Signals & Systems - Chapter 2: K N X K H

Download as pdf or txt
Download as pdf or txt
You are on page 1of 13

Signals & Systems - Chapter 2

1S. Let x[n] = δ[n] + 2δδ[n – 1] – δ[n – 3] and h[n] = 2δδ[n + 1] + 2δ


δ[n – 1]
Compute and plot each of the following convolutions:
a) y1[n] = x[n] * h[n] b) y2[n] = x[n+2] * h[n] c) y3[n] = x[n] * h[n+2]

Solution:

a)

Theory—We have y1[n] = x[n] * h[n] = ∑ h[k ]x[n − k ]
k = −∞

2 x[n] 2 2 h[n]
1
n n
0 1 2 3 4 -2 -1 0 1 2
-1

y1[n] = h[-1]x[n+1] + h[1]x[n-1] = 2x[n+1] + 2x[n-1]


y1[n]= 2δ[n+1] + 4δ[n] – 2δ[n – 2] + 2δ[n-1] + 4δ[n – 2] – 2δ[n – 4]
y1[n]= – 2δ[n – 4] + 2δ[n – 2] + 2δ[n-1] + 4δ[n] + 2δ[n+1]

b) Note x[n] x[n+2] which is a shift of +2


Since we have linear time invariant system

y2[n] = x[n+2] * h[n] = ∑ h[k ]x[(n + 2) − k ]
k = −∞
y2[n] = y1[n+2] Therefore

y2[n]= – 2δ[n – 2] + 2δ[n] + 2δ[n+1] + 4δ[n+2] + 2δ[n+3]

c) Note h[n] h[n+2] which is a shift of +2


Since we have linear time invariant system
∞ ∞
y3[n] = x[n] * h[n+2] = ∑ h[k + 2]x[n − k ] = ∑ h[k ]x[n + 2 − k ]
k = −∞ k = −∞
y3[n] = y1[n+2] Therefore

y3[n]= – 2δ[n – 2] + 2δ[n] + 2δ[n+1] + 4δ[n+2] + 2δ[n+3]

1U. Let x[n] = δ[-n] + 2δδ[n + 1] – δ[n – 4] and h[n] = δ[n + 2] + 3δ


δ[n + 1]
Compute and plot each of the following convolutions:
a) y1[n] = x[n] * h[n] b) y2[n] = x[n-2] * h[n] c) y3[n] = x[n] * h[n-2]
Solution:

2S. Consider an input x[n] and a unit impulse response h[n] given by
1
x[n] = ( ) n− 2 u[n − 2]
2
h[n] = u[n + 2]
Determine and plot the output y[n] = x[n]*h[n].

Solution:

www.EngrCS.com, ik Signals and Systems page 16


a)

Theory—We have y1[n] = x[n] * h[n] = ∑ x[k ]h[n − k ]
k = −∞
k −2
∞ ∞
1
y[ n] = x[ n] * h[ n] = ∑ x[k ]h[n − k ] =
k = −∞
∑  
k = −∞  2 
u[ k − 2]u[ n + 2 − k ]

taking a look at plot of the two step function we see that


u[k-2] = 1 when k-2≥0 or k≥2
0 otherwise
u[n+2-k] = 1 when n+2-k≥0 or k≤n+2
0 otherwise

u[k-2] k
2

u[n+2-k]
k
n+2

There are two conditions:


1) when n+2 < 2 y[n] = 0 for n<0
k −2 k
n+2
1 1n
2) when n+2 ≥ 2 y[ n] = ∑   =∑   for n≥0
k =2  2  k =0  2 
n
1 − a n +1
Apply finite sum formula ∑
k =0
a k
=
1− a
for n≥0 & 0< |a| <1

1 − (1 / 2) n +1
y[n] = u[ n] = 2[1 – (1/2)n+1]u[n]
1 − 1/ 2

2U. Consider an input x[n] and a unit impulse response h[n] given by
1
x[n] = ( ) n−3 u[n + 2]
4
h[n] = 2u[n − 2]
Determine and plot the output y[n] = x[n]*h[n].

