 0} X( k )
x[2n] = [1 + 5] = 3
n=0
2 Suppose,
DFT
29. (d) y(n) = {1, –1, 1, –1} Y( k )
[Y(k)] = [ WN ] N = 4 [ y(n)]
j n j n
x(n) = c1 e 2 + c2 e 2
1 1 1 1 1
rad/sec.
= 1 j 1 j 1
o 2 = = [0, 0, 4, 0]
1 1 1 1 1
H(e j ) = 1 + ae–j + be–j2
1 j 1 j 1
 GATE Previous Years Solved Paper 121

Now, as we know, if for 1 4 3 2 0

DFT 2 1 4 3 1
{a, b, c, d} { A, B, C , D}
=
3 2 1 4 2
DFT
Then for {a, 0, b, 0, c, 0, d, 0} {A, B, C, D, 4 3 2 1 3
A, B, C, D} z(n) = {16, 18, 16, 10 }
Similarly, for z(3) = 10
DFT Here, y(n) and z(n) are equal for n = 3
y(n) = {1, –1, 1, –1} Y(k) = {0, 0, 4, 0}
[n = N – 1 = 4 – 1]
Here, for x(n) = {1, 0, –1, 0, 1, 0, –1, 0}
where, ‘N’ is length of the sequences.
X(k) = {0, 0, 4, 0, 0, 0, 4, 0}
33. (8)
Given:
32. (a)
F.T.
Given, y(n) = x(n) h(n) x(n) = 2 n 1
u( n + 2) X( e j )
where, ‘ ’ represents circular convolution n+2 F.T.
y(n) = 2 u( n + 1) Y(e j )
Given, z(n) = IDFT [X(k) H(k)]
For any real signal,
z(n) = x(n) N h(n) X(e j ) = X (e–j )
X (e j ) = X(e–j )
where, N represents circular convolution 2
1
Value of X( e j )Y ( e j
)d
The length of linear convolution is N1 + N2 – 1. 2
0
Given, 16 point sequences 2
1
So, N1 + N2 – 1 = 31 = X( e j )Y ( e j ) d
y(n) length is 31. 2
0
The length of circular convolution is max(N1, N2) From Plancherel’s theorem
= max(16, 16) = 16 +
So, z(n) length is 16. = x(n) y (n)
n=
z(k) and y(k) are equal for k = 15[16 – 1]
i.e., z(15) = y(15) Overlap of the signals x(n) and y(n) in the range
For your understanding purpose, consider two –1 to +2,
4-point sequences as: x(n) = 2n – 1; n 2
x(n) = {1, 2, 3, 4} y(n) = 2–n + 2; n –1
h(n) = {0, 1, 2, 3} 2 2
n+2
= 2n 1
2 = 2
y(n) = x(n) h(n) n= 1 n= 1
0 1 2 3 = 2[1 + 1 + 1 + 1] = 8
1 0 1 2 3
34. (8)
2 0 2 4 6
= y(n) = DFT [DFT (x(n))]
3 0 3 6 9
Using duality property,
4 0 4 8 12
DFT DFT
x( n ) Nx( k ) = y( n)
y(n) = {0, 1, 4, 10 , 16, 17, 12}
y(n) = Nx(–k) = Nx(N – k)
y(3) = 10 y(n) = 8(1, 0, 0, 0, 2, 0, 0, 0)
z(n) = x(n) N h(n) At x = 0, y(0) = 8 × 1 = 8
 8 Sampling

ELECTRO NICS EN GINEERIN G (b) 800 Hz and 900 Hz components.

