,7

Signals and Linear Systems X(f)

Chapter 2

Problems

105

-l

Equations (2.7. - 7) and (2.7.11) represent two methods for xpressing a bandthe signal in terms of pass signal in terms of two lowpass signals. We can express components or in terms of the envelope and phase of the in-phase and uadrature

the bandpass signal.

-fW-f,
+w

f' - Wf.f, + w

2.8 FURTHER READING

X(f) Numerous references cover the analysis of LTI systems in both the time and freWillsky, and Young (1983) contains wide coverage of

J
X(f)

time- and frequency-domain analysis of both discrete-time and contin of the Fouriertems. Papoulis (1962) and Bracewell (1 provide in-Ueptn a lysis systems based series and transform techniques. A more advanced treatment 01 uneai studied on linear operator theory can be found in Franks (1969). Digital filters are (1996). The book by ingle and Proakis (1997) in detail by Proakis and Manolakis

emphasizes design techniques for digital filters using MATLAB.

-fc
.51 Spectra of three bandpass signa's.

.1

PROBLEMS

quation (2.7.11) is a very useful relation; it basically says that a bandpass an be represented in terms of two lowpass signals, namely, its in-phase and
ure components. i this case, the complex lowpass signal xj(t) = xC(t) + jxs(t) d the lowpass equivalent of the bandpass si oordinates, we have

2.1 Plot the f lowing signals:

1. x1(t) = Fl(2t + 5) 2. x2(t) = fl(-2t + 8)

g
n

a a

(2.7.12)

3. x3(t) = E 0A (t - '/)

1 x(t). If we represent x1(t) in

c

4. X4(t) = 2 Ei A ; E n=-c'c A(t , 4n)

--

x(t) r t n Xs(t) x1(t) = 7x (t) +x (t)ej we define the envelope and the phase of the bandpass signal as

5. x5(t) = sgn(2t) - sgn(t) 6. x6(t) E 'I)IzA(t 7. X7(t) = sinc( lot) 8. x8(t) = sine

2n)

lxi (t)l = A(,) = 7x (t) + x (t)

-

/x1(t) = 9(t) = arctan

(2.7.13)

9. x9(t) = 41J

cos(4rt)

2.2 Plot the discrete version of the given signals. Assume they are sampled at
multiples of T0, i.e., x[n] = x (n T0)

t express xi(I) as
xj(t) = A(t)e18(t).

(2.7.14)

1. x(t) = sinc(3t) and T0 =

2. x(t) = I-I

Equations (2.7.14) and (2.7.11), we have

x (t) = Re[xj (t)eI22Tit]
= Re[A (t)e' 2rft+9 (t)

and T0 =

(2.7. ] 5) (2.7.16) (2.7.17)

3. : = tu,i(t) ' (t 1)u, (t 1) and T0 =

= A(t) cos(27rft + 9(t)).

= 0 l , 2, 2.3 Two signals, xl (t) = 1 and X2(t) = cot and x2[n]. Verify that The resulting discrete-time signals are denoted by x1 from this observation? xl = X2 I What can you conclude

Signals and Linear Systems

2.4 Show that the sum of two discrete periodic signals is periodic, whereas the Sum of two Continuous periodic condition is the snm nf tx,,-, p^t signals is not necessarily periodic. Under what riodic signals periodic?
2.5 Determine whether the

Chapter 2 -

Problems

107

mine the period

given signals are periodic. For periodic signals, deter e

2. x2(t) = E A(t ' n) 3. xi(t) = sin t + sin 27rt

4. X4[fl] = sin n

5. xs(t) = E - y(t ; nT), where y(t) is an arbitrary signal and T is an

arbitrary constant

1. sin(4OOO t) + cos(l I .000Jrt)

2. sin(4OOOj-t) + cos( " , 000t) 3. sin[4OOO ,1j + cos[l I ,OOOrii]

-

2.9 Using the definition of power-type and energy-type signals,

1. Show that x(t) = Ae3(2J0t+9) is a power-type signal and its power
content is A I e e

4. sin[4OQOj,1j + cos[1 l,000nj

2.6 Classify the sign als that follow are neither even nor odd. In the !nto even signals, odd signals, and signals that latter case, find the even and odd parts of the signals.
t >
0 0 0

2. Show that x(t) = A cos(27rfht + is power-type and its power is 3. Show that the unit-step signal U_I (t) is a power-type signal and find its power content.

