(HW6) Solution

Download as pdf or txt
Download as pdf or txt
You are on page 1of 7

1.

For the discrete-time signal shown in Figure 1, sketch each of the following:

(a) 𝑥(3𝑛 − 4)

(b) 𝑥(2 − 𝑛) + 𝑥(3𝑛 − 4)

2. Determine whether each of the following signals is periodic, and if it is, find its period.
𝜋𝑛 𝜋
(a) 𝑥(𝑛) = sin ( + )
4 8
𝜋
Ω0 4 1
= = is a rational number, the given signal is periodic;
2𝜋 2𝜋 8

N=8
3𝜋𝑛 𝜋
(b) 𝑥(𝑛) = sin ( ) + cos ( 𝑛)
4 3
3𝜋 𝜋
Ω01 4 3 Ω02 3 1
= = , = =
2𝜋 2𝜋 8 2𝜋 2𝜋 6

3 1
The provided signal is periodic because and are the rational numbers.
8 6

𝑁1 = 8, 𝑁2 = 6

N = Lowest common multiple of 𝑁1 and 𝑁2 = 24

𝜋
3. The signal 𝑥(𝑡) = 5cos (120𝑡 − ) is sampled to yield uniformly spaced samples T seconds apart.
3
What values of 𝑇 cause the resulting discrete-time sequence to be periodic? What is the period?

𝜋
𝑥(𝑛) = 5cos (120𝑛𝑇 − )
3
2𝜋𝑚 𝜋𝑚
Ω0 = 120𝑇, 𝑁 = =
120𝑇 60𝑇

𝜋𝑚
Therefore, the period of the signal is equal to 𝑁 =
60𝑇

4. Repeat Problem 3 if 𝑥(𝑡) = 3sin (100𝜋𝑡) + 4cos (120𝑡)

𝑥(𝑛) = 3sin (100𝜋𝑛𝑇) + 4cos (120𝑛𝑇)

Ω1 = 100𝜋𝑇, Ω2 = 120𝑇
Ω1 100𝜋
=
Ω2 120

It is clear that there is 𝜋 terms present in the ration which is irrational terms it means that the ration of
the whole term is irrational. Therefore, the defined function is aperiodic in nature

It is known that if 𝑥(𝑛) is the sum of two periodic sequences with periods 𝑁1 and 𝑁2 , then period of the
entire sequence is 𝑁 = 𝑝𝑁1 = 𝑞𝑁2 where p and q are integers.

5. Such concepts from continuous-time systems as memory, time invariance, linearity, and causality carry
over to discrete-time systems. In the following, x(n) refers to the input to a system and y(n) refers to
output. Determine whether the systems are (i) linear, (ii) memoryless, (iii) shift invariant, and (iv) causal.
Justify your answer in each case.

(a) 𝑦(𝑛) = log [𝑥(𝑛)]

- non-linear system

𝑦3 (𝑛) = 𝑎𝑦1 (𝑛) + 𝑏𝑦2 (𝑛) = 𝑎𝑙𝑜𝑔[𝑥1 (𝑛)] + 𝑏𝑙𝑜𝑔[𝑥2 (𝑛)]

𝑥3 = 𝑎𝑥1 (𝑛) + 𝑏𝑥2 (𝑛)

𝑦3 (𝑛) = log[𝑎𝑥1 (𝑛) + 𝑏𝑥2 (𝑛)] ≠ 𝑎𝑦1 (𝑛) + 𝑏𝑦2 (𝑛)


- Memoryless system

The value of y(n) only depends on the present variable.

- Shifts invariant.

𝑥2 (𝑛) = 𝑥1 (𝑛 − 𝑛0 )

𝑦2 (𝑛) = log[𝑥2 (𝑛)] = log[𝑥1 (𝑛 − 𝑛0 )] = 𝑦1 (𝑛 − 𝑛0 )

- Causal system

The value of the output depends upon the future value of the input.

(b) 𝑦(𝑛) = 𝑥(𝑛) + 2𝑥(𝑛 − 1)

- Linear system

𝑦3 (𝑛) = 𝑎𝑦1 (𝑛) + 𝑏𝑦2 (𝑛) = 𝑎𝑥1 (𝑛) + 𝑏𝑥2 (𝑛) + 2𝑎𝑥1 (𝑛 − 1) + 2𝑏𝑥2 (𝑛 − 1)

𝑥3 (𝑛) = 𝑎𝑥1 (𝑛) + 𝑏𝑥2 (𝑛)

𝑥3 (𝑛 − 1) = 𝑎𝑥1 (𝑛 − 1) + 𝑏𝑥2 (𝑛 − 1)

𝑦3 (𝑛) = 𝑥3 (𝑛) + 2𝑥3 (𝑛 − 1) = 𝑎𝑦1 (𝑛) + 𝑏𝑦2 (𝑛)

- Not memoryless System

The value of the output depends on the past information(x(n-1))

- Shift invariant

𝑦1 (𝑛) = 𝑥1 (𝑛) + 2𝑥1 (𝑛 − 1)

𝑥2 (𝑛) = 𝑥1 (𝑛 − 𝑛0 )

𝑦2 (𝑛) = 𝑥1 (𝑛 − 𝑛0 ) + 2𝑥1 (𝑛 − 𝑛0 − 1) = 𝑦1 (𝑛 − 𝑛0 )

- Causal system

The value of output of the system 𝑦(𝑛) at any instant 𝑛0 depends on value of the present input x(n) as
well as past value of input that is x(n-1).