(c) 800 Hz and 1000 Hz components.
(GATE Previous Years Solved Papers)
(d) 800 Hz, 900 Hz and 1000 Hz components.
Q.1 A signal containing only two frequency [EC-1995 : 1 Mark]
components (3 kHz and 6 kHz) is sampled at
the rate of 8 kHz, and then passed through a Q.4 Flat top sampling of low pass signals
low pass filter with a cut-off freuqency of 8 kHz. (a) gives rise to aperture effect
The filter output (b) implies oversampling
(a) is an undistorted version of the original (c) leads to aliasing
signal. (d) introduces delay distortion
(b) contains only the 3 kHz component. [EC-1998 : 1 Mark]
(c) contains the 3 kHz component and a
Q.5 Let x(t) = 2 cos(800 t) + cos(1400 t). x(t) is
spurious component of 2 kHz.
sampled with the rectangular pulse train shown
(d) contains both the components of the original
in the figure. The only spectral components
signal and two spurious components of
(in kHz) present in the sampled signal in the
2 kHz and 5 kHz.
frequency range 2.5 kHz to 3.5 kHz are
[EC-1988 : 2 Marks]
p (t )
Q.2 Increased pulse-width in the flat-top sampling,
leads to 3

m(t ) M(f )

t
–T0 –T0/6 0 T0/6 T0
–3
f T = 10 sec
–W 0 W
(a) 2.7, 3.4
(a) attenuation of high frequencies in
reproduction. (b) 3.3, 3.6
(b) attenuation of low frequencies in (c) 2.6, 2.7, 3.3, 3.4, 3.6
reproduction. (d) 2.7, 3.3
(c) greater aliasing errors in reproduction. [EC-2003 : 2 Marks]
(d) no harmful effects in reproduction.
Q.6 A 1 kHz sinusoidal signal is ideally sampled at
[EC-1994 : 1 Mark] 1500 samples/sec and the sampled signal is
Q.3 A 1.0 kHz signal is flat-top sampled at the rate passed through an ideal low-pass filter with
of 1800 samples/sec and the samples are cut-off frequency 800 Hz. The output signal has
applied to an ideal rectangular LPF with cut-off the frequency
frequency of 1100 Hz, then the output of the (a) 0 Hz (b) 0.75 Hz
filter contains (c) 0.5 kHz (d) 0.25 kHz
(a) only 800 Hz component. [EC-2004 : 2 Marks]
 GATE Previous Years Solved Paper 123

Q.7 A signal m(t) with bandwidth 500 Hz is first (a) 5 Hz and 15 Hz only
multiplied by a signal g(t) where, (b) 10 Hz and 15 Hz only
(c) 5 Hz, 10 Hz and 15 Hz only
g(t ) = ( 1)k (t 0.5 × 10 4
k) (d) 5 Hz only
k=
[EC-2014 : 1 Mark]
The resulting signal is then passed through an
ideal low pass filter with bandwidth 1 kHz. The Q.12 For a given sample-and-hold circuit, if the value
output of the low pass filter would be of the hold capacitor is increased, then
(a) (t) (b) m(t) (a) drop rate decreases and acquisition time
decreases.
(c) 0 (d) m(t) (t)
(b) drop rate decreases and acquisition time
[EC-2006 : 2 Marks]
increases.
Q.8 An LTI system having transfer function (c) drop rate increases and acquisition time
decreases.
s2 + 1
and input x(t) = sin(t + 1) is in steady- (d) drop rate increases and acquisition time
s2 + 2s + 1
increases.
state. The output is sampled at a rate s rad/sec [EC-2014 : 1 Mark]
to obtain the final output {y(k)}. Which of the
following is true?
Q.13 The signal cos 10 t + is ideally sampled at
(a) y is zero for all sampling frequencies s. 4
(b) y is non-zero for all sampling frequencies
a sampling frequency of 15 Hz. The sampled
s. signal is passed through a filter with impulse
(c) y is non-zero for s > 2 but zero for s < 2.
(d) y is non-zero for > 2 but non-zero for sin( t )
s response cos 40 t . The filter
t 2
s < 2.
[EC-2009 : 2 Marks] output is
Q.9 A band-limited signal with a maximum 15
(a) cos 40 t
frequency of 5 kHz is to be sampled. According 2 4
to the sampling theorem, the sampling frequency
15 sin( t )
which is not valid is (b) cos 10 t +
2 t 4
(a) 5 kHz (b) 12 kHz
15
(c) 15 kHz (d) 20 kHz (c) cos 10 t
2 4
[EC-2013 : 1 Mark]
15 sin( t )
Q.10 Consider two real valued signals, x(t) band- (d) cos 10 t
2 t 2
limited to [–500 Hz, 50 Hz] and y(t) band-limited
[EC-2015 : 1 Mark]
to [–1 kHz, 1 kHz]. For z(t) = x(t) y(t), the Nyquist
sampling frequency (in kHz) is ______ . Q.14 Consider a continuous time signal defined as,
[EC-2014 : 1 Mark]
sin t /2
x(t ) = (t 10 n)
Q.11 Let x(t) = cos(10 t) + cos(30 t) be sampled at t /2 n=
20 Hz and reconstructed using an ideal low-
where ‘ ’ denotes the convolution operation
pass filter with cut-off frequency of 20 Hz. The
and ‘t’ is in seconds. The Nyquist sampling rate
frequency/frequencies present in the
(in samples/sec) for x(t) is _______ .
reconstructed signal is/are
[EC-2015 : 2 Marks]
 124 Electronics Engineering Signals & Systems

