Chapter 10: Time Domain Analysis of DT Systems

Problem 10.1
(a) Expressing the difference equation as
y[k ] = 2 y[k 1] + x[k 1]
= 2 y[k 1] + 2u ( k 1) { x[k ] = 2u[k ]}
2 y[k 1] k <1
=
2 y[k 1] + 2 k 1

Starting from k = 0, we iterate to calculate the output as

y[0] = 2 y[ 1] = 4
y[1] = 2 y[0] + 2 = 10
y[2] = 2 y[1] + 2 = 22
y[3] = 2 y[2] + 2 = 46
y[4] = 2 y[3] + 2 = 94
y[5] = 2 y[4] + 2 = 190
(b) and (c) The zero-state and zero-input responses are calculated below

Zero state response Zero input response

2 y zs [k 1] k <1
, y[ 1] = 0
y zi [ k ] = 2 y zi [ k 1] with y[ 1] = 2
yzs [k ] =
2 yzs [k 1] + 2 k 1

Iterating with respect to k, we get Iterating with respect to k, we get

y zs [0] = 2 y zs [ 1] = 0 y zi [0] = 2 y[ 1] = 4
y zs [1] = 2 y zs [0] + 2 = 2 y zi [1] = 2 y zi [0] = 8
y zs [2] = 2 y zs [1] + 2 = 6 y zi [2] = 2 y zi [1] = 16
y zs [3] = 2 y zs [2] + 2 = 14 y zi [3] = 2 y zi [2] = 32
y zs [4] = 2 y zs [3] + 2 = 30 y zi [4] = 2 y zi [3] = 64
y zs [5] = 2 y zs [4] + 2 = 62 y zi [5] = 2 y zi [4] = 128
(d) It can be easily verified that y[k] = yzi[k] + yzs[k].

Problem 10.2
(a) Expressing the difference equation as
y[k ] = y[k 1] 0.5 y[k 2] + x[k 2]

y[ k 1] 0.5 y[k 2], k<2

or, y[k ] = k 2
y[k 1] 0.5 y[k 2] + 0.5 k 2.
2 Chapter 10

Starting from k = 0, we iterate to calculate the output as

y[0] = y[1] 0.5 y[ 2] = 0 0.5(1) = 0.5
y[1] = y[0] 0.5 y[ 1] = 0.5 0.5(0) = 0.5
y[2] = y[1] 0.5 y[0] + 1 = 0.5 0.5(0.5) + 1 = 0.75
y[3] = y[2] 0.5 y[1] + 0.5 = 0.75 0.5(0.5) + 0.5 = 1.5
y[4] = y[3] 0.5 y[2] + 0.5 2 = 1.5 0.5(0.75) + 0.25 = 1.375
y[5] = y[4] 0.5 y[3] + 0.5 3 = 1.375 0.5(1.5) + 0.125 = 0.75.
(b) and (c) The zero-state and zero-input responses are calculated below

Zero state response Zero input response

y[ k 1] 0.5 y[k 2], k<2 y[k ] = y[k 1] 0.5 y[k 2]
y[k ] = k 2
y[k 1] 0.5 y[k 2] + 0.5 k 2.

with y[1] = y[2] = 0. with y[1] = 0 and y[2] = 1.

Iterating with respect to k, we get Iterating with respect to k, we get
y zs [0] = 0 y zi [0] = y zi [1] 0.5 y zi [ 2] = 0.5
y zs [1] = 0 y zs [1] = y zi [0] 0.5 y zi [ 1] = 0.5
y zs [2] = y zs [1] 0.5 y zs [0] + 1 = 1 y zs [2] = y zs [1] 0.5 y zs [0] = 0.25
y zs [3] = y zs [2] 0.5 y zs [1] + 0.5 = 1.5 y zs [3] = y zs [2] 0.5 y zs [1] = 0
2
y zs [4] = y zs [3] 0.5 y zs [2] + 0.5 = 1.25 y zs [4] = y zs [3] 0.5 y zs [2] = 0.125
3
y zs [5] = y zs [4] 0.5 y zs [3] + 0.5 = 0.625. y zs [5] = y zs [4] 0.5 y zs [3] = 0.125.

(d) By adding the zero-state and zero input responses obtained in parts (b) and (c) and comparing with
the output in (a), it is verified that y[k] = yzi[k] + yzs[k].

Problem 10.3
(a) Expressing the difference equation as
y[k ] = 0.75 y[k 1] 0.125 y[k 2] + x[k 2]

0.75 y[k 1] 0.125 y[k 2], k<2

or, y[k ] = k 2
0.75 y[k 1] 0.125 y[k 2] + (1) k 2.
Starting from k = 0, we iterate to calculate the output as
y[0] = 0.75 y[ 1] 0.125 y[ 2] = 0.75 0.125(1) = 0.875
y[1] = 0.75 y[0] 0.125 y[ 1] = 0.75(0.875) 0.125(1) = 0.53125
y[2] = 0.75 y[1] 0.125 y[0] + 1 = 0.75(0.53125) 0.125(0.875) + 1 = 1.28906
y[3] = 0.75 y[ 2] 0.125 y[1] 1 = 0.75(1.28906) 0.125(0.53125) 1 = 0.09961
y[4] = 0.75 y[3] 0.125 y[2] + 1 = 0.75(0.09961) 0.125(1.28906) + 1 = 0.76416
y[5] = 0.75 y[ 4] 0.125 y[3] 1 = 0.75(0.76416) 0.125(0.09961) 1 = 0.41443.
(b) and (c) The zero-state and zero-input responses are calculated below
 Solutions 3

Zero state response Zero input response

0.75 y zs [k 1] 0.125 y zs [k 2], k<2 y zi [k ] = 0.75 y zi [k 1] 0.125 y zi [k 2]
y zs [k ] = k 2
0.75 y zs [k 1] 0.125 y zs [k 2] + (1) k2
with yzs[1] = yzs [2] = 0. with yzi[1] = 1 and yzi [2] = 1.
Iterating with respect to k, we get Iterating with respect to k, we get
y zs [0] = 0 y zi [0] = 0.75 y zi [1] 0.125 y zi [2] = 0.875
y zs [1] = 0 y zs [1] = 0.75 y zi [0] 0.125 y zi [1] = 0.53125
y zs [ 2] = 0.75 y zs [1] 0.125 y zs [0] + 1 = 1 y zs [2] = 0.75 y zs [1] 0.125 y zs [0] = 0.28906
y zs [3] = 0.75 y zs [2] 0.125 y zs [1] 1 = 0.25 y zs [3] = 0.75 y zs [2] 0.125 y zs [1] = 0.15039
y zs [ 4] = 0.75 y zs [3] 0.125 y zs [ 2] + 1 = 0.6875 y zs [4] = 0.75 y zs [3] 0.125 y zs [2] = 0.07869
y zs [5] = 0.75 y zs [4] 0.125 y zs [3] 1 = 0.45312. y zs [5] = 0.75 y zs [4] 0.125 y zs [3] = 0.03988.