1 1 1

4. Show that the signal

-

I ,f,, 3
!' *'l`'' , -,

<

X
L

Kt

t>0 t<0

cos (l20t +
1 1

t0

is neither an energy- nor a power-type signal. 2.10 Find the even and odd parts of the signal x(t) = A(t)u_1 (t). 2.11 Using the definition of even and odd signals, 1. Show that the decomposition of a signal into even and odd parts is que. odd signals is even, whereas 2. Show that the product of two even or al odd. the product of an even and an odd sign 3. Try to find an example of two signals that are neither even nor odd, but
have an even product.

X2 (t) = e
I '

4. x3(t)

0
1

t=0

5. x(t)
6 . 7

l `

'

I>0
I<0

()

X5(t) = sin t + cos t . X6 (I) = x] (t) . x2(t), where x1 (t) is even and x2(t) is odd

2.7 Classify these signals into energy-type signals, power-type signals, and that are neither energy type nor power type signals. For energy-type and signals powertype signals, find the energy or the power content of the signal.
1. x1(t) == (e I cost) u 1 (t) 2. x2(t) = e,t cost 3. x3(t) = sgn(t) 4. x4(t) = A cos 2rJj t + B cos 2rtt

2.12 Plot the f lowing signals:

1. x1(t) = LI (t) + fT(-t)

--

2. x2(t) = LI (t) , U (t - I) 3. x3(t) = A(t)IJ(t)

4. x4(t) = E _ A(t 2n) 5. x5(t) = E -l A(t ,

6. x(t) = sgn (t) + sgn(l - t) 7. x7(t) = 1 + sgn(t)
8. x8(t) = sgn2 (t)

.8 Classify these signals into periodic and nonperiodic:

1. x1(t) = 2 E A L) , E A(t . 4n)

9. x(t) = sinc(t) sgn(t)

 Ub

Signals and Linear Systems

10. 11.

Chapter 2

Problems

109

x10(t) = E (-1 )'n8 (t , n)

xl(t) = :00 n=1

fl()

3. y(t) 4. y(t)

0
e s
r

2 s (
X X

2.13 By using the properties of the impulse function, find the values of these

expressions:

5. y(t) a

t) > 0
t) <0

`n l )
x (1)

1. xl(t) = sinc(f) 8(t)

2. x1(t) = sinc(t) 8(t , 3)

6. y(t)=-

( 0(t)l

x(t) 0
x (t) = 0

3. x1(t) = sinc(t - 2)8(t) 4. x(t) = A(I)* E 8(t . 2n) 5. x3(t) = A (t) 8'(t) 6. x4(t) = cost 8(3t) 7. x4(t) = cos (2t + 8(3t) 8. x4(t) = cost 8(3t + 1)

7. y(t)

(t)

lx(1) 0

x(t) 0
otherwise

8. y(t) = e-'x(t)
9. y(t) = x(t)u1 (t)

10. y(t) = x(t)6(t)

11. y(t) = x (t) E 5(t T)

r0 l
1

,) = 8(5t) * 8(4i')
.

L2 L 3.

= 8(5t) * 8'(3t) x7(t) = cos I 8'(t)

x6(t) X8 (t)

x (t)
12. y(t) =
-

t>0 t<0
t = 0

r (t)
0

f - L

fI(t)8(2t - 1) dt
sinc(t) 8(t) (It

13. y(t) = x (t) + y (t - 1)

14. y(t) = Algebraic sum of jumps in x(t) in the interval (-oc, t]

1 5 1" 6)
.

si nc (t + 1)8(t) (It

2.17 Prove that a system is linear if and only if

1. It is nomogeneous, 1.l for all input signals x(t) and all real numbers i.e., 1. It 15horn

LEoc, -fI()j8(t)dt Z_n=J 2

cos t [E00 I 8(2't)] dt fl =

2.14 Show that the impulse signal is even. What can you say about evenness or oddness of its th derivatjve?

we have ''[ax(t)] = a .T [x (I)]. 2. It is additive, i.e., for all input signals x1(t) and X2(1), we have

2.15 We have seen that x(t) * 8(t)

TIixi(t) + x2(t)j = .T [Xi (t)1 + I?T [ x2(t)I .