Q.15 A continuous-time sinusoid of frequency 33 Hz ELECTRICAL EN GINEERIN G

is multiplied with a periodic Dirac impulse
(GATE Previous Years Solved Papers)
train of frequency 46 Hz. The resulting signal is
passed through an ideal analog low-pass filter Q.1 The frequency spectrum of a signal is shown in
with a cut-off frequency of 23 Hz. The the figure. If this is ideally sampled at intervals
fundamental frequency (in Hz) of the output is of 1 ms, then the frequency spectrum of the
________ . sampled signal will be
[EC-2016 : 1 Mark] |U(j )|
Q.16 Consider the signal x(t) = cos(6 t) + sin(5 t),
where ‘t’ is in seconds. The Nyquist sampling
rate (in samples/second) for the signal
y(t) = x(2t + 5) is 1 kHz
(a) 8 (b) 12
(c) 16 (d) 32 |U(j )|

[EC-2016 : 1 Mark]

Q.17 The signal x(t) = sin(14000 t), where ‘t’ is in (a)

seconds is sampled at a rate of 9000 samples
per second. The sampled signal is the input to
an ideal low-pass filter with frequency response
H(f ) as follows:
1, f 12 kHz (b)
H( f ) =
0, f > 12 kHz
What is the number of sinusoids in the output
and their frequencies (in kHz)?
(a) Number = 1, frequency = 7
(b) Number = 3, frequencies = 2, 7, 11
(c)
(c) Number = 2, frequencies = 2, 7
(d) Number = 2, frequencies = 7, 11
[EC-2017 : 2 Marks]

Q.18 A band-limited low-pass signal x(t) of

bandwidth 5 kHz is sampled at a sampling rate
(d)
fs. The signal x(t) is reconstructed using the
reconstruction filter H(f ) whose magnitude
response is shown below.
[EE-2007 : 1 Mark]
H( f )
K Q.2 A band-limited singal with a maximum
frequency of 5 kHz is to be sampled. According
to the sampling theorem, the sampling frequency
in kHz which is not valid is
f (kHz)
–8 –6 0 6 8 (a) 5 (b) 12
The minimum sampling rate fs (in kHz) for (c) 15 (d) 20
perfect reconstruction of x(t) is ______ . [EE-2013 : 1 Mark]
[EC-2018 : 2 Marks]
 GATE Previous Years Solved Paper 125

Q.3 A sinusoidal x(t) of unknown frequency is (a) 2B1 (b) 2(B1 + B2)
sampled by an impulse train of period 20 ms. (c) 4(B1 + B2) (d)
The resulting sample train is next applied to an [EE-2016 : 2 Marks]
ideal low-pass filter with a cut-off at 25 Hz. The
filter output is seen to be a sinusoid of frequency Q.6 The output y(t) of the following system to be
20 Hz. This means that x(t) has a frequency of sampled, so as to reconstruct it from its samples
uniquely. The required minimum sampling rate
(a) 10 Hz (b) 60 Hz
is
(c) 30 Hz (d) 90 Hz
X( )
[EE-2014 : 2 Marks]

Q.4 For the signal f(t) = 3 sin8 t + 6 sin12 t + sin14 t,

the minimum sampling frequency (in Hz)
–1000 1000
satisfying the Nyquist criterion is _____ . sin (1500 t ) y(t)
h (t ) =
x (t ) X( ) t
[EE-2014 : 1 Mark]

