3 Signals and Systems: Part II

Solutions to
Recommended Problems
S3.1
(a)
x[n]= 8[n] + 8 [n - 3]

n
0

Figure S3.1-1

(b)
x[n] = u[n]-u[n-- 5]

1111

0041T
-1

Figure S3.1-2

(c)
x [n]=S[n] + -1 [n --

-3 -2 -1

+(_1)2S[n] +(_1)3 5[n -3]

n
2

Figure S3.1-3

(d)
x(t)= u(t +3) - u(t - 3)

-3

Figure S3.1-4

S3-1

Signals and Systems

S3-2

(e)
x(t) =6(t + 2)

-2

0
Figure S3.1-5

(f)

S3.2
(1) h
(2) d
(3) b
(4) e
(5) a, f
(6) None

S3.3
(a)

x[n] = b[n -

1] -

26[n -

2] + 36[n -

(b) s[n] = -u[n + 3] + 4u[n + 1] -

S3.4
We are given Figure S3.4-1.

3] -

26[n 4] + b[n -

4u[n 2] + u[n -

4]

5]

Signals and Systems: Part II / Solutions

S3-3

x(- t) and x(1 - t) are as shown in Figures S3.4-2 and S3.4-3.

x (-t)

-12
Figure S3.4-2

x(1--t)
x1-0

-11

1
Figure S3.4-3

(a) u(t + 1) - u(t - 2) is as shown in Figure S3.4-4.

-1

Figure S3.4-4
Hence, x(1 -

t)[u(t + 1) -

u(t - 2)]1 looks as in Figure S3.4-5.

5
6
t

-1

1
Figure S3.4-5

Signals and Systems

S3-4

(b) -u(2 - 3t) looks as in Figure S3.4-6.

2
3
t

Figure S3.4-6
Hence, u(t + 1) -

u(2 -

3t) is given as in Figure S3.4-7.

Figure S3.4-7
So x(1 t)[u(t + 1) - u(2 - 3t)] is given as in Figure S3.4-8.

S3.5
(a) y[n] = x2[n] + x[n] - x[n - 1]
(b) y[n] = x2[n] + x[n] - x[n - 1]
(c) y[n] = H[x[n] x[n - 1]]
x 2[n] + x 2 [n - 1] - 2x[nJx[n 1]
(d) y[n] = G[x 2[n]]
x 2[n] - X 2 [n - 1

Signals and Systems: Part 11/ Solutions

S3-5

(e) y[n] = F[x[n] - x[n - 1]]

= 2(x[n] - x[n - 1]) + (x[n - 1] y[n] = 2x[n] - x[n - 1] - x[n - 2]

x[n -

2])

(f) y[n] = G[2x[n] + x[n - 11]

11 - 2x[n - 1] - x[n 1] - x[n - 2]

= 2x[n] + x[n = 2x[n] - x[n -

21

(a) and (b) are equivalent. (e) and (f) are equivalent.
S3.6
Memoryless:
(a) y(t) = (2 + sin t)x(t) is memoryless because y(t) depends only on x(t) and not
on prior values of x(t).
(d) y[n] = Ek=. x[n] is not memoryless because y[n] does depend on values of
x[-] before the time instant n.
(f) y[n] = max{x[n], x[n - 1],

...

, x[-oo]} is clearly not memoryless.

Linear:
(a)

y(t) = (2 + sin t)x(t) = T[x(t),

T[ax 1 (t) + bx 2 (t)] = (2 + sin t)[axi(t) + bx 2 (tt)

= a(2 + sin t)x 1(t) + b(2 + sin t)x 2 (t)
= aT[x 1 (t)] + bT[x 2 (t)]
Therefore, y(t) = (2 + sin t)x(t) is linear.
(b)

y(t) = x(2t) = T[x(t)],

T[ax 1(t) + bx 2 (t)] = ax1 (2t) + bx 2 (t)

= aT[x1 (t)) + bT[x 2 (t)]

Therefore, y(t) = x(2t) is linear.

y[n] =

(c)

x[k] = T[x[n]],

k=

T[ax1 [n] + bx2[n]] = a T

x1 [k] + b L

k= -x

k=-w

x 2[k]

= aT[x1[n]] + bT[x2[n]]

Therefore, yin] = E=_ x[k] is linear.

(d) y[n] =

>

x[k] is linear (see part c).

k= -o

dxt
d

T[ax 1(t) + bx 2(t)] = -[ax 1 (t)+bx

2 (t)]
= a

dx 1 (t)
dx (t)
dt + b dt2

aT[x 1 (t)] + bT[x 2 (t)]

Therefore, y(t) = dx(t)/dt is linear.

y[n] = max{x[n], . . . , x[-oo]} = T1x[n]],
(f)
T[ax 1[n] + bx2[n]] = max{ax 1 [n] + bx 2[n], . . . , ax 1[-oo] + bx2[ - o]}
# a max{x[n], . . . , x1[-oo]} + b max{x 2[n], . . . , X2[-00])
Therefore, y[n] = max{x[n],

.. .