Solution:

3S. Compute and plot y[n] = x[n] * h[n], where


x[n] = 1 for 3 ≤ n ≤ 8
0 otherwise

h[n] = 1 for 4 ≤ n ≤ 15
0 otherwise

Solution:

www.EngrCS.com, ik Signals and Systems page 17


a)

x[n]

n
3 4 5 6 7 8
h[n]

n
4 5 6 7 8 9 10 15


y[ n] = x[ n] * h[n] = ∑ x[k ]h[n − k ]
k = −∞
Let’s take the graphic path to find out the intervals of interest --- first need to plot h[n-k]

x[k]

k
3 8
h[n-k]

k
(n-15) (n-4)

Look for segment where the signal have the same amount of overlap
for 3 ≤ n-4 ≤ 8 or 7 ≤ n ≤ 12 y[n] = n -6 Partial front-end overlap
for 8 ≤ n-4 & n-15 ≤ 3 or 12 ≤ n ≤ 18 y[n] = 6 Full overlap
for 3 ≤ n-15 ≤ 8 or 18 ≤ n ≤ 23 y[n] = 24 - n Partial back-end overlap
otherwise y[n] = 0 or no overlap.

3U. Compute and plot y[n] = x[n] * h[n], where


x[n] = 1 for 2≤n≤5
0 otherwise

h[n] = 1 for −3≤ n ≤ 8


0 otherwise

Solution:

4S. Compute and plot the convolution y[n] = x[n] * h[n], where
−n
1
x[ n] =   u[ −n − 1] and h[ n] = u[ n − 1]
 3

www.EngrCS.com, ik Signals and Systems page 18


Solution:

−k

1 ∞
y[ n] = x[ n] * h[n] = ∑ x[ k ]h[ n − k ] = ∑   u[ −k − 1]u[ n − k − 1]
k = −∞ k = −∞  3 
u[-k-1] = 1 for –k-1 ≥0 k≤-1
0 otherwise
Condition 1 – when k≤-1 the summation is not 0 & the equation for y[n] can be writer as:

−k
−1
1
y[n ] = ∑   u[ n − k − 1]
k = −∞ 3 

u[n-k-1] = 1 for n–k-1 ≥0 k ≤ n-1


0 otherwise
Condition 2 – when k≤-n-1 the summation is not 0 & the equation for y[n] can be writer as:

−k
n −1
1
y[n ] = ∑  
k = −∞ 3 

Considering Condition 1 and 2, we have two intervals for output:


−k
−1
1
For n≥0 the above equation reduces y[n ] = ∑  
k = −∞ 3 
−k
1
n −1
For n<0 the above equation reduces y[n ] = ∑  
k = −∞ 3 

These equations can be further simplified by applying the Infinite Geometric Sum Series.

4U. Compute and plot the convolution y[n] = x[n] * h[n], where
−n
1
x[ n] =   u[ n + 1] and h[ n] = u[ n − 5]
8
Solution:

5S. Compute the convolution y[n] = x[n] * h[n] of the following pairs of signals:
x[n] = α n u[n] 1
n n
x[n] = (− ) n u[n − 4]
a) h[n] = β u[n] b) x[ n] = h[ n] = α u[ n] c) 2
n
whenα ≠ β h[n] = 4 u[2 − n]

d) x[n] and h[n] are as in following figure

x[n] h[n]
1 1

n n
-1 0 1 2 3 4 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16

Hint: first draw x(t) and h(t). reflect h(t) about x=0 and then walk the signal to find the limits.