Q.5 Let x1(t) X1( ) and x2(t) X2( ) be two cos(100 t)

signals whose Fourier transforms are shown in (a) 1000 samples/sec
the figure below. In the figure, h(t ) = e 2 t (b) 1500 samples/sec
(c) 2000 samples/sec
denotes the impulse response.
(d) 3000 samples/sec
X1( ) X 2( )
[EE-2017 : 2 Marks]

Q.7 Consider the two continuous-time signals

defined below:
These signals are sampled with a sampling
–B1 B1 B1 B1 –B2 B2
period of T = 0.25 seconds to obtain discrete-
2 2
time signals x1[n] and x2[n], respectively. Which
X1(t ) one of the following statements is true?
–2|t| (a) The energy of x1[n] is greater than the energy
h (t ) = e y (t )
of x2[n].
X2(t ) (b) The energy of x2[n] is greater than the energy
For the system shown above, the minimum of x1[n].
sampling rate required to sample y(t), so that (c) x1[n] and x2[n] have equal energies.
y(t) can be uniquely reconstructed from its (d) Neither x1[n] nor x2[n] is finite energy signal.
samples, is [EE-2018 : 2 Marks]
126 Electronics Engineering Signals & Systems

Electronics & Electrical Engineering

GATE Previous Years Solved Paper

A n swe rs & Expl a n a t i o n s

Answers
EC Sampling

1. (d) 2. (a) 3. (c) 4. (a) 5. (d) 6. (c) 7. (b) 8. (a)

9. (a) 10. (3) 11. (a) 12. (b) 13. (a) 14. (0.4) 15. (13) 16. (c)

17. (b) 18. (13)

Solutions
EC Sampling

1. (d) 3. (c)
f s = 8000 samples/sec f s = 1800 samples/sec
fm1 = 3 kHz fm = 1000 Hz
fm2 = 6 kHz The spectrum of sampled signal would have
The spectrum of sampled signal would have nfs ± fm
nfs ± fm So, 1000 Hz, 1800 ± 1000 Hz, 3600 ± 1000 Hz ...
So, 3 kHz, 8 ± 3, 16 ± 3, ... = 3 kHz, 5 kHz, So, 1000 Hz, 800 Hz, 2800 Hz, 2600 Hz,
11 kHz, ... 4600 Hz, ...
6 kHz, 8 ± 3, 16 ± 6, ... = 6 kHz, 2 kHz, The cut-off frequency of LPF is 1100 Hz.
14 kHz, ... So, the output of filter will contains 800 Hz and
Cut-off frequencies of LPF = 8 kHz 1000 Hz components.
So, the filter output would have 3 kHz, 6 kHz,
4. (a)
2 kHz and 5 kHz.
Flat top sampling of low pass signals gives rise
2. (a) to aperture effect.

x(t ) X( f ) = sin c ( f ) 5. (d)

From Fourier series expansion,
1 F.T.
To /6
1 jn ot
t f Cn = Ae dt
To
To /6
As pulse width is increased, the width 1/ of
the first lobe of the spectrum is decreased. A n
= sin
Hence, increased pulse-width in the flat-top n 3
sampling, leads to attenuation of high From Cn its clear that 1, 2, 4, 5, 7,... harmonics
frequencies in reproduction. are present.
 GATE Previous Years Solved Paper 127

Frequency of p(t) corresponding to 1, 2, 4, 5, 7, ... x(t) = sin(t + 1)

are 103, × 2 × 103, 4 × 103.... = 12 rad/sec
x(t) has frequency components 0.7 k and 0.4 k 1 es
p(t) × x(t) gives (1 ± 0.7) k, (2 ± 0.7) k, (4 ± 0.7) k, X(s) = 2
es = = 2
+ s2 12 + s 2 s +1
(1 ± 0.4) k, (2 ± 0.4) k, ....
Y (s )
Frequency present in range of 2.5 k to 3.5 are H(s) =
X( s )
2.7, 3.3.
Y(s) = H(s) X(s)
6. (c)
s2 + 1 es
f s = 1.5 kHz Y(s) = ×
s2 + 2s + 1 s2 + 1
fm = 1 kHz
es
Available frequency components are nfs ± fm. =
So, 1 kHz, 2.5 kHz, 0.5 kHz, .... s2 + 2s + 1
Q LPF has fc = 0.8 k es
Y(s) =
only 0.5 k will appear at output. (s + 1)2
y(t) = (t + 1) e–(t + 1)
7. (b)
ses
M(f ) y( ) = lim sY (s) = lim =0
s 0 s 0 ( s + 1)2