6. Find the convolution 𝑦(𝑛) = ℎ(𝑛) ∗ 𝑥(𝑛) for each of the following pairs of finite sequences:

1 1 1 1
(a) 𝑥(𝑛) = {1, − , , − , }, h(n)={1,-1,1,-1}
2 4 8 16
(b) Plot the resulting y(n)

1 1 1 1
7. Consider the periodic convolution of the sequences 𝑥(𝑛) = {3, , − , 1,4} and ℎ(𝑛) = {2, −1, ,− }
2 4 2 2

(5-point periodic convolution)

If the two sequences are not of the same length, we can still define their convolution by augmenting the
shorter sequence with zeros to make the two sequences the same length

Perform circular shift for x(n) to shift its origin to extreme left
1 1
𝑥(𝑛) = {− , 1,4,3, }
4 2

Zero-pad the h(n) to match the length of h(n) with x(n)


1 1
ℎ(𝑛) = {2, −1, , − , 0}
2 2
3 53 21 1
𝑦𝑝 (𝑛) = {− , 1, , , − }
2 8 8 2

8. Determine the characteristic function and the homogeneous solution of the following difference
equation:

1 1 1
𝑦(𝑛) − 𝑦(𝑛 − 1) − 𝑦(𝑛 − 2) = 𝑥(𝑛) + 𝑥(𝑛 − 1), 𝑛 ≥ 0
6 6 2
𝑦(−1) = 1, 𝑦(−2) = 0

1 1
1 − 𝛼 −1 − 𝛼 −2 = 0
6 6
1 1
𝛼2 − 𝛼 − = 0
6 6

1 1 2 1
−(− ) ± √(− ) − 4 ∙
6 6 6 1 1
𝛼= = ,−
2 2 3
1 𝑛 1 𝑛
𝑦ℎ (𝑛) = 𝐴1 (− ) + 𝐴2 ( )
3 2

𝑦(−1) = 1, 𝑦(−2) = 0

𝑦ℎ (−1) = −3𝐴1 + 2𝐴2 = 1

𝑦ℎ (−2) = 9𝐴2 + 4𝐴2 = 0

2 1 𝑛 3 1 𝑛
𝑦ℎ (𝑛) = − (− ) + ( )
15 3 10 2
9. Find the total solution to the following difference equation:
8 1
𝑦(𝑛) + 𝑦(𝑛 − 1) + 𝑦(𝑛 − 2) = 𝑥(𝑛), 𝑛≥0
15 15
1 𝑛
𝑖𝑓 𝑦(−1) = 𝑦(−2) = 1 𝑎𝑛𝑑 𝑥(𝑛) = ( )
3

Characteristic equation
8 −1 1 −2
1+ 𝛼 + 𝛼 =0
15 15
8 1
𝛼2 + 𝛼 + =0
15 15
1 1
𝛼1 = − , 𝛼2 = −
5 3
1 𝑛 1 𝑛
𝑦ℎ (𝑛) = 𝐴 (− ) + 𝐵 (− )
5 3

Particular solution

1 𝑛
𝑦𝑝 (𝑛) = 𝐶 ( )
3
1 𝑛−1 1 𝑛−2
𝑦𝑝 (𝑛 − 1) = 𝐶 ( ) , 𝑦𝑝 (𝑛 − 2) = 𝐶 ( )
3 3
1 𝑛 8 1 𝑛−1 1 1 𝑛−2 1 𝑛
𝐶( ) + 𝐶( ) + 𝐶( ) =( )
3 15 3 15 3 3
1 𝑛 13 3 1 𝑛
𝐶( ) ( + ) = ( )
3 5 5 3
5
𝐶=
16
The complete solution of difference equation is

1 𝑛 1 𝑛 5 1 𝑛
𝑦(𝑛) = 𝐴 (− ) + 𝐵 (− ) + ( )
5 3 16 3

Use the initial condition

𝑦(−1) = 𝑦(−2) = 1
15
𝑦(−1) = −5𝐴 − 3𝐵 + =1
16
45
𝑦(−2) = 25𝐴 + 9𝐵 + =1
16
13 1
𝐴=− ,𝐵 =
80 4
13 1 𝑛 1 1 𝑛 5 1 𝑛
𝑦(𝑛) = − (− ) + (− ) + ( )
80 5 4 3 16 3

10. Find the impulse response of the following system


3 1
𝑦(𝑛) − 𝑦(𝑛 − 1) + 𝑦(𝑛 − 2) = 𝑥(𝑛), 𝑛 ≥ 0
2 2
𝑦(−1) = 0, 𝑦(−2) = 0

For homogeneous solution: 𝑛 ≥ 1


3 1
𝛼2 − 𝛼 + = 0
2 2
1
𝛼1,2 = ,1
2
1 𝑛
ℎ(𝑛) = 𝐴1 ( ) + 𝐴2
2

For homogeneous solution for n=0 and n=1:

𝑦(0) = 𝐴1 + 𝐴2 = 1
3 3 1
𝑦(1) = 𝛿(1) + 𝑦(0) = = 𝐴1 + 𝐴2
2 2 2
𝐴 = −1, 𝐵=2

1 𝑛
ℎ(𝑛) = (− ( ) + 2) 𝑢(𝑛)
2

You might also like