Advanced Digital Systems Tutorial Sheet 4 Solutions S.

Nasser, 2022/2023

Tutorial Sheet 4 + Solutions

DTFT

1. The discrete-time system shown consists of one unit delay element and one scalar multiplier.
Write a difference equation that relates the output 𝑦[𝑛] and the input 𝑥[𝑛].
a)

Solution:
𝑦[𝑛 ] = 𝑎𝑦 [𝑛 − 1] + 𝑥 [𝑛]
𝑦[𝑛 ] − 𝑎𝑦 [𝑛 − 1] = 𝑥[𝑛]
b)

Solution:

Let the input to the unit delay element be q[nl.

1
𝑞 [𝑛] = 𝑞 [𝑛 − 1] + 𝑥 [𝑛] → (1)
2
𝑦 [𝑛] = 2𝑞 [𝑛] + 𝑞 [𝑛 − 1] → (2)
Solving (1) and (2) for q[n] and q[n-1] in terms of x[n] and y[n]:
4𝑞 [𝑛] = 2𝑥 [𝑛] + 𝑦 [𝑛] → (3)
2𝑞 [𝑛 − 1] = 𝑦 [𝑛] − 2𝑥[𝑛] → (4)
Changing n to n-1 in (3):
4𝑞 [𝑛 − 1] = 2𝑥 [𝑛 − 1] + 𝑦 [𝑛 − 1] → ( 5)
Multiplying (4) by 2 then equating (5) and (4):
2𝑦 [𝑛] − 4𝑥 [𝑛] = 2𝑥 [𝑛 − 1] + 𝑦[𝑛 − 1]

Page 1 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Rearranging the terms:

2𝑦 [𝑛] − 𝑦 [𝑛 − 1] = 2𝑥 [𝑛 − 1] + 4𝑥[𝑛]
Which is the required difference equation.

c)

Solution:
Let the input to the unit delay element be q[nl.
𝑞 [𝑛] = 2𝑞 [𝑛 − 1] + 𝑥[𝑛 ] → (1)
𝑦 [𝑛] = 𝑞 [𝑛] + 3𝑞 [𝑛 − 1] → (2)
Solving (1) and (2) for q[n] and q[n-1] in terms of x[n] and y[n]:
2 3
𝑞[𝑛 ] = 𝑦 [𝑛] + 𝑥 [𝑛] → (3)
5 5
1 1
𝑞[𝑛 − 1] = 𝑦 [𝑛] − 𝑥 [𝑛] → (4)
5 5
Changing n to n-1 in (3):
2 3
𝑞[𝑛 − 1] = 𝑦 [𝑛 − 1] + 𝑥 [𝑛 − 1] → (5)
5 5
Equating (5) and (4):
1 1 2 3
𝑦 [𝑛] − 𝑥 [𝑛] = 𝑦[𝑛 − 1] + 𝑥[𝑛 − 1]
5 5 5 5

Multiplying both sides by 5 and rearranging the terms:

𝑦 [𝑛] − 2𝑦 [𝑛 − 1] = 𝑥[𝑛] + 3𝑥[𝑛 − 1]
Which is the required difference equation.

d)

Page 2 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Solution:
In the figure above, the output of the first (from the right) unit delay element is y[n – 1] and the
output of the second (from the right) unit delay element is y[n – 2]. Thus, we
see that:
𝑦 [𝑛] = 𝑎1 𝑦[𝑛 − 1] + 𝑎2 𝑦[𝑛 − 2] + 𝑥[𝑛]
𝑦 [𝑛] − 𝑎1 𝑦[𝑛 − 1] − 𝑎2 𝑦[𝑛 − 2] = 𝑥[𝑛]
which is the required second-order linear difference equation.
Note that, in general, the order of a discrete-time LTI system consisting of the interconnection
of unit delay elements and scalar multipliers is equal to the number of unit delay elements
in the system.