In other words, show that the two definitions of linear systems given by Equations (2.1.39) and (2.1.40) are equivalent.

= x(t). Show that

d" dt'

x (t) * (t) = j-X (t) and

2.18 Verify whether any (or both) of the conditions described in Problem 2.17 are satisfied by the systems given in Problem 2.16.

x (t) (t) = 1'

dr.

2.19 Prove that if a system satisfies the additivity property described in Problem

2.17. then it is homogeneous for all rational 2.20 Show that the system described by
l -

2.16 Classify these systems into linear and nonlinear:

1. y(t) = 2x(t) , 3

x
'0 ( )

x'(t) 0

2. y(t) ==

r(t) = 0

is homogeneous but nonlinear. Can you give another example of such a system? t

110 , .

'

Signals and Linear Systems

Chapter 2

Pro ems

111

2.21 Show that the response of a linear system to the input which is identically zero is an output which is identically zero.
2.22 The system defined by the input-output relation y (t) = x(t) cos(2jrf0t),

2.28 Let a system be defined by

y (t)

_ 2T J' t- I X(T)

r+T

at.

Is this system causal? 2.29 For an LTT system to be causal, it is required that /](t) be zero for t < 0. Give an example of a nonlinear system which is causal, but its impulse response is

where is a constant, is called a modulator. Is this system linear? Is it time invariant?

2.23 Are these statements true or false? Why? 1. A system whose components are nonlinear is necessarily nonlinear. 2. A system whose components are time variant is necessarily a time-variant system. 3. The response of a causal system to a causal signal is itself causal. 2.24 Determine whether the following systems are time variant or time invariant: 1. y(t) = 2x(t) + 3 2. y(t) = (t + 2)x(t) 3. y(t) = x (t) + t 4. y(t) =
-1)

nonzero for t <

2.30 Determine whether the impulse response of these LTI systems is causal:

1. h(t) = sinc (t)

2. h(t) = 3. h(t) = sine (t)u_i (t) 2.31 Using the convolution integral, show that the response of an LTI system to
u_i(t) is given by j h(r)dt.

2.32 What is the impulse response of a differentiator? Find the output of this system to an arbitrary input x (t) by finding the convolution of the input and the impulse response. Repeat for the delay system. 2.33 The system defined by
y (t)= [ x(r)dT.

5. y(t) = x(t)u-1 (t) 6. y(t) = x(t)8(t)

7. y(t) = x (t) E '(l . nT)
Jt- 7

8. y(t) = f oox(r)dr 9. y(t) = x(t) + y (t - 1)

x(t)

(T is a constant) is a finite-time integrator. Is this system LTI? Find the impulse response of this system. 2.34 Compute the following convolution integrals: 1. e_tu_i(t) 2. e-tu_i(t)

x(t) 0
x(t) = 0

etul(t) u_1(t)

10. y(t):I o

2.25 Prove that if the response f an LTI system to x(t) is y(t), then the response d of this system to x(t) is y (t).

3. rI(t) * A (1)

2.26 Prove that if the response of an LTI system to x(t) is y(t), then the response of this system to J ,0x(r)dr is fyd 2.27 The response of a linear time-invariant system to the input x(t) = e'atu[j (t)
is (t). Using time-domain analysis and the result of Problem 2.25, determine the impulse response of this system. What is the response of the system to a general input x(t)?

4. (A (t)sgn(t)) *u1(t) 5. A(t) * sgn (t)

6. (A(t)u-1(t)) * H (t)

2.35 Show that in a causal LTI system, the convolution integral reduces to
r+oc ,. ,, , , , r
y tI) = - t)/1 a = / )Ctj)fltt - t) U t