(d) By adding the zero-state and zero input responses obtained in parts (b) and (c) and comparing with
the output in (a), it is verified that y[k] = yzi[k] + yzs[k].

Problem 10.4

a k u[k ] b k u[k ] = a
m =
m
u[m] b k mu[k m] = a m b k mu[k m] = b k (a / b) m u[k m]
m=0 m=0

Noting that u[k m] = u[(m k)] = 0 for m > k, we obtain

k (k + 1)b k a = b
a u[k ] b u[k ] = b (a b) = k 1( a b )k +1
k k k m

m=0 b 1( a b ) a b
(k + 1)a a = b (k + 1)a k a = b
k

= k +1 1( a b )k +1 = k +1 k +1
b ba a b b b aa ab

(k + 1)a a=b
k
= a k +1 bk +1
a b ab

Problem 10.5

(a) The graphical approach is illustrated in Fig. 10.5(a).

Case (k < 6), Fig. S10.5(a) subplot (v): x1[k ] x2 [k ] = 0 .
Case (6 k 2), Fig. S10.5(b) subplot (vi):
k +4
x1[k ] x 2 [k ] = 1 = ( k + 7 ) .
m = 2

Case (2 k 6), Fig. S10.5(b) subplot (vii):

4 Chapter 10

2
x1 [k ] x 2 [ k ] = 1 = (6 k ) .
m=k 4

Case (k > 6), Fig. S10.5(b) subplot (viii): x1[k ] x2 [k ] = 0

The convolution sum is plotted as a function of k in subplot (ix) of Fig. S10.5(a).

x1[ m] = u[m + 2] u[ m 3] x2 [ m] = u[ m + 4] u[ m 5]

m m
5 4 3 2 1 0 1 2 3 4 5 5 4 3 2 1 0 1 2 3 4 5

(i) Sequence x1[m] (ii) Sequence x2[m]

x2 [ m] = u[m + 4] u[m 5] x2 [ k m] = x2 [( m k )]

m
m

k1

k+4
k4
k3
k2

k
k+1
k+2
k+3
k5

k+5
5 4 3 2 1 0 1 2 3 4 5

(iii) Sequence x2[m] (iv) Sequence x2[km]

x1[ m] x2 [ k m]
x1[m] x2 [k m]

m
m
k
k+1
k+2

k+4
k+3
2 1 0 1 2 3 4 5
k+3
k+4

2 1 0 1 2 3 4 5

(v) Convolution x1[m] * x2[m] for k < 6. (vi) Convolution x1[m] * x2[m] for (6 k 2)
x1[ m] x2 [k m] x1[ m] x2 [k m]

m m
k4
k3
k2

k
k+1
k1

k2
k4
k3

2 1 0 1 2 3 4 5 2 1 0 1 2 3 4 5

(viii) Convolution x1[m] * x2[m] for (2 k 6) (viii) Convolution x1[m] * x2[m] for k > 6.
x1[k ] x2 [ k ]
5
4 4
3 3
2 2
1 1
k
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7

(ix) Convolution x1[m] * x2[m] as a function of k.

Fig. S10.5(a) Convolution sum for part (a) in Problem 10.5.

(b) From Problem 10.4, we know that

a k u[k ] b k u[k ] = 1
( a b ) (a k +1 b k +1 )u[k ] .

Setting a = 0.8 and b = 0.5, we get

0.8k u[k ] 0.5k u[k ] = 103 (0.8k +1 0.5k +1 )u[k ] .
 Solutions 5

Using the time shifting property,

0.8k 5 u[k 5] 0.5k u[k ] = 103 (0.8k 5+1 0.5k 5+1 )u[k 5] ,

or, 0.8k 5 u[k 5] 0.5k u[k ] = 103 (0.8k 4 0.5k 4 )u[k 5] ,

which implies 0.8k u[k 5] 0.5k u[k ] = 0.85 103 (0.8k 4 0.5k 4 )u[k 5] ,

or, 0.8k u[k 5] 0.5k u[k ] = 1.09227(0.8k 4 0.5k 4 )u[k 5] .

(c) The graphical approach is illustrated in Fig. S10.5(c).

Case (k 6), Fig. S10.5(c) subplot (v):
k 4
16
x1[k ] x2 [ k ] = 0.4 k m 7 m = 0.4 k (2 / 35)m = 33 353 7 k .
m = m = ( k 4)

Case (2 k 6), Fig. S10.5(b) subplot (vii):

2
353
x1[k ] x2 [k ] = 0 .4 k m 7 m = 0 .4 k (2 / 35) m =
132
0 .4 k .
m = m = 2

The convolution sum is plotted as a function of k in subplot (viii) of Fig. S10.5(c).

x1[m] = 7 m u[m + 2] x2 [m] = 0.4 m u[m 4]

m m
6 5 4 3 2 1 0 1 2 3 4 5 6 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7

(i) Sequence x1[m] (ii) Sequence x2[m]

x2 [ m] = 0.4 m u[ m 4] x2 [k m] = 0.4 k m u[k m 4]

m m
6 5 4 3 2 1 0
k6
k5

k2
k1
k
k+1

k+3

k+6
k4
k3

k+2

k+4
k+5

k+7

1 2 3 4 5 6 7

(iii) Sequence x2[m] (iv) Sequence x2[km]

x1[m] x2 [m] x1[m] x2 [m]

m m
6 5 4 3 2 1 0 1 2 3 4 5 6 7 6 5 4 3 2 1 0 1 2 3 4 5 6 7
k11
k14
k13
k12

k10
k9
k8
k7
k6

k4
k5
k7
k8

k5
k6

k4

(v) Convolution x1[m] * x2[m] for k 6 (vi) Convolution x1[m] * x2[m] for k 6
y[k ]
1.33
0.53
0.19
0.004 0.02
k
6 5 4 3 2 1 0 1 2 3 4 5 6 7
6 Chapter 10

(vii) Convolution x1[m] * x2[m] as a function of k.

Fig. S10.5(c) Convolution sum for part (c) in Problem 10.5.

(d) The graphical approach is illustrated in Fig. S10.5(d).