So, at steady-state ‘y’ remains zero for all

sampling frequencies s.
f
–500 500
9. (a)
m(t) g(t) M(f ) G(f)
(fs)min = 2fm
1 (fs)min = 2 × 5 = 10 kHz
4 = 20 kHZ
0.5 × 10 So, f s 10 kHz
G(f ) 10. Sol.
Multiplication in time domain = convolution in
frequency domain,
x1(t) x2(t) X1( ) X2( )
f
–20 kHz 20 kHz So, highest frequency component contained by
the convolved signal z(t) = 1500 Hz.
M(f ) G(f ) Nyquist rate = 2 × 1500
= 3000 Hz = 3 kHz

11. (a)
f
x(t) = cos(10 t) + cos(30 t)
–20 kHz 20 kHz
Given sampling frequency,
After low pass filtering with fc = 1 kHz f s = 20 Hz
Output is zero.
s = 40 rad/sec.

8. (a) X( )

s2 + 1
X(s) H (s ) = X(s )
2
s + 2s + 1
–30 –10 10 30
128 Electronics Engineering Signals & Systems

1 (f 20) (f + 20) j
Y( ) = X( k o) H(f ) = rect + rect e 2
2 2 2
k=
After sampling waveform will be Output, Y(f ) = Xs(f ) × H(f )
15
Y( ) y(t) = cos 40 t
2 4

14. Sol.
–70 –50 –30 –10 10 30 50 70 t
sin
–40 40 2
x(t) = [t 10 n]
t n=
Applying an ideal low-pass filter of cut-off
2
frequency of 20 Hz or 40 rad/sec, we get the
frequencies in reconstructed signal as, 10 and
30 rad/sec. or 5 Hz and 15 Hz.
X(f ) = ×
12. (b) f
–0.1 0 0.1 0.2 0.3
–1/4 1/4
We know that, fmax = 0.2 Hz
Q = CV = i t ...(i) f s = 2fmax = 0.4 Hz
CV
t= ...(ii) 15. Sol.
i
If x(t) is a message signal and y(t) is a sampled
From equation (ii),
signal, then y(t) is related to x(t) as,
t C and it is clear that if value of capacitor
increases then the acquisition time increases.
y(t) = x(t ) (t nTs )
For capacitor the drop rate is given as dv/dt, n=
dv
i= C
dt Y(f ) = f s X( f nTs )
dv i n=
=
dt C Spectrum of X(f ) and Y(f ) are as shown
dv 1 X(f )
dt C
From above relation it is clear that if capacitor
value increases drop rate decreases.
f
13. (a) –33 Hz 0 33 Hz

x(t) = cos 10 t +
4 Y(f )

1
[ f 5) + ( f + 5)] e j
X(f ) =
/4
2
After sampling x(t) by 15 Hz, we get –79 Hz –33 Hz –13 Hz 13 Hz 33 Hz 79 Hz
f

Xs(f ) = 15 X( f 15n) –23 Hz 23 Hz

n=
Cut-off frequency of LPF = 23 Hz
sin( t ) Hence, frequency at the output is 13 Hz.
Given, h(t) = cos 40 t
t 2
 GATE Previous Years Solved Paper 129

16. (c) So, three sinusoids will be there at the output of

X(t) = cos(6 t) + sin(8 t) the LPF and the frequencies of those sinusoids
Y(t) = x(2t + 5) are 2 kHz, 7 kHz and 11 kHz.
Y(t) = cos[6 (2t + 5) + sin(8 (2t + 5)]
18. Sol.
= cos(12 t + 30 ) + sin(16 t + 40 )
Let us assume an arbitrary spectrum for x(t) as
fm1 = 6 HZ, fm2 = 8 Hz
shown below.
Nyquist sampling rate,
f s = 2 fmax X(f )