, x[-oo]} is not linear.

Signals and Systems

S3-6

Time-invariant:
(a)

y(t) = (2 + sin t)x(t) = T[x(t)],

T[x(t - T 0 )] = (2 + sin t)x(t - T0 )
9 y(t - T0 ) = (2 + sin (t - T0 ))x(t - T0 )

Therefore, y(t) = (2 + sin t)x(t) is not time-invariant.

(b)

y(t) = x(2t) = T[x(t)],

# x(2t - TO) = y(t - T 0 )

Therefore, y(t) = x(2t) is not time-invariant.
T[x(t -

T0 )] = x(2t -

2T0)

y[n] =

x[kJ = T[x[n]],

T[x[n - NO]J =

x[k -

(c)

NO] = y[n - N 0 ]

Therefore, y[n] = Ek'= _.x[k] is time-invariant.

(d)

y[n] =

x[k] = T[x[n]],

k=
n

T[x[n - NO]] = E

n-NO

x[k - N] =

k=-

x[l] = y[n - N
0 ]
=-w0

Therefore, y[n] = E" _.x[k] is time-invariant.

(e)

y(t)
T[x(t -

=dx(t)

To)] =

dt

x(t
dt

T[x(t)],

To) = y(t - To)

Therefore, y(t) = dx(t)/dt is time-invariant.

Causal:
(b) y(t) = x(2t),

y(1) = x(2)

The value of y(-) at time = 1 depends on x(-) at a future time = 2. Therefore,

y(t) = x(2t) is not causal.
(d) y[n] =

x[k]

k=

Yes, y[n] = E .x[k] is causal because the value of y[-] at any instant n
depends only on the previous (past) values of x[-].
Invertible:
(b) y(t) = x(2t) is invertible; x(t) = y(t/2).

(c) y[n] = E _.x[k] is not invertible. Summation is not generally an invertible

operation.
(e) y(t) = dx(t)/dt is invertible to within a constant.
Stable:

(a) If Ix(t) I < M, Iy(t) I < (2 + sin t)M. Therefore, y(t) = (2 + sin t)x(t) is stable.
(b) If |x(t)| < M, |x(2t)I < M and ly(t)| < M. Therefore, y(t) = x(2t) is stable.
(d) If |x[k]| 5 M,

ly[n]j 5 M - E_,, which is unbounded. Therefore, y[n]

E"Lx[k] is not stable.

Signals and Systems: Part II / Solutions

S3-7

S3.7
(a) Since H is an integrator, H-1 must be a differentiator.
dx(t)
d
dt

H~':

y(t) =

G:
G 1:

y(t) = x(2t)
y(t) = x(t/2)

(b)

Solutions to
Optional Problems
S3.8
(a) x 2(t) = xi(t) - xi(t

2)

2(t) =

1 (t)

y)(t - 2)

Figure S3.8-1

Signals and Systems

S3-8

(b)

xA(t) = Xi( t)

+ xI(t + 1)

y 3(t)=y (t)+y 1 (t + 1)

10

-1

Figure S3.8-2
(c) x(t) = u(t -

y(t)=e-(t-1)u
-1)+u(-t)+
u(t
-2)
-u(1
-t)

1) - u(t -

2)

Figure S3.8-3
(d) y[n] = 3y 1 [n] 2y2[n] + 2y 3 [n]

p p14

-3

-2

-1

-4
Figure S3.8-4

Signals and Systems: Part 11/ Solutions

S3-9

(e) y2[n] = y,[n] + yi[n -

y2

11

[n]

__

Figure S3.8-5
Y3[n] = y 1 [n

+ 1]

2 3 4

-1 0

Figure S3.8-6

(f) From linearity,

y1(t) =

1r

+ 6 cos(2t) - 47 cos(5t) + '/e cos(6t),

4"

0t

(-t

x 2(t) = 1 + t2 =

)".

n=O

So y 2(t)

1 - cos(2t) + cos(4t) - cos(6t) + cos(8t).

S3.9
(a) (i)

The system is linear because

Tlaxi(t)

+ bx 2 (t)]

[ax

(t) + bx (t )](t - nT)

1
2
n=

a T3 x 1 (t)b(t - nT) + b

xst

nT)

= aT[xi(t)] + bT[x 2 (t)]

(ii)

The system is not time-invariant. For example, let xi(t) = sin(22rt/T).