www.EngrCS.com, ik Signals and Systems page 19


Solution:

a)

y [ n ] = x[ n ] * h[ n ] = ∑ x[ k ]h[ n − k ]
k = −∞
n
y[ n ] = ∑a
k =0
k
β n−k
for n ≥ 0

y[ n ] = β n
∑ (a / β )
k =0
k
for n ≥ 0
n
1 − a n +1
or apply inifinit series ∑a
k =0
k
=
1− a
β −α n +1 n +1

y[ n ] =   u[ n ] for α ≠ β
 β −α 
b)
Note, this part is the same as “part a” except for the fact that a=β
n 
y[n] = α n ∑ (1) k u[n] = (n + 1)a n u[n]
 k =0 
c)

y [ n ] = x [ n ] * h[ n ] = ∑ x[k ]h[n − k ]
k = −∞
∞ ∞
∞ 3

y[n ] = ∑ ( −1 / 2) k ( 4) n− k =( 4) n ∑ ( −1 / 8) k =(4) n ∑ ( −1 / 8) k − ∑ ( −1 / 8) k  for n ≤ 6
k =4 k =4  k =0 k =0 
∞ ∞
∞ n −1

y[n ] = ∑ ( −1 / 2) k ( 4) n− k =( 4) n ∑ ( −1 / 8) k =(4) n ∑ ( −1 / 8) k − ∑ ( −1 / 8) k  for n > 6
k =n k =n  k =0 k =0 
n n +1 ∞
1− a 1
To Further Simplify apply inifinit series ∑ a = since a = −1 / 8 ⇒ ∑ a k =
k

k =0 1− a k =0 1− a
d)

x[n] h[n]
1 1

n n
-1 0 1 2 3 4 5 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16


y [ n ] = x[ n ] * h [ n ] = ∑ x [ k ]h[ n − k ]
k = −∞

y [ n ] = x[ 0 ]h[ n − 0 ] + x [1] h[ n − 1] + x[ 2 ] h[ n − 2 ] + x[ 3 ]h[ n − 3 ] + x [ 4 ] h[ n − 4 ]


y [ n ] = h[ n ] + h[ n − 1] + h [ n − 2 ] + h[ n − 3 ] + h[ n − 4 ]

www.EngrCS.com, ik Signals and Systems page 20


y[n] 5

1
n
0 1 2 3 4 5 6 7 8 9 11 13 15 17 19

5U. Compute the convolution y[n] = x[n] * h[n] of the following pairs of signals:
1
x[n] = ( ) n u[n + 2]
a) 3
h[n] = 8 n u[3 + n]

b)

x[n] h[n]
1 1
... ... ...
...
n n
-1 0 1 2 3 4 5 6 7 0 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17

6S. For each of the following pairs of waveforms, use the convolution integral to find response y(t) of
the LTI system with impulse response h(t) and x(t). Sketch your results.
x(t ) = e −αt u (t )
a) (Do this both when α ≠ β and α = β .)
h(t ) = e − βt u (t )

b) x(t) = u(t) – 2u(t – 2) + u(t – 5) and h(t) = e2tu(1 – t)

c) x(t) and h(t) shown below:

x(t) h(t)
1 2
One period of sinπt
t t
1 2 1 2 3

d) x(t) and h(t) shown below:

www.EngrCS.com, ik Signals and Systems page 21


x(t) h(t)
4/3
b Slope = a
t t
1 2

-1/3

e) x(t) and h(t) shown below:

x(t) h(t)

1 1

t t
-3 -2 -1 1 2 3 1

-1

Hint: first draw x(t) and h(t). reflect h(t) about x=0 and then walk the signal to find the limits.

Solutions:

a) The desired Convolution is:



y (t ) = ∫ x(τ )h(t −τ )dτ
−∞

y (t ) = 0(no overlap ) for t ≥ 0


t
y (t ) = ∫ e −ατ e − β ( t −τ ) dτ for t ≥ 0
0

Therefore
t 
y (t ) = e − βt  ∫ e ( β −α )τ dτ u (t )
0 
t 
for α = β ⇒ y (t ) = e − βt  ∫ dτ u (t 0 = te − βt u (t )
0 

for α ≠ β ⇒ y (t ) =
e e ( β −α ) t − 1
− βt
[ ]
β −α

b) The desired Convolution is:


First draw x(t) and h(-t) to find the number of unique overlap sections.

www.EngrCS.com, ik Signals and Systems page 22



y (t ) = ∫ x(τ )h(t −τ )dτ
−∞

Considering the intervals :


t − 1 < 0 → t < 1 → y (t ) = 0
t −1

∫e
2 ( t −τ )
1 ≤ t < 3 → y (t ) = dτ
0
2 t −1
3 ≤ t ≤ 6 → y (t ) = ∫ e 2 ( t −τ ) dτ − ∫ e 2( t −τ ) dτ
0 2

t>6 → y (t ) = 0 for 6 < t

Therefore

t < 1 → y (t ) = 0
t −1
1 1
1 ≤ t < 3 → y (t ) = e [ − e −2τ
2t
= − e 2t [e −2 ( t −1) − 1)
2 0 2
2 t −1
1 1 1 2t
for 3 ≤ t ≤ 6 y (t ) = e [ − e −2τ
2t
+ e − 2τ = e [ −( e −2 − 1) + ( e −2( t −1) − e − 2 )
2 0 2 2 2
t > 6 → y (t ) = 0

c) The desired Convolution is:


First draw x(t) and h(-t) to find the number of unique overlap sections.
∞ 2
y (t ) = ∫ x(τ )h(t −τ )dτ = ∫ sin(πt )h(t − τ )dτ
−∞ 0

Considering the overlap types :


y (t ) = 0 for t ≤ 1
y (t ) = (2 / π )[1 − cos{π (t − 1)}] for 1 < t < 3
y (t ) = (2 / π )[1 − cos{π (t − 3)} − 1] for 3 < t < 5
y (t ) = 0 for 5 < t

d) The desired Convolution is:

let h(t) = h1(t) – (1/3)δ(t-2) where h1(t)= 4/3 for 0≤t≤1


0 otherwise

so we can use the system linearity to write:


y(t) = h(t)*x(t)= h1(t)*x(t) –(1/3)x(t-2)

we have

www.EngrCS.com, ik Signals and Systems page 23


t
4 4 1 1 
∫ 3 (aτ + b)dτ = 3 ( 2 at
2
h1 (t ) * x (t ) = − a (t − 1) 2 + bt − b(t − 1)
t −1
2 
Therefore
4 1 2 1  1 1 
y (t ) =  ( at − a (t − 1) 2 + bt − b(t − 1) − ( a (t − 2) + b
3 2 2  3 3 

e) The desired Convolution is:


for LTI system when the input is periodic, it implies that output y(t) is also periodic so determine one period.
But remember to include effect of other periods on calculation of single period output.

So we will take the period -1/2 < t < 3/2


x(t)=1 for -1/2 < t < 1/2
0 for 1/2 < t < 3/2

h(t) = t+1 for 0 < t < 1


=0 otherwise

First draw x(t) and h(-t) to find the number of unique overlap sections.

y (t ) = ∫ x(τ )h(t −τ )dτ
−∞
−1 / 2 t

∫ (t − τ − 1)dτ − ∫ (t − τ − 1)dτ = −t
2
y (t ) = + t + 1/ 4 for − 1 / 2 < t < 1 / 2
t −1 −1 / 2
1/ 2 t
y (t ) = − ∫ (t − τ − 1)dτ + ∫ (t − τ − 1)dτ = t 2 − 3t + 7 / 4 for 1 / 2 < t < 3 / 2
t −1 1/ 2

Note: Output period is also 2.