= 16 samples/second

17. (b)
x(t) = sin(14000 t) f (kHz)
–5 0 5
fm = 7 kHz
f s = 9000 samples per second The spectrum of the sampled signal can be given
= 9 kHz as,
The spectrum of the sampled signal can be given
as shown below.
S(f )
Frequency response of
LPF with fc = 12 kHz f (kHz)
1 –fs – 5 –fs –fs + 5 –8 –6 –5 0 5 6 8 fs – 5 –fs fs + 5

For proper reconstruction of the signal,

fs – 5 8
–25 –16 –11 –7 –2 0 2 7 11 16 25 f(kHz) f s 8 + 5 = 13 kHz
–12 12
So, fs(min) = 13 kHz

Answers
EE Sampling

1. (b) 2. (a) 3. (c) 4. (14) 5. (b) 6. (b) 7. (a)

Solutions
EE Sampling

1. (b) Therefore sampling frequency,

1
|U(j )| fs =
Ts
1
= 1 kHz
=
10 3
After sampling new signal in frequency domain,
–1 kHz 1 kHz
Given that, sampling interval = 1 msec 1
UT(f ) = U( f nf s )
i.e., Ts = 1 msec = 10–3 sec Ts n=
130 Electronics Engineering Signals & Systems

Spectrum of sampled signal will be Then the minimum sampling frequency

satisfying the Nyquist criterion is = 14 Hz.
|UT(j )|
Method-II:
Sampling frequency = 2 (Highest frequency
component of f(t)).
Given, f(t) = 3 sin8 t + 6 sin12 t + sin14 t
Different frequency components are:
2. (a) 1=8 :

(fs)min = 2 fm 8
f1 = = 4 Hz
(fs)min = 2 × 5 = 10 kHz 2
So, f s 10 kHz 2 = 12 :
12
3. (c) f2 = = 6 Hz
2
Given, impulse train of period 20 ms. and = 14 :
2
1 14
Then, sampling frequency = 3
= 50 Hz f3 = = 7 Hz
20 × 10 2
If the input signal x(t) = cos m (t) having Clearly, highest frequency component of the
spectrum. signal,
X(f ) f(t) = f3 = 7 Hz
Minimum sampling frequency,
f s = 2 × f3
= 2 × 7 = 14 Hz
f
–fm fm 5. (b)
The filtered out sinusoidal signal has 20 Hz Given that:
frequency has sampling must bounded Bandwidth of X1( ) = B1
sampling. The output signal which is an under Bandwidth of X2( ) = B2
sampled signal with sampling frequency
System has h(t ) = e 2 t and input to the system
50 Hz is
is x1(t) x2(t).
X(f )
The bandwidth of x1(t) x2(t) is B1 + B2.
The bandwidth of output will be B1 + B2.
So sampling rate will be 2(B1 + B2).

f
6. (b)
–fm –50 + fm 50 + fm fm
sin(1500 t ) y(t)
x(t) h (t ) =
and 50 – fm = 20 Hz t
X( )
fm = 30 Hz
1 cos(1000 t)
4. Sol.
Method-I: –1000 1000

fm1 = 4 Hz From the above block diagram,

fm 2 = 6 Hz Z(t) = x(t) cos1000 t
fm 3 = 7 Hz By using modulation property of Fourier
transform,
 GATE Previous Years Solved Paper 131

1
Z( ) = [ X( + 1000 ) + X( 1000 )]
2 1
Z( ) 0.75
0.5
1/2 0.25
t

0
–2000 0 2000

sin 1500 t –0.25 0.25

Now, h(t) = = 1500 Sa (1500 t ) –0.5 0.5
t
–0.75 0.75
H( )
–1 1
1

1 t; 1 t 1
Now, x2(t) =
–1500 0 1500 0; otherwise
Thus, H( ) is a low pass filter and it will pass x2(t)
frequency, component of Z( ) upto 1500 rad/s. 1
Y( )