The corresponding output yi(t) = 0. Now let us shift the input xi(t) by
r/2 to get
x2

(t) = sin (

+r =

cos (2)

Now the output

+00

Y2(0

>7 b(t - nT) =Ay,

n = -oo

+ 2=

Signals and Systems

S3-10

(b) y(t)

x(t)b(t - nT)

cos(2,rt)b(t - nT)

cos(27rt)

0<

Figure S3.9-1

T=1
0

tt

~11
2

Figure S3.9-2

Signals and Systems: Part II / Solutions

S3-11

y (t )

T=

T~__

-1
Figure S3.9-5

y (t)

Figure S3.9-6

(c) y(t) =

e t cos(27rt)b(t - nT)

etcos(2nrt)

et
t

22

Figure S3.9-7

y (t)

T= I

T=1

tt
1
Figure S3.9-8

3
2

Signals and Systems

S3-12

y (t)
T= Ie3
2

-e1/2

2-

3/2

Figure S3.9-9

y (t)
YWe3

T=
4t

r2

2
-e 3 12

-el2

Figure S3.9-10

T=

Y3
12

y~t)

Figure S3.9-12

S3.10

(a) True. To see that the system is linear, write

y2 (t) = T 2 [T1 [x(t)]]

T[x(t)],

T 1[ax 1 (t) + bx 2(t)] = aT 1 [x1(t)] + bT[x 2(t)]

T2[T1[ax 1(t) + bx 2 (t)]] = T2[aT1 [x1(t)] + bT1[x 2(t)]I
=

aT2[T1 [x1 (t)]] + bT2[(T[x2t)]]

aT[x 1 (t)] + bT[x 2(t)J

Signals and Systems: Part 11/ Solutions

S3-13

We see that the system is time-invariant from

T 2[T 1[x(t - T)]] = T 2[y 1(t - T)l
= y 2(t

-T),

Tx(t - T)] = y 2(t

T)

(b) False. Two nonlinear systems in cascade can be linear, as shown in Figure S3.10.
The overall system is identity, which is a linear system.

x(t)

Reciprocal

Reciprocal

0 y(t)=x(t)

x(t)

Figure S3.10
(c) y[n] = z[2n] = w[2n] + {w[2n - 1] + {w[2n - 21
= x[n] + {x[n - 11

The system is linear and time-invariant.

(d) y[n] = z[-nl = aw[-n 11 + bw[-n] + cw[-n + 1]
= ax[n + 11 + bx[nl + cx[n - 1]

(i)
(ii)
(iii)

The overall system is linear and time-invariant for any choice of a, b,

and c.
a= c
a= 0

S3.11
(a) y[n] = x[n] + x[n - 11 = T[x[n]]. The system is linear because
T[ax 1 [n] + bx 2 [n]| = ax1[n] + ax1[n - 1] + bx 2[n] + bx 2[n - 1]
=

aT[x1 [n]] + bT[x 2[n - 1]]

The system is time-invariant because

y[n] = x[n] + x[n - 1] = Tjx[n]],
T[x[n - N]] = x[n - N] + x[n 1 - N]
= y[n - N]

(b) The system is linear, shown by similar steps to those in part (a). It is not
time-invariant because
T[x[n N]]

x[n - N] + x[n - N - 1] + x[O]

# y[n - N] = x[n N] + x[n - N - 1] + x[-NJ

S3.12
(a) To show that causality implies the statement, suppose
x1 (t) - yl(t)

(input x 1(t) results in output y 1(t)),

x 2(t) - y2),

where y 1(t) and y 2(t) depend on x 1(t) and

(t) for t < to. By linearity,

xI(t) - x 2(t) -+ y 1 (t) - y 2(t)

Signals and Systems

S3-14

If Xi(t)

x 2 (t) for t < to, then y 1 (t) = y 2 (t) for t < to. Hence, if x(t) = 0 for

t < to, y(t) = 0 for t < to.

(b) y(t) = x(t)x(t + 1),

x(t) = 0
for t < to =* y(t) = 0,
for t < to
This is a nonlinear, noncausal system.
(c) y(t) = x(t) + 1 is a nonlinear, causal system.
(d) We want to show the equivalence of the following two statements:
Statement 1 (S1): The system is invertible.
Statement 2 (S2): The only input that produces the output y[n] = 0 for all n is
x[n] = 0 for all n.

To show the equivalence, we will show that

S2 false
S1 false
S2 false
S1 false
X1

(e) y[n]

==

S1 false
S2 false

S1 false: Let x[n] # 0 produce y[n] = 0. Then cx[n] == y[n] = 0.

S2 false: Let xi => yi and x 2 =* Y2. If x 1 # X 2 but y1 = Y2, then

X2 0 0 but yi - yi = 0.
=

and

x 2 [n] is nonlinear and not invertible.

MIT OpenCourseWare
http://ocw.mit.edu

Resource: Signals and Systems

Professor Alan V. Oppenheim

The following may not correspond to a particular course on MIT OpenCourseWare, but has been
provided by the author as an individual learning resource.

For information about citing these materials or our Terms of Use, visit: http://ocw.mit.edu/terms.