Case (k 1), Fig. S10.5(d) subplot (v):

1 1
x1[ k ] x 2 [ k ] = 0.6 m sin(0.5(k m)) =
j 2 m =0

0.6 m e j 0.5( k m )
j 2 m =0

0.6 m e j 0.5( k m ) ,
m =0

which reduces to

1 j 0.5k 1 j 0.5k
x1[ k ] x 2 [k ] =
j2
e
(0.6e j 0.5 ) m
j 2
e (0.6e j 0.5 ) m ,
m =0 m =0

1 j 0.5k 1 1 j 0.5k 1
or, x1[ k ] x 2 [k ] = e
e .
j2 1 0.6e j 0. 5 j2 1 0.6e j 0.5

x1[m] = 0.6 m u[ m] x2 [ m] = sin(0.5m)u[ m]

5 1
m m
5 4 3 2 1 0 1 2 3 4 5 4 3 2 0 1 2 3 4 5

(i) Sequence x1[m] (ii) Sequence x2[m]

x2 [ m] = sin(0.5m)u[m] x2 [ k m ]

1 5
m m
5 4 3 2 1 0
k3
k2
k1

k+2

k+6
k+7
k

k+3
k+4

2 3 4

(iii) Sequence x2[m] (iv) Sequence x2[km]

x1[m] x2 [ k m] x1[ m] x2 [k m]

m m
5 4 3 2 1 0 1 2 3 4 5 5 4 3 2 1 0 1 2 3 4 5
k3
k2

k+3
k+4
k1

k+2

k5

k2

k+4
k4
k3

k1

k+2
k+3
k+6
k+7
k

(v) Convolution x1[m] * x2[m] for k 1 (vi) Convolution x1[m] * x2[m] for k 0
y[k ]
0.6
0.5
5 4 1 0 1 2 3 4 5
k
3 2 0.13
0.22
0.5 0.5 0.36
0.6 0.6

(vii) Convolution x1[m] * x2[m] as a function of k.

Fig. S10.5(d) Convolution sum for part (d) in Problem 10.5.

Simplifying, we get
 Solutions 7

1 j 0.5k jk 1 1
x1[ k ] x 2 [k ] = e e .
j2 1 + j 0.6 1 j 0.6

If k is even, exp(jk) = 1, therefore,

1 j 0.5k 1 1 j 0.5k
x1 [ k ] x 2 [k ] = e = 0.6e .
j2 1 + j 0.6 1 j 0.6
If k is odd, exp(jk) = 1, therefore,
1 j 0.5k 1 1 j 0.5 k
x1 [k ] x 2 [k ] = e = je .
j2 1 + j 0. 6 1 j 0 . 6
Case (k 0), Fig. S10.5(d) subplot (vi):

1 1
x1[ k ] x 2 [k ] =
j 2 m=k
0.6 m e j 0.5( k m )
j 2 m=k

0.6 m e j 0.5( k m ) ,

which reduces to

1 j 0.5k 1 j 0.5k
x1 [k ] x 2 [k ] =
j2
e
(0.6e j 0.5 ) m
j 2
e
(0.6e j 0.5 ) m ,
m=k m=k

1 j 0.5k 0.6 k e j 0.5k 1 j 0.5k 0.6 k e j 0.5k

or, x1[ k ] x 2 [k ] = e e ,
j2 1 0.6e j 0.5 j2 1 0.6e j 0.5

0 .6 k 1 1 0 .6 k
or, x1[ k ] x 2 [k ] = = (1.2 j sin(0.5)) ,
j 2 1 0.6e j 0.5 1 0.6e j 0.5
j2

0 .6 k
or, x1 [k ] x 2 [ k ] = (1.2 j sin(0.5)) = 0.6 k +1 .
j2
The convolution sum is plotted as a function of k in subplot (vii) of Fig. S10.5(d). Note that the
result of convolution sum is always real as two real valued sequences are being convolved.
(e) The graphical approach is illustrated in Fig. 10.5(e).
Case (k < 0), Fig. S10.5(e) subplot (v):
k 1 0
x1[ k ] x 2 [k ] = 0.5 m 0.8 ( k m) + 0.5 m 0.8 ( k m) + 0 .5 m 0 .8 ( k m ) ,
m = m=k m =1

which reduces to
k 1 0
x1[ k ] x 2 [ k ] = 0.8 k (0.4) m + 0.8 k 0.625 m + 0.8 k 0.4 m ,
m = m=k m =1

k
or, x1[ k ] x 2 [k ] = 0.8 k 0.4 m + 0.8 k 0.625 m + 0.8 k 0.4 m ,
m = ( k 1) m=0 m =1
8 Chapter 10

0.4 ( k 1) (1 0.625 k +1 ) 0.4

or, x1[ k ] x 2 [k ] = 0.8 k + 0 .8 k + 0 .8 k
1 0.4 1 0.625 1 0 .4

or, x1[ k ] x 2 [ k ] =
2
3
[
2 k + 4 0.8 k (1 0.625 k +1 ) + 0.8 k . ]
Case (k 0), Fig. S10.5(e) subplot (v):
1 k
x1[ k ] x 2 [k ] = 0.5 m 0.8 ( k m ) + 0.5 m 0.8 ( k m ) + 0 .5 m 0 .8 ( k m ) ,
m = m =0 m = k +1

which reduces to
1 k
x1[ k ] x 2 [ k ] = 0.8 k 0.4 m + 0.8 k 0.625 m + 0.8 k 0.4 m ,
m = m =0 m = k +1

k
or, x1[ k ] x 2 [ k ] = 0.8 k
0 .4 m
+ 0 .8 k
0.625 m
+ 0.8 k
0.4 m ,
m =1 m =0 m = k +1

m m
x1[m] = 0.5 x2 [m] = x2 [m] = 0.8

m m
6 5 4 3 2 1 0 1 2 3 4 5 6 6 5 4 3 2 1 0 1 2 3 4 5 6

(i) Sequence x1[m] (ii) Sequence x2[m]

k m
x2 [m] = x2 [m] = 0.8
m x2 [k m] = 0.8

m m
k1

k+3
k+4
k3
k2

k+2

6 5 4 3 2 1 0 1 2 3
k+1
k6
k5
k4

k+5
k+6
k

4 5 6

(iii) Sequence x2[m] (iv) Sequence x2[km]

x1[m] x2 [ k m] x1[m] x2 [ k m]

m m
6 5 4 3 2 1 0 1 2 3 4 5 6 6 5 4 3 2 1 0 1 2 3 4 5 6
k+3
k+4

k+7

k3
k2

k+3
k+4
k3
k2
k1

k+2

k+9
k+8

k1

k+2
k+1

k+1
k4

k+7

k4
k5

k+5
k+6

k5

k
k

(v) Convolution x1[m] * x2[m] for k < 0. (vi) Convolution x1[m] * x2[m] for k 0
2.172.332.17 y[k ]
1.88 1.88
1.58 1.58
1.30 1.30
1.06 1.06
0.690.86 0.860.69

k
7 6 5 4 3 2 1 0 1 2 3 4 5 6 7

(vii) Convolution x1[m] * x2[m] as a function of k.