6U. For each of the following pairs of waveforms, use the convolution integral to find response y(t) of
the LTI system with impulse response h(t) and x(t). Sketch your results.

a) x(t) = 3u(-t) – 5u(t + 2) + 3u(-t + 3) and h(t) = e-2tu(-2 + t)

b) x(t) and h(t) shown below:

x(t) h(t)
4
5 Slope = 4
t t
6 12

-4

Solutions:

7S. Compute the convolution y[n]=x[n]*h[n] of the following pair of signals:


x[n] = a n u[n − 4]
h[n] = b n u[n − 1] where 0 < a < b

www.EngrCS.com, ik Signals and Systems page 24


Solution:

x[k]
...
4
h[n-k]

...
n-1

x[n] = a n u[n − 4]
h[n] = b n u[n − 1]

y[n] = x[n] * h[n] = ∑ x[k ]h[n − k ]
k = −∞

y[n] = 0 for n − 1 < 4 ⇒ n < 5


n −1
y[n] = ∑ a k β n −k for n ≥ 5
k =4
n −5
y[n] = β n ∑ (a / β ) k for n ≥ 0
k =0

7U. Compute the convolution y[n]=x[n]*h[n] of the following pair of signals:


1
x[n] = ( ) n u[n − 2]
3
1
h[n] = ( ) n u[n + 3]
2
Solution:

8S. The following are the impulse response of discrete-time LTI systems. Determine whether each
system is causal and/or stable. Justify your answers.
a) h[n] = (1/5)n u[n]
b) h[n] = (0.8)n u[n + 2]
c) h[n] = (1/2)n u[-n]
d) h[n] = (5)n u[3 – n]
e) h[n] = (-1/2)n u[n] + (1.01)n u[n – 1]
f) h[n] = (-1/2)n u[n] + (1.01)n u[1 – n]
g) h[n] = n (1/3)n u[n – 1]

Solution:

Notes “System is causal if h[n]=0 for n<0 --- System is stable if ∑ | h[n ] |< ∞ ”
n = −∞
a) h[n] = (1/5)n u[n]
h[n]=0 for n<0 therefore system is causal

www.EngrCS.com, ik Signals and Systems page 25



1
∑ (1/5)
n =0
n
=
1 − (1 / 5)
= 5 / 4 < ∞ therefore the System is stable

b) h[n] = (0.8)n u[n + 2]


h[n]≠0 for n<0 therefore system is non-causal
∞ ∞
(0.8) -2
∑ (0.8)
n = −2
n
= ∑ (0.8) n -2 =
n =0 1 − ( 0.8)
< ∞ therefore the System is stable

c) h[n] = (1/2)n u[-n]


h[n]≠0 for n<0 therefore system is non-causal
0

∑ (1/2)
n = −∞
n
= ∞ therefore the System is not stable

d) h[n] = (5)n u[3 – n]


h[n]≠0 for n<0 therefore system is non-causal
3 ∞ 0
1
∑ (5)
n = −∞
n
= ∑ (1/5) n −
n =0
∑ (1/5)
n = −3
n
=
1 − (1 / 5)
− (1 / 5) −3 − (1 / 5) − 2 − (1 / 5) −1 − (1 / 5) 0 = 625 / 4 < ∞
therefore the System is stable

e) h[n] = (-1/2)n u[n] + (1.01)n u[n – 1]


h[n]=0 for n<0 therefore system is causal
∞ ∞
1
∑ | (-1/2)
n =0
n
| + ∑ (1.01) n =
n =1 1 − (1 / 2)
+ ∞ = ∞ therefore the System is not stable

f) h[n] = (-1/2)n u[n] + (1.01)n u[1 – n]


h[n]≠0 for n<0 therefore system is non-causal
∞ 1 ∞
1 1
∑ | (-1/2)
n =0
n
|+ ∑ (1.01)
n = −∞
n
= + ∑ (1.01) -n − 1.01 = 2 / 3 +
1 − (1 / 2) n = 0 1 − (1 / 1.01)
− 1.01 < ∞
therefore the System is stable

g) n (1/3)n u[n – 1]
h[n]=0 for n<0 therefore system is causal

1
∑ n(1/3)
n =1
n
=
1 − (1 / 3)
= 3 / 4 = 0.75 < ∞ therefore the System is stable