t
–1 1
(rad/sec)
x2(n) = {0, 0.25, 0.5, 0.75, 1, 0.75, 0.5, 0.25, 0}
–1500 1500
Therefore, maximum frequency component of
1
y(t) is
0.75
m = 1500 rad/sec 0.5
or, fm = 750 Hz 0.25
So, the minimum sampling rate for y(t) is t
fs min = 2fm = 2 × 750
0
= 1500 Hz
= 1500 samples/sec. –0.25 0.25
–0.5 0.5
7. (a)
–0.75 0.75
t; 1 t 1
x1(t) = –1 1
0; otherwise
Since x1(n) is having one more non-zero sample
x1(t)
of amplitude ‘1’ as compared to x2(n). Therefore,
1 energy of x1(n) is greater than energy of x2(n).

t
–1 1
Ts = sampling time-period
= 0.25 sec.
x1(n) = {1, 0.75, 0.5, 0.25, 0, 0.25, 0.5, 0.75, 1}
 9 Digital Filters

ELECTRO NICS EN GINEERIN G The filter can be used to approximate a

(GATE Previous Years Solved Papers) (a) low-pass filter (b) high-pass filter
(c) band-pass filter (d) band-stop filter
Q.1 A continuous-time filter with transfer function [EC-2016 : 2 Marks]
2s + 6 Q.3 Let h[n] be a length 7 discrete-time finite impulse
H (s) = 2
is converted to a discrete time
s + 6s + 8 response filter, given by:
filter with transfer function, h[0] = 4, h[1] = 3, h[2] = 2, h[3] = 1
h[–1] = –3, h[–2] = –2, h[–3] = –1
2 z 2 0.5032 z
G( s ) = 2 . So that the impulse
z 0.5032 z + k and h[n] is zero for n 4. A length-3 finite
impulse response approximation g[n] of h[n] has
response of the continuous-time filter, sampled
to be obtained such that,
at 2 Hz, is identical at the sampling instants to
the impulse response of the discrete time filter. 2
E( h , g ) = H ( e j ) G( e j ) d
The value of ‘k’ is ______ .
[EC-2016 : 2 Marks]
is minimized, where H(e j ) and G(e j ) are the
Q.2 The direct form structure of an FIR (Finite discrete time Fourier transforms of h[n] and g[n],
Impulse Response) filter is shown in the figure. respectively. For the filter that minimizes E(h, g),
the value of 10g[–1] + g[1] rounded off to two
Unit Unit
x[n]
delay delay decimal places, is _______ .
[EC-2019 : 2 Marks]

5 5

–
+
y[n]
 GATE Previous Years Solved Paper 133

Answers
EC Digital Filters

1. (0.049) 2. (c) 3. (–27)

Solutions
EC Digital Filters

1. Sol.
= 0; H(e j ) = 0
2s + 6 1 1
Given, H(s) = 2
= +
s + 6s + 8 s+2 s+4 = ; H ( e j ) = 10
2
h(t) = e–2t u(t) + e–4t u(t)
Given, f s = 2 Hz = , H(e j ) = 0
For discrete time,
The given one is band-pass filter.
n
t = nTs =
2 3. Sol.
–n –2n
h[n] = (e + e ) u[n] From Parseval’s theorem,
1 1
H(z) = + 2 1 2
1 e 1
Z 1
1 e 2Z 1 x[n] = X( e j ) d
n=
2
Z Z
= 1
+ 2 2 3
2
Z e Z e So, H ( e j ) G( e j ) d = 2 h( n ) g( n )
n= 3
2Z 2 0.5032 Z
= The solution of g(n) that minimizes E(h, g) also
Z2 0.5032 Z + 0.049
3
K = 0.049 2
minimizes, h(n) g(n) .
n= 3
2. (c)
y[n] = 5x[n] – x[n – 2]] 3
2
h(n) g(n) = 4 g(0) 2 + 3 g(1) 2
Y ( z) 2
H(z) = = 5[1 z ] n= 3
X( z)
2
But, z = ej + 3 g( 1) + 10
H(e ) = 5[1 – e–j2 ]
j
The solution of g(n) that minimizes the above
equation is,
H(e j )
g(n) = { 3, 4, 3}
10
So, 10g(–1) + g(1) = 10(–3) + 3
= –27

0 /2