Fig. S10.5(e): Convolution sum for part (e) in Problem 10.5.
 Solutions 9

0 .4 (1 0.625 k +1 ) 0.4 k +1
or, x1[ k ] x 2 [k ] = 0.8 k + 0.8 k + 0 .8 k
1 0.4 1 0.625 1 0.4

or, x1 [k ] x 2 [k ] =
2
3
[
0.8 k + 4 0.8 k (1 0.625 k +1 ) + 2 k .]
The convolution sum is plotted as a function of k in subplot (vii) of Fig. S10.5(e).

Problem 10.6
(a) Direct Convolution Approach:
Expressing x[k ] = [k 1] + 2[k 2] + 3[k 3]
and h[k ] = 2[k + 1] + 2[k ] + 2[k 1] + 2[k 2] ,
the convolution is given by
x[k ] h[k ] = ([ k 1] + 2[ k 2] + 3[k 3]) (2[k + 1] + 2[ k ] + 2[k 1] + 2[k 2]) ,
which reduces to
x[k ] h[k ] = (2[k ] + 2[k 1] + 2[k 2] + 2[k 3]) + (4[k 1] + 4[k 2]
,
+ 4[k 3] + 4[k 4]) + (6[k 2] + 6[k 3] + 6[k 4] + 6[k 5])
or, x[k ] h[k ] = 2[k ] + 6[k 1] + 12[k 2] + 12[k 3] + 10[k 4] + 6[k 5] .

Sliding tape method: The convolution using the sliding tape method is shown in Table S10.6(a).

Table S10.6(a): Convolution of x[k] and h[k] using the sliding tape method Problem 10.6(a).
5 4 3 2 1 0 1 2 3 4 5 6 7 k y[k]
x[m] 1 2 3
h[m] 2 2 2 2
h[m] 2 2 2 2
h[1 m] 2 2 2 2 1 0
h[0 m] 2 2 2 2 0 2
h[1 m] 2 2 2 2 1 6
h[2 m] 2 2 2 2 2 12
h[3 m] 2 2 2 2 3 12
h[4 m] 2 2 2 2 4 10
h[5 m] 2 2 2 2 5 6
h[6 m] 2 2 2 2 6 0

The results from the two approaches are the same.

(b) Direct Convolution Approach:

x[k ] = 2[k + 2] + [k + 1] + [k 1] + 2[k 2]
h[k ] = [k ] + 0.5[k 1] + 0.25[k 2] + 0.125[k 3] ,
The convolution of x[k ] and h[k ] is given by
10 Chapter 10

x[k ] h[k ] = (2 [k + 2] + [k + 1] + [k 1] + 2 [k 2]) ( [k ] + 0.5 [k 1] + 0.25 [k 2] + 0.125 [k 3])

= (2 [k + 2] + [k + 1] + 0.5 [k ] + 0.25 [k 1])
+ ( [k + 1] + 0.5 [k ] + 0.25 [k 1] + 0.125 [k 2])
,
+ ( [k 1] + 0.5 [k 2] + 0.25 [k 3] + 0.125 [k 4])
+ (2 [k 2] + [k 3] + 0.5 [k 4] + 0.25 [k 5])
= 2 [k + 2] + 2 [k + 1] + [k ] + 1.5 [k 1] + 2.625 [k 2] + 1.25 [k 3] + 0.625 [k 4]
+ 0.25 [k 5]
or, x[k ] h[k ] = 2[k + 2] + 2[k + 1] + [k ] + 1.5[k 1] + 2.625[k 2] + 1.25[k 3] + 0.625[k 4] + 0.25[k 5] .
Sliding tape method: The convolution using the sliding tape method is shown in Table S10.6(b).
The results from the two approaches are the same.

Table S10.6(a): Convolution of x[k] and h[k] using the sliding tape method Problem 10.6(b).
6 5 4 3 2 1 0 1 2 3 4 5 6 k y[k]
x[m] 2 1 0 1 2
h[m] 1 0.5 0.25 0.125
h[m] 0.125 0.25 0.5 1
h[3 m] 0.125 0.25 0.5 1 3 0
h[2 m] 0.125 0.25 0.5 1 2 2
h[1 m] 0.125 0.25 0.5 1 1 2
h[0 m] 0.125 0.25 0.5 1 0 1
h[1 m] 0.125 0.25 0.5 1 1 1.5
h[2 m] 0.125 0.25 0.5 1 2 2.625
h[3 m] 0.125 0.25 0.5 1 3 1.25
h[4 m] 0.125 0.25 0.5 1 4 0.625
h[5 m] 0.125 0.25 0.5 1 5 0.25
h[6 m] 0.125 0.25 0.5 1 6 0

Problem 10.7

The convolution using the sliding tape method is shown in Table S10.7.

Table S10.7: Convolution of x[k] and h[k] using the sliding tape method Problem 10.7.
5 4 3 2 1 0 1 2 3 4 5 6 7 k y[k]
h[m] 1 2 3 4 5
x[m] 2 2 2
x[m] 2 2 2
x[1 m] 2 2 2 1 0
x[0 m] 2 2 2 0 2
x[1 m] 2 2 2 1 6
x[2 m] 2 2 2 2 12
x[3 m] 2 2 2 3 18
x[4 m] 2 2 2 4 24
x[5 m] 2 2 2 5 18
x[6 m] 2 2 2 6 10
x[7 m] 2 2 2 7 0
 Solutions 11

The result matches with that obtained in Example 10.8.

Problem 10.8

The convolution using the sliding tape method is shown in Table S10.8.

Table S10.8: Convolution of x[k] and h[k] using the sliding tape method Problem 10.8.
6 5 4 3 2 1 0 1 2 3 4 5 6 k y[k]
x[m] -1 1 2
h[m] 3 1 -2 3 -2
h[m] -2 3 -2 1 3
h[3 m] -2 3 -2 1 3 3 0
h[2 m] -2 3 -2 1 3 2 -3
h[1 m] -2 3 -2 1 3 1 2
h[0 m] -2 3 -2 1 3 0 9
h[1 m] -2 3 -2 1 3 1 -3
h[2 m] -2 3 -2 1 3 2 1
h[3 m] -2 3 -2 1 3 3 4
h[4 m] -2 3 -2 1 3 4 -4
h[5 m] -2 3 -2 1 3 5 0

The result matches with that obtained in Example 10.9.

Direct Convolution Approach:

x[k ] = [ k + 1] + [k ] + 2[k 1]
h[k ] = 3[k + 1] + [ k ] 2[k 1] + 3[k 2] 2[k 3] ,
The convolution is given by
x[k ] h[k ] = ( [k + 1] + [k ] + 2 [k 1]) (3 [k + 1] + [k ] 2 [k 1] + 3 [k 2] 2 [k 3])
= (3 [k + 2] [k + 1] + 2 [k ] 3 [k 1] + 2 [k 2])
+ (3 [k + 1] + [k ] 2 [k 1] + 3 [k 2] 2 [k 3])
+ (6 [k ] + 2 [k 1] 4 [k 2] + 6 [k 3] 4 [k 4])
= 3 [k + 2] + 2 [k + 1] + 9 [k ] 3 [k 1] + [k 2] + 4 [k 3] 4 [k 4].

Problem 10.9

The periodic convolution with K0 = 10 is shown in Table S10.9(a).

Table S10.9(a): Periodic convolution of xp[k] and hp[k] in Problem 10.9 with K0 set to 10.
0 1 2 3 4 5 6 7 8 9 k yp[k]
hp[k] 1 2 3 4 5 0 0 0 0 0
xp[k] 2 2 2 0 0 0 0 0 0 0
xp[k] 2 0 0 0 0 0 0 0 2 2 0 2
xp[1k] 2 2 0 0 0 0 0 0 0 2 1 6
xp[2k] 2 2 2 0 0 0 0 0 0 0 2 12
12 Chapter 10

xp[3k] 0 2 2 2 0 0 0 0 0 0 3 18
xp[4k] 0 0 2 2 2 0 0 0 0 0 4 24
xp[5k] 0 0 0 2 2 2 0 0 0 0 5 18
xp[6k] 0 0 0 0 2 2 2 0 0 0 6 10
xp[7k] 0 0 0 0 0 2 2 2 0 0 7 0
xp[8k] 0 0 0 0 0 0 2 2 2 0 8 0

The periodic convolution with K0 = 13 is shown in S10.9(b).

Table S10.9(b): Periodic convolution of xp[k] and hp[k] in Problem 10.9 with K0 set to 13.
0 1 2 3 4 5 6 7 8 9 10 11 12 k yp[k]
hp[k] 1 2 3 4 5 0 0 0 0 0 0 0 0
xp[k] 2 2 2 0 0 0 0 0 0 0 0 0 0
xp[k] 2 0 0 0 0 0 0 0 0 0 0 2 2 0 2
xp[1k] 2 2 0 0 0 0 0 0 0 0 0 0 2 1 6
xp[2k] 2 2 2 0 0 0 0 0 0 0 0 0 0 2 12
xp[3k] 0 2 2 2 0 0 0 0 0 0 0 0 0 3 18
xp[4k] 0 0 2 2 2 0 0 0 0 0 0 0 0 4 24
xp[5k] 0 0 0 2 2 2 0 0 0 0 0 0 0 5 18
xp[6k] 0 0 0 0 2 2 2 0 0 0 0 0 0 6 10
xp[7k] 0 0 0 0 0 2 2 2 0 0 0 0 0 7 0
xp[8k] 0 0 0 0 0 0 2 2 2 0 0 0 0 8 0
xp[9k] 0 0 0 0 0 0 0 2 2 2 0 0 0 9 0
xp[10k] 0 0 0 0 0 0 0 0 2 2 2 0 0 10 0
xp[11k] 0 0 0 0 0 0 0 0 0 2 2 2 0 11 0
xp[12k] 0 0 0 0 0 0 0 0 0 0 2 2 2 12 0

Since in both cases the fundamental period K 0 K1 + K 2 1 (= 7) , the results of periodic convolution over
one period are the same as that of linear convolution.

Problem 10.10
The periodic convolution with K0 = 10 is shown in Table S10.10.
Table S10.10: Periodic convolution of xp[k] and hp[k] in Problem 10.10 with K0 set equal to 10.
0 1 2 3 4 5 6 7 8 9 k yp[k]
hp[k] 1 2 3 4 5 0 0 0 0 0
xp[k] 2 2 2 0 0 0 0 0 0 0
xp[k] 2 0 0 0 0 0 0 0 2 2 0 2
xp[1k] 2 2 0 0 0 0 0 0 0 2 1 6
xp[2k] 2 2 2 0 0 0 0 0 0 0 2 12
xp[3k] 0 2 2 2 0 0 0 0 0 0 3 18
xp[4k] 0 0 2 2 2 0 0 0 0 0 4 24
xp[5k] 0 0 0 2 2 2 0 0 0 0 5 18
xp[6k] 0 0 0 0 2 2 2 0 0 0 6 10
xp[7k] 0 0 0 0 0 2 2 2 0 0 7 0
xp[8k] 0 0 0 0 0 0 2 2 2 0 8 0
 Solutions 13

Since the fundamental period K 0 (= 10) K1 + K 2 1 (= 7) of the two sequences, the result of periodic
convolution over one period is the same as that of linear convolution.

Problem 10.11

The periodic convolution with K0 = 15 is shown in Table S10.11.

Table S10.11: Periodic convolution of xp[k] and hp[k] in Problem 10.9 with K0 set to 15.
0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 k yp[k]
hp[k] 1 2 3 4 5 0 0 0 0 0 0 0 0 0 0
xp[k] 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0
xp[k] 2 0 0 0 0 0 0 0 0 0 0 0 0 2 2 0 2
xp[1k] 2 2 0 0 0 0 0 0 0 0 0 0 0 0 2 1 6
xp[2k] 2 2 2 0 0 0 0 0 0 0 0 0 0 0 0 2 12
xp[3k] 0 2 2 2 0 0 0 0 0 0 0 0 0 0 0 3 18
xp[4k] 0 0 2 2 2 0 0 0 0 0 0 0 0 0 0 4 24
xp[5k] 0 0 0 2 2 2 0 0 0 0 0 0 0 0 0 5 18
xp[6k] 0 0 0 0 2 2 2 0 0 0 0 0 0 0 0 6 10
xp[7k] 0 0 0 0 0 2 2 2 0 0 0 0 0 0 0 7 0
xp[8k] 0 0 0 0 0 0 2 2 2 0 0 0 0 0 0 8 0
xp[9k] 0 0 0 0 0 0 0 2 2 2 0 0 0 0 0 9 0
xp[10k] 0 0 0 0 0 0 0 0 2 2 2 0 0 0 0 10 0
xp[11k] 0 0 0 0 0 0 0 0 0 2 2 2 0 0 0 11 0
xp[12k] 0 0 0 0 0 0 0 0 0 0 2 2 2 0 0 12 0
xp[13k] 0 0 0 0 0 0 0 0 0 0 0 2 2 2 0 13 0
xp[14k] 0 0 0 0 0 0 0 0 0 0 0 0 2 2 2 14 0
Since the fundamental period K 0 (= 15) K1 + K 2 1 (= 7) of the two sequences, the result of periodic
convolution over one period is the same as that of linear convolution.