8U. The following are the impulse response of discrete-time LTI systems. Determine whether each
system is causal and/or stable. Justify your answers.
a) h[n] = (1/5)n u[-n]
b) h[n] = (0.8)n u[n - 2]
c) h[n] = (1/2)n u[n]
d) h[n] = (5)n u[3 + n]
e) h[n] = n (1/3)n u[n + 1]

Solution:

9S. The following are the impulse responses of continuous-time LTI systems. Determine whether each
system is causal and/or stable. Justify your answers.

a) h(t) = e -4t u(t – 2)


b) h(t) = e -6t u(3 – t)
c) h(t) = e -2t u(t + 50)
d) h(t) = e 2t u(–1 – t)
e) h(t) = e -6|t|
www.EngrCS.com, ik Signals and Systems page 26
f) h(t) = t e –t u(t)
g) h(t) = (2 e –t – e (t – 1000)/100) u(t)
−∞
Notes “System is causal if h(t)=0 for t<0 --- System is stable if ∫ | h(t ) | dt ”
−∞
Solution:

a) h(t) = e -4t u(t – 2)


h(t)=0 for t<0 therefore System is causal

∫| e
−4t
| dt = −1 / 4(e −∞ − e −8 ) = 1 / 4(e −8 ) < ∞ therefore System is stable
2
b) h(t) = e -6t u(3 – t)
h(t)≠0 for t<0 therefore System is not causal
3

∫| e
−6t
| dt = 1 / 6(e −18 − e ∞ ) = ∞ therefore System is unstable
−∞

c) h(t) = e -2t u(t + 50)


h(t)≠0 for t<0 therefore System is not causal

∫| e
− 2t
| dt = −1 / 2(e −∞ − e100 ) = 1 / 2e100 < ∞ therefore System is stable
−50

d) h(t) = e 2t u(–1 – t)
h(t)≠0 for t<0 therefore System is not causal
−1

∫| e | dt = 1 / 2(e − 2 − e −∞ ) = e − 2 / 2 < ∞ therefore System is stable


2t

−∞
e) h(t) = e -6|t|
h(t) ≠0 for t<0 therefore System is not causal

2∫ | e −6|t| | dt = −1 / 3(e −∞ − e 0 ) = 1 / 3 < ∞ therefore System is stable
0

f) h(t) = t e –t u(t)
h(t) =0 for t<0 therefore System is causal

∫ | te
−t
| dt = 1 < ∞ therefore System is stable (Use integral by part to solve)
0

g) h(t) = (2 e –t – e (t – 1000)/100) u(t)


h(t) =0 for t<0 therefore System is causal

∫ | (2 e
-t (t -1000)/100
-e ) | dt = ∞ therefore System is unstable
0
9U. The following are the impulse responses of continuous-time LTI systems. Determine whether each
system is causal and/or stable. Justify your answers.

a) h(t) = e -3t u(t + 4)


b) h(t) = e -5t u(13 – t)
c) h(t) = e -2t u(-t + 10)
d) h(t) = e 2t u(–3 – t)
e) h(t) = e-10|t|

Solution:

www.EngrCS.com, ik Signals and Systems page 27


10S. h[n] = (-1/2)n u[n] + (1.01)n u[n – 1] is the impulse of response of a system. Determine if the system is
Causal and/or Stable

Solution:
h[n] = (-1/2)n u[n] + (1.01)n u[n – 1]

Notes “System is causal if h[n]=0 for n<0 --- System is stable if ∑ | h[n] |< ∞ ”
n=∞
h[n]=0 for n<0 therefore system is causal
∞ ∞
1

n =0
| (-1/2) n
| + ∑
n =1
(1.01) n =
1 − (1 / 2)
+ ∞ = ∞ therefore the System is not stable

10U. h[n] = (-1/8)n u[n + 1] + (1.21)n u[n] is the impulse of response of a system. Determine if the system is
Causal and/or Stable

Solution:

www.EngrCS.com, ik Signals and Systems page 28

You might also like