2. Find the DTFT of the following sequences:

1 𝑛
a) 𝑥1[𝑛] = ( ) 𝑢 [𝑛 + 3]
2

Solution:

∞
𝑋( 𝑒𝑗𝜔 ) = ∑ 𝑥[𝑛 ] 𝑒 −𝑗𝜔𝑛
𝑛=−∞
∞ 𝑛
1
=∑ ( ) 𝑢 [𝑛 + 3]𝑒 −𝑗𝜔𝑛
𝑛=−∞ 2
But,
1
𝑛 ≥ −3
𝑢 [𝑛 + 3] = { }
0
𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
∞ 1 𝑛 −𝑗𝜔𝑛
∴= ∑ ( ) 𝑒
𝑛=−3 2
Let m = n+3, 𝑎𝑡 𝑛 = −3 → 𝑚 = 0 𝑎𝑛𝑑 𝑛 = 𝑚 − 3
∞ 1 𝑚−3 −𝑗𝜔(𝑚−3)
= ∑ ( ) 𝑒
𝑚=0 2
1 −𝑗𝜔 −3 ∞ 1 −𝑗𝜔 𝑚
=( 𝑒 ) ∑ ( 𝑒 )
2 𝑚=0 2
Using the closed form of the geometric series:

−3
1 1
= ( 𝑒 −𝑗𝜔 ) ( 1 )
2 1 − ( 𝑒 −𝑗𝜔 )
2
3𝑗𝜔
8𝑒
= 1
1 − 𝑒 −𝑗𝜔
2

b) 𝑥2 [𝑛] = 𝛼 𝑛 sin(𝜔0 𝑛) 𝑢 [𝑛]

Page 3 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Solution:
The best way to find the DTFT of x(n) is to express the sinusoid as a sum of two complex
exponentials as follows:
1
𝑥 [𝑛] = ( 𝛼 𝑛 𝑒𝑗𝜔0 𝑛 − 𝛼 𝑛 𝑒 −𝑗𝜔0𝑛 )𝑢[𝑛]
2𝑗
∞ 1
𝑋(𝑒𝑗𝜔 ) = ∑ (𝛼 𝑛 𝑒𝑗 𝜔0𝑛 − 𝛼 𝑛 𝑒 −𝑗𝜔0𝑛 )𝑢 [𝑛] 𝑒 −𝑗𝜔
𝑛=−∞ 2𝑗
∞ 1 ∞ 1
∑ ( 𝑛 𝑗𝜔 0 𝑛 ) −𝑗𝜔 ∑ (𝛼 𝑛 𝑒 −𝑗𝜔0𝑛 )𝑢[𝑛] 𝑒 −𝑗𝜔
= 𝛼 𝑒 𝑢[𝑛] 𝑒 −
𝑛=−∞ 2𝑗 𝑛=−∞ 2𝑗

But,
1 𝑛≥0
𝑢 [𝑛 ] = { }
0
𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒
∞ 1 ∞ 1
∴= ∑ (𝛼 𝑛 𝑒𝑗𝜔0 𝑛 ) 𝑒 −𝑗𝜔 − ∑ ( 𝛼 𝑛 𝑒 −𝑗𝜔0𝑛 ) 𝑒 −𝑗𝜔
𝑛=0 2𝑗 𝑛=0 2𝑗
The DTFT of the first term is:
∞ 1
=∑ (𝛼 𝑛 𝑒𝑗 𝜔0𝑛 ) 𝑒 −𝑗𝜔
𝑛=0 2𝑗
1 ∞ 𝑛
= ∑ (𝛼𝑒 −𝑗(𝜔−𝜔0) )
2𝑗 𝑛=0
Using the closed form of the geometric series:
1 1
= ( )
2𝑗 1 − 𝛼𝑒 −𝑗 (𝜔−𝜔0)
Similarly, for the second term we have:
1 ∞
= − ∑ (𝛼 𝑛 𝑒 −𝑗𝜔0 𝑛 ) 𝑒 −𝑗𝜔
2𝑗 𝑛=0
1 ∞ 𝑛
= − ∑ (𝛼𝑒 −𝑗 (𝜔+𝜔0 ) )
2𝑗 𝑛=0
1 1
=− ( )
2𝑗 1 − 𝛼𝑒 −𝑗 (𝜔+𝜔0 )
Therefore,
1 1 1
𝑋(𝑒𝑗𝜔 ) = (( ) (
(𝜔 −𝜔 0 ) −
))
2𝑗 1 − 𝛼𝑒 −𝑗 1 − 𝛼𝑒 (𝜔+𝜔0)
−𝑗

(𝛼 sin 𝜔0 )𝑒 −𝑗𝜔
=
1 − (2𝛼 cos 𝜔0 )𝑒 −𝑗𝜔 + 𝛼 2 𝑒 −2𝑗𝜔

c) 𝑥3 [𝑛] = 𝛿 [𝑛 − 1] + 𝛿 [𝑛 + 1]

Solution:
1 𝑛=1
𝛿[𝑛 − 1] = { }
0 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

Page 4 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

∞
∴ 𝑋1 (𝑒𝑗𝜔 ) = ∑ 𝑥[𝑛] 𝑒 −𝑗𝜔𝑛
𝑛=−∞
But 𝛿[𝑛 − 1] exists only at n=1 and is equal to 1 at that instant:
∴ 𝑋1 (𝑒𝑗𝜔 ) = 𝑒 −𝑗𝜔(1) = 𝑒 −𝑗𝜔
Similarly for 𝛿[𝑛 + 1] :
𝑋2 (𝑒𝑗𝜔 ) = 𝑒 −𝑗𝜔( −1) = 𝑒𝑗𝜔
𝑋 (𝑒𝑗𝜔 ) = 𝑋1(𝑒𝑗𝜔 ) + 𝑋2 (𝑒𝑗𝜔 )
= 𝑒 −𝑗𝜔 + 𝑒𝑗𝜔
𝑒𝑗𝜔 +𝑒 −𝑗𝜔
From Euler’s rule: cos(𝜔) = , we see that:
2
𝑋 (𝑒𝑗𝜔 ) = 2 cos 𝜔

d) 𝑥4[𝑛] = 𝛿 [𝑛 + 2] − 𝛿 [𝑛 − 2]

Solution:
Similar to part (c):
𝑋(𝑒𝑗𝜔 ) = 𝑒 −𝑗2𝜔 − 𝑒 −𝑗2𝜔
𝑒𝑗𝜔 −𝑒 −𝑗𝜔
From Euler’s rule: sin(𝜔) = , we see that:
2𝑗
𝑋(𝑒𝑗𝜔 ) = 2𝑗 sin (2𝜔)

1 𝑛−1
e) 𝑥5[𝑛] = ( ) 𝑢 [𝑛 − 1]
2

Solution:
∞ ∞ 1 𝑛−1
=∑ 𝑥[𝑛] 𝑒 −𝑗𝜔𝑛 = ∑ ( ) 𝑢 [𝑛 − 1]𝑒 −𝑗𝜔𝑛
𝑛=−∞ 𝑛=−∞ 2

But:
1 𝑛≥1
𝑢 [𝑛 − 1] = { }
0 𝑒𝑙𝑠𝑒𝑤ℎ𝑒𝑟𝑒

1 𝑛−1 𝑗𝜔𝑛 ∞
∴ 𝑋(𝑒𝑗𝜔 ) = ∑
( ) 𝑒
𝑛=1 2
let 𝑚 = 𝑛 − 1. 𝑎𝑡 𝑛 = 1 → 𝑚 = 0 𝑎𝑛𝑑 𝑛 = 𝑚 + 1.
Rewrite the summation:
( )
1 𝑚+1 −1 −𝑗𝜔(𝑚+1)
∞
=∑ ( ) 𝑒
𝑚=0 2

−𝑗𝜔 ∑
∞ 1 𝑚 −𝑗𝜔𝑚
=𝑒 ( ) 𝑒
𝑚=0 2

−𝑗𝜔 ∑
∞ 1 −𝑗𝜔 𝑚
=𝑒 ( 𝑒 )
𝑚=0 2

Page 5 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Using the closed form of the geometric series:

1
= (𝑒 −𝑗𝜔 ) ( 1 )
1 − ( ) 𝑒 −𝑗𝜔
2

1 |𝑛−1|
f) 𝑥[𝑛 ] = ( )
2

Solution:
1 𝑛−1
( ) 𝑛≥1
𝑥[𝑛] = 2
1 −(𝑛−1)
( ) 𝑛≤0
{ 2 }
∞
𝑋(𝑒𝑗𝜔 ) = ∑ 𝑥[𝑛] 𝑒 −𝑗𝜔
𝑛=−∞
0 −(𝑛−1 ) ∞ 𝑛−1
1 −𝑗𝜔𝑛
1
=∑ ( ) 𝑒 +∑ ( ) 𝑒 −𝑗𝜔𝑛
𝑛=−∞2 𝑛=1 2
The second summation in the right-hand side of the equation is exactly the same as the result of
part (e). Now,
0 1 −𝑛+1 −𝑗𝜔𝑛
=∑ ( ) 𝑒
𝑛=−∞ 2
Replacing n by -n and exchanging the limits of the summation:
∞ 1 𝑛+1 −𝑗𝜔𝑛
=∑ ( ) 𝑒
𝑛=0 2
1 ∞ 1 𝑛
= ( )∑ ( ) 𝑒 −𝑗𝜔𝑛
2 𝑛=0 2
1 ∞ 1 −𝑗𝜔 𝑛
= ( )∑ ( 𝑒 )
2 𝑛=0 2
Using the closed form of the geometric series:
1 1
= ( )( )
2 1 − (1 𝑒 −𝑗𝜔 )
2
1 1 1
∴ 𝑋(𝑒𝑗𝜔 ) = ( ) ( ) + ( 𝑒 −𝑗𝜔 ) ( )
2 1 − (1 𝑒 −𝑗𝜔 ) 1
1 − ( ) 𝑒 −𝑗𝜔
2 2
−𝑗𝜔
0.75𝑒
=
1.25 − cos(𝜔)

3. Consider the following sequences. Evaluate the DTFT, then sketch it.
a) 𝑥[𝑛 ] = 𝑎 |𝑛| , | 𝑎| < 1

Page 6 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Solution:
𝑎𝑛 𝑛≥0
𝑥 [𝑛] = { }
𝑎 −𝑛 𝑛 ≤ −1
∞ −1
𝑋(𝑒𝑗𝜔 ) = ∑ 𝑛 −𝑗𝜔𝑛
𝑎 𝑒 +∑ 𝑎 −𝑛 𝑒 −𝑗𝜔𝑛 → (1)
𝑛=0 𝑛=−∞
The second term:
Replacing n by -n:
∞
=∑ 𝑎𝑛 𝑒𝑗𝜔𝑛
𝑛=1
Let 𝑚 = 𝑛 − 1, 𝑎𝑡 𝑛 = 1 → 𝑚 = 0 𝑎𝑛𝑑 𝑛 = 𝑚 + 1. Rewriting:
∞
=∑ 𝑎𝑚+1 𝑒 −𝑗𝜔(𝑚+1)
𝑚=0
∞
= 𝑎𝑒 −𝑗𝜔 ∑ 𝑎𝑚 𝑒 −𝑗𝜔𝑚
𝑚=0
Back in (1):
∞ ∞
𝑋(𝑒𝑗𝜔 ) = ∑ 𝑎𝑛 𝑒 −𝑗𝜔𝑛 + 𝑎𝑒 −𝑗𝜔 ∑ 𝑎𝑚 𝑒 −𝑗𝜔𝑚
𝑛=0 𝑚=0
Both of these summations are infinite geometric series that we can evaluate in closed
form, yielding:

Signal 𝑥[𝑛] (a) and its DTFT (b) sketched for 0 < 𝑎 < 1

1 |𝑛| ≤ 𝑁1
b) 𝑥[𝑛 ] = { }, consider 𝑁1 = 2.
0 |𝑛| > 𝑁1

Solution:

Page 7 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

𝑁1
𝑋(𝑒𝑗𝜔 ) = ∑ 𝑒 −𝑗𝜔𝑛
𝑛=−𝑁1
Let 𝑚 = 𝑛 + 𝑁1 , 𝑎𝑡 𝑛 = −𝑁1 → 𝑚 = 0; 𝑎𝑡 𝑛 = 𝑁1 → 𝑚 = 2𝑁1 𝑎𝑛𝑑 𝑛 = 𝑚 − 𝑁1
2𝑁1 2𝑁1
=∑ 𝑒 −𝑗𝜔(𝑚−𝑁1 ) = 𝑒𝑗𝜔 𝑁1 ∑ 𝑒 −𝑗𝜔𝑚
𝑚=0 𝑚=0
Using the closed form of this geo. Series:
𝑗𝜔 𝑁1 (
1 − 𝑒 −𝑗𝜔(2𝑁1 +1)
=𝑒 )
1 − 𝑒 −𝑗𝜔
𝑒𝑗𝜔 𝑁1 − 𝑒 −𝑗𝜔(𝑁1 +1)
=
1 − 𝑒 −𝑗𝜔
𝜔 1 1
−𝑗
𝑒 2 (𝑒 𝑗𝜔(𝑁1 +2) − 𝑒 −𝑗𝜔(𝑁1 +2) )
= 𝜔 𝜔 𝜔
−𝑗 𝑗 −𝑗
𝑒 2 (𝑒 2 −𝑒 2)
𝜔 𝜔
𝑗 𝑁1 𝑗 𝑁1
(because 𝑒𝑗𝜔 𝑁1 = 𝑒 2 𝑒 2 𝑎𝑛𝑑 𝑒𝑗𝜔(𝑁1 +1) = 𝑒𝑗𝜔 𝑁1 𝑒𝑗𝜔 ).
1
2𝑗 sin (𝜔 (𝑁1 + ))
2
= 𝜔
2𝑗 sin ( )
2
Consider 𝑁1 = 2

(a) Rectangular pulse signal for 𝑁1 = 2 and

(b) its Fourier transform.

c) 𝑥[𝑛 ] = sin [𝜔0 𝑛]

Page 8 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

Solution:
1
𝑥[𝑛] = sin[𝜔0 𝑛] = (𝑒𝑗 𝜔0𝑛 − 𝑒 −𝑗𝜔0𝑛 )
2𝑗
1
𝑋(𝑒𝑗𝜔 ) = (𝐷𝑇𝐹𝑇 {𝑒𝑗 𝜔0𝑛 } − 𝐷𝑇𝐹𝑇 {𝑒 −𝑗𝜔0𝑛 })
2𝑗

1
= (2𝜋𝛿[𝜔 − 𝜔0 ] − 2𝜋𝛿[𝜔 + 𝜔0 ])
2𝑗
𝑋(𝑒𝑗𝜔 ) = −𝑗( 𝜋𝛿[𝜔 − 𝜔0 ] − 𝜋 [𝜔 + 𝜔0 ]) , −𝜋 < 𝜔 < 𝜋
and X(𝑒𝑗𝜔 ) repeats periodically with a period of 2𝜋,

Sketch of |𝑋 (𝑒𝑗𝜔 )|

4. Let 𝑥[𝑛] be the sequence:

𝑥[𝑛] = 2𝛿[𝑛 + 2] − 𝛿 [𝑛 + 1] + 3𝛿 [𝑛] − 𝛿 [𝑛 − 1] + 2𝛿 [𝑛 − 2]
Evaluate the 𝑋(𝑒𝑗𝜔 )|𝜔=0 without explicitly finding (𝑋 (𝑒𝑗𝜔 )).
Solution:

Page 9 of 10
Advanced Digital Systems Tutorial Sheet 4 Solutions S. Nasser, 2022/2023

5. Let 𝑥[𝑛] be a sequence with a DTFT 𝑋 (𝑒𝑗𝜔 ). For each of the following sequences that are
related to 𝑥[𝑛], express the DTFT in terms of 𝑋(𝑒𝑗𝜔 ).

a) 𝑥[𝑛 ] − 𝑥[𝑛 − 2]
b)
Solution:
= 𝑋 (𝑒𝑗𝜔 ) − 𝑒 −2𝑗𝜔 𝑋(𝑒𝑗𝜔 )
= (1 − 𝑒 −2𝑗𝜔 )𝑋(𝑒𝑗𝜔 )

c