Problem 10.12
The periodic convolution with K0 = 8 is shown in Table S10.12.
Table S10.12: Periodic convolution of xp[k] and hp[k] in Problem 10.12 with K0 set to 8.
0 1 2 3 4 5 6 7 k yp[k]
hp[k] 5 5 0 0 0 0 0 0
xp[k] 0 1 2 3 0 0 0 0
xp[k] 0 0 0 0 0 3 2 1 0 0
xp[1k] 1 0 0 0 0 0 3 2 1 5
xp[2k] 2 1 0 0 0 0 0 3 2 15
xp[3k] 3 2 1 0 0 0 0 0 3 25
xp[4k] 0 3 2 1 0 0 0 0 4 15
xp[5k] 0 0 3 2 1 0 0 0 5 0
xp[6k] 0 0 0 3 2 1 0 0 6 0
xp[7k] 0 0 0 3 2 1 0 3 7 0
14 Chapter 10

Note that the non-zero values of the periodic convolution are the same as the ones obtained in Example
10.12. Since the period has been increased, additional zeros are added.

Problem 10.13
The unit step response s[k] of the DT system is the output of the system when a unit step u[k] is applied at
the input of the system. In other words,

s[k ] = h[k ] u[k ] = h[ p]u[k p]
p =
.
k 1 pk
= h[ p] u[k p] =
p = 0 otherwise
In other words, the unit step response is the running sum of the impulse response.

0 k < 7
k k

(a) s[k ] = h[ p ] = ( u[ p + 7] u[ p 8]) = (k + 8) 7 k 7
p = p = 15 k > 7.


0 k <0
k
0 k <0
(b) s[k ] = 0.4 u[ p ] = k
p
= 5 k +1
(1 0.4 ) k 0.
0.4 p k 0 3

p =

p =0

k p
2 k<0
k
p = 2k +1 k<0
(c) s[k ] = 2 u[ p ] = 0
p
=
2 k 0.
k0
p
p =
2p
=

0.6 p k <0
2.5 0.6 k k <0
k

s[k ] = 0.6 = 0
p p =
(d) =
2.5 + 1.5 (1 0.6 ) k 0.
k k

p
p = p
0.6 + 0.6 p k 0
= p =1

1 k = 4n; 0, 4, 8,....

(e) h[k ] = (1) (k 2m) = 1
m
k = 4n + 2; 2, 6,....
m = 0 otherwise

The plot of h[k ] is shown in Fig. S10.13. It is observed that h[k ] is a zero-mean periodic signal with
k
period 4. The running sum s[ k ] = h[ p] will depend on how h[k ]
p =
starts at k = . If h[] = 1 ,

the sequence s[k ] will have the form 1 ,1, 0, 0,1,1, 0, 0,1,1,......... . On the other hand, if
at k =
 Solutions 15

h[] = 1 , the sequence s[k ] will have the form 1 , 1, 0, 0, 1, 1, 0, 0, 1, 1,......... . As we do
at k =
not know if = 4n or (4n + 1) or (4n + 2) or (4n + 3) , we cannot calculate s[k ] exactly. The only
observation we can make with certainty is that s[k ] will have three possible values 1, 0,1 .

Problem 10.14

(a) (x[k] + 2[k 1]) * [k 2] = x[k] * [k 2] + 2[k 1] * [k 2] [using distributive property]

= x[k 2] + 2[k 3].

(b) Using the distributive property

(x[k] + 2[k 1]) * ([k + 1] + [k 2])
= x[k] * [k + 1] + 2[k 1] * [k + 1] + x[k] * [k 2] + 2[k 1] * [k 2]
= x[k + 1] + 2[k] + x[k 2] + 2[k 3]
= x[k + 1] + x[k 2] + 2[k] + 2[k 3].

(c) (x[k] u[k 1]) * [k 2] = x[k] * [k 2] u[k 1] * [k 2] [using distributive property]

= x [k 2] u[k 3].

(d) ( x[k ] x[k 1]) u[k ] = ( x[k ] x[k 1]) [k m]
m=0

= x[k m] x[k 1 m] [using distributive property]
m=0 m=0

= x[k ]

Problem 10.15
The result is straightforward to prove because zero padding ensures that there is no overlap between two
replicas of the linear convolution of x[k] and h[k].

Problem 10.16
Using the Eq. (10.25) (convolution with impulse function property), the left hand side of Eq. (10.24) can
be expressed as
x1[ k k1 ] x 2 [k k 2 ] = ( x1[k ] [k k1 ]) ( x 2 [k ] [k k 2 ])
Using the commutative and associative property, the above expression can be expressed as
x1[k k1 ] x2 [k k2 ] = ( x1[k ] x2 [k ]) ( [k k1 ]) [k k2 ])
g[k ] [ k k1 k2 ]

= g[k ] [k k1 k2 ]
= g[k k1 k2 ] .
16 Chapter 10

Problem 10.17
Note that the finite duration DT sequences x1[k] and x2[k] can be expressed as
x1 [k ] = x1 [k 1 ][k k 1 ] + x1[k 1 + 1][k k 1 1] + + x1 [k u1 ][ k k u1 ]
and x 2 [k ] = x 2 [k 2 ][k k 2 ] + x 2 [k 2 + 1][k k 2 1] + + x 2 [k u 2 ][ k k u 2 ] .

When x1[k] is convolved by x2[k], the earliest sample will be the result of the convolution of the term
x1[k 1 ][k k 1 ] with x 2 [ k 2 ][k k 2 ] , while the latest sample will be the result of the convolution of
the term x1[ku1 ][k ku1 ] with x2 [ku 2 ][k ku 2 ] . In other words,
Earliest term = x1[ k 1 ][k k 1 ] x 2 [k 2 ][ k k 2 ] = x1[k 1 ]x 2 [k 2 ][k k 1 k 2 ],
and Latest term = x1[ k u1 ][k k u1 ] x 2 [k u 2 ][k k u 2 ] = x1 [k u1 ] x 2 [k u 2 ][ k k u1 k u 2 ] ,
which proves that the linear convolution y[k] is time-limited within the range (k1 + k2 k ku1 + ku2).

Problem 10.18

1 7 k 7
(a) h[k ] = u[k + 7] u[k 8] =
0 otherwise.
We observe that:
(i) h[k] 0 for all k 0. Therefore, the system is NOT memoryless.
(ii) h[k] 0 for k < 0, e.g., h[1] = 1. Therefore, the system is NOT causal.
(iii) |h[k]| = 15 < . Therefore, the system is bounded-input-bounded-output (BIBO) stable.

sin( k / 8) k 0
(b) h[k ] = sin( k / 8)u[k ] =
0 k < 0.
We observe that:
(i) h[k] 0 for all k 0. Therefore, the system is NOT memoryless.
(ii) h[k] = 0 for k < 0. Therefore, the system is causal.
(iii) |h[k]| = . Therefore, the system is NOT stable.

6 k k 0
(c) h[k ] = 6k u[ k ] =
0 k > 0.
We observe that:
(i) h[k] 0 for all k 0. Therefore, the system is NOT memoryless.
(ii) h[k] 0 for k < 0, e.g., h[2] = 1/36. Therefore, the system is NOT causal.
(iii) |h[k]| = 1/(1 1/6) = 1.2 < . Therefore, the system is stable.
 Solutions 17

k 0.9k k 0
(d) h[k ] = 0.9 = k
0.9 k < 0.
we observe that:
(i) h[k] 0 for all k 0. Therefore, the system is NOT memoryless.
(ii) h[k] 0 for k < 0, e.g., h[2] = 0.81. Therefore, the system is NOT causal.
(iii) |h[k]| = 21 < . Therefore, the system is BIBO stable.

1 k = 4m

(e) h[k ] = (1) (k 2m) = 1
m
k = 4m + 2
m = 0 k = odd

we observe that:
(i) h[k] 0 for all k 0. Therefore, the system is NOT memoryless.
(ii) h[k] 0 for k < 0, e.g., h[2] = 1. Therefore, the system is NOT causal.
(iii) Since h[k] includes an infinite number of impulse functions, |h[k]| = .Therefore, the system
is NOT stable.

Problem 10.19
(a) h1[k ] h2 [k ] = u[k 1] ( [k 1] [k ]) = u[ k 1] [k 1] u[ k 1] [k ]
= u[ k ] u[ k 1] = [k ]
Therefore, the pair corresponds to inverse systems.
(b) h1[k ] h2 [k ] = 0.5k u[k ] ( [k ] 0.5 [k 1]) = 0.5k u[k ] [k ] 0.5 0.5k u[k ] [k 1]

= 0.5k u[k ] 0.5k u[k 1] = [k ]

Therefore, the pair corresponds to inverse systems.
(c) h1[k ] h2 [k ] = 0.8k ku[k ] (0.8 [k 1] 2 [k ] + 1.25 [k + 1])

= 0.8k ku[k ] 0.8 [k 1] 2 0.8k ku[k ] [k ] + 1.25 0.8k ku[k ] [k + 1]

= 0.8k (k 1)u[k 1] 2 0.8k ku[k ] + 1.25 0.8k +1 (k + 1)u[k + 1]
= 0.8k u[k 1] ((k 1) 2k + (k + 1)) + 1.25 0.81 [k ]
= [k ].
Therefore, the pair corresponds to inverse systems.
(d) h1[k ] h2 [k ] = ku[k ] ( [k + 1] 2 [k ] + [k 1])
18 Chapter 10

= ku[k ] [k + 1] 2ku[k ] [k ] + ku[k ] [ k 1]

= ( k + 1)u[k + 1] 2ku[k ] + (k 1)u[k 1]
= [ k ] + u[ k 1] ((k + 1) 2k + (k 1))
= [ k ].
Therefore, the pair corresponds to inverse systems.
(e) h1[k ] h2 [k ] = (k + 1)0.8k u[k ] ( [k ] 1.6 [k 1] + 0.64 [k 2])

= (k + 1)0.8k u[k ] [k ] 1.6(k + 1)0.8k u[ k ] [ k 1] + 0.64(k + 1)0.8k u[k ] [k 2])

= (k + 1)0.8k u[k ] 1.6k 0.8k 1 u[k 1] + 0.64(k 1)0.8k 2 u[k 2]
= [( k + 1) 2k + (k 1)]0.8k u[k 2] + [k ] + 2 0.8 [ k 1] 1.6 [k 1]
=0 =0

= [k ]
Therefore, the pair corresponds to inverse systems.

Problem 10.20
The MATLAB code for computing the first 50 samples of the output response y[k] in Problems 10.1
10.3 is given below:

>> % Problem 10.1: y[k] 2y[k-1] = x[k-1]

>> k1 = [0:49]; % time index k = [-1, 0, 1, 49]
>> x1 = 2*(k1 >= 0); % Input Signal x[k] = 2*u[k]
>> A1 = [1 -2]; % Coefficients for y[k]
>> B1 = [0 1]; % Coefficients for x[k]
>> zi1 = filtic(B1,A1,[2]); % Initial Conditions
>> y1 = filter(B1,A1,x1,zi1); % Calculate output
>> subplot(3,1,1); stem(k1,y1,'fill'); title('Problem 10.1'); xlabel('k');
>> axis tight; grid on
>>
>> % Problem 10.2: y[k] y[k-1] + 0.5y[k-2] = x[k-2]
>> k2 = [0:49]; % time index k = [-1, 0, 1, 49]
>> x2 = (0.5.^k2).*(k2 >= 0); % Input Signal x[k] = 0.5^k u[k]
>> A2 = [1 -1 0.5]; % Coefficients for y[k]
>> B2 = [0 0 1]; % Coefficients for x[k]
>> zi2 = filtic(B2,A2,[0 1]); % Initial Conditions
>> y2 = filter(B2,A2,x2,zi2); % Calculate output
>> subplot(3,1,2); stem(k2,y2,'fill'); title('Problem 10.2'); xlabel('k');
>> axis tight; grid on
>>
>> % Problem 10.3: y[k] 0.75y[k-1] + 0.125y[k-2] = x[k-2]
>> k3 = [0:49]; % time index k = [-1, 0, 1, 49]
>> x3 = ((-1).^k3).*(k3 >= 0); % Input Signal x[k] = (-1)^k u[k]
>> A3 = [1 -0.75 0.125]; % Coefficients for y[k]
>> B3 = [0 0 1]; % Coefficients for x[k]
>> zi3 = filtic(B3,A3,[1 -1]); % Initial Conditions
>> y3 = filter(B3,A3,x3,zi3); % Calculate output
>> subplot(3,1,3); stem(k3,y3,'fill'); title('Problem 10.3'); xlabel('k');
>> axis tight; grid on
The stem plots for the three outputs are shown in Fig. S10.20.
 Solutions 19

15
x 10 Problem 10.1

2.5

1.5

0.5

0
0 5 10 15 20 25 30 35 40 45
k

Problem 10.2
1.5

0.5

-0.5
0 5 10 15 20 25 30 35 40 45
k

Problem 10.3

0.5

-0.5
0 5 10 15 20 25 30 35 40 45
k

Fig. S10.20: Stem plots for the outputs of the DT systems in Problem 10.20.

Problem 10.21
The MATLAB code for computing the convolution sums in Problem 10.5 is given below:
>> % Problem 10.5(a)
>> kx1 = [-2:2]; % time indices where x1 is nonzero
>> x1 = 1*ones(length(kx1),1); % Sample values for DT sequence x1
>> kx2 = [-4:4]; % time indices where x2 is nonzero
>> x2 = 1*ones(length(kx2),1); % Sample values for DT sequence x2
>> y1 = conv(x1,x2); % Convolve x1 with x2
>> ky1 = kx1(1)+kx2(1):kx1(length(kx1))+kx2(length(kx2));% ky1=indices for y1
>> subplot(5,1,1); stem(ky1,y1,'fill'); title('Problem 10.5(a)');
>> xlabel('k'); axis tight; grid on
>> % Problem 10.5(b)
>> kx1 = [0:ceil(log(0.001)/log(0.5))]; % time indices where x1 is nonzero
>> x1 = 0.5.^kx1.*(kx1>=0); % Sample values for DT sequence x1
>> kx2 = [0:ceil(log(0.001)/log(0.8))]; % time indices where x2 is nonzero
>> x2 = 0.8.^kx2.*(kx2>=0); % Sample values for DT sequence x2
>> y2 = conv(x1,x2); % Convolve x1 with x2
>> ky2 = kx1(1)+kx2(1):kx1(length(kx1))+kx2(length(kx2));% ky2=indices for y2
>> subplot(5,1,2); stem(ky2,y2,'fill'); title('Problem 10.5(b)');
>> xlabel('k'); axis tight; grid on
20 Chapter 10

>> % Problem 10.5(c)

>> kx1 = [floor(log(0.049)/log(7)):2]; % time indices where x1 is nonzero
>> x1 = 7.^kx1.*(kx1<=2); % Sample values for DT sequence x1
>> kx2 = [4:ceil(log(0.001*0.4^4)/log(0.4))];
% time indices where x2 is nonzero
>> x2 = 0.4.^kx2.*(kx2>=4); % Sample values for DT sequence x2
>> y3 = conv(x1,x2); % Convolve x1 with x2
>> ky3 = kx1(1)+kx2(1):kx1(length(kx1))+kx2(length(kx2));% ky3=indices for y3
>> subplot(5,1,3); stem(ky3,y3,'fill'); title('Problem 10.5(c)');
>> xlabel('k'); axis tight; grid on
>> % Problem 10.5(d)
>> kx1 = [0:ceil(log(0.001)/log(0.6))]; % time indices where x1 is nonzero
>> x1 = 0.6.^kx1.*(kx1>=0); % Sample values for DT sequence x1
>> kx2 = [-30:0]; % time indices where x2 is nonzero
>> x2 = sin(pi*kx2/2).*(kx2<=0); % Sample values for DT sequence x2
>> y4 = conv(x1,x2); % Convolve x1 with x2
>> ky4 = kx1(1)+kx2(1):kx1(length(kx1))+kx2(length(kx2));% ky4=indices for y4
>> subplot(5,1,4); stem(ky4,y4,'fill'); title('Problem 10.5(d)');
>> xlabel('k'); axis tight; grid on
>> % Problem 10.5(e)
>> kx1 = [-floor(log(0.001)/log(0.5)):ceil(log(0.001)/log(0.5))];
% time indices where x1 is nonzero
>> x1 = 0.5.^abs(kx1); % Sample values for DT sequence x1
>> kx2 = [-floor(log(0.001)/log(0.8)):ceil(log(0.001)/log(0.8))];
% time indices where x2 is nonzero
>> x2 = 0.8.^abs(kx2); % Sample values for DT sequence x2
>> y5 = conv(x1,x2); % Convolve x1 with x2
>> ky5 = kx1(1)+kx2(1):kx1(length(kx1))+kx2(length(kx2));% ky5=indices for y5
>> subplot(5,1,5); stem(ky5,y5,'fill'); title('Problem 10.5(e)');
>> xlabel('k'); axis tight; grid on

The stem plots for the five convolution sums are shown in Fig. S10.21.
 Solutions 21

Problem 10.5(a)

0
-6 -4 -2 0 2 4 6
k
Problem 10.5(b)

0.5

0
0 5 10 15 20 25 30 35 40
k
Problem 10.5(c)

0.5

0
2 4 6 8 10 12 14
k
Problem 10.5(d)

0.5

-0.5

-1
-30 -25 -20 -15 -10 -5 0 5 10
k
Problem 10.5(e)

0
-30 -20 -10 0 10 20 30 40
k

Fig. S10.21: Stem plots for the results of convolution sums in Problem 10.21.

Problem 10.22
The MATLAB code for determining the impulse responses of the systems specified in Problem 10.1
10.3 is as follows:

>> % Problem 10.1: y[k] 2y[k-1] = x[k-1]

>> A1 = [1 -2]; % Coefficients for y[k]
>> B1 = [0 1]; % Coefficients for x[k]
>> [h1,k1] = impz(B1,A1,50); % Calculate impulse response
>> subplot(3,1,1); stem(k1,h1,'fill');
>> title('Problem 10.1'); xlabel('k');
>> axis tight; grid on
>> % Problem 10.2: y[k] y[k-1] + 0.5y[k-2] = x[k-2]
>> A2 = [1 -1 0.5]; % Coefficients for y[k]
>> B2 = [0 0 1]; % Coefficients for x[k]
>> [h2,k2] = impz(B2,A2,50); % Calculate impulse response
>> subplot(3,1,2); stem(k2,h2,'fill');
>> title('Problem 10.2'); xlabel('k');
>> axis tight; grid on
>> % Problem 10.3: y[k] 0.75y[k-1] + 0.125y[k-2] = x[k-2]
>> A3 = [1 -0.75 0.125]; % Coefficients for y[k]
>> B3 = [0 0 1]; % Coefficients for x[k]
22 Chapter 10

>> [h3,k3] = impz(B3,A3,50); % Calculate impulse response

>> subplot(3,1,3); stem(k3,h3,'fill');
>> title('Problem 10.3'); xlabel('k');
>> axis tight; grid on

The impulse responses of the DT systems are shown in Fig. S10.22.

14 Problem 10.1
x 10

2.5

1.5

0.5

0
0 5 10 15 20 25 30 35 40 45
k

Problem 10.2

0.8

0.6

0.4

0.2

-0.2
0 5 10 15 20 25 30 35 40 45
k

Problem 10.3
1

0.8

0.6

0.4

0.2

0
0 5 10 15 20 25 30 35 40 45
k

Fig. S10.22: Impulse responses for systems defined using difference equations in Problem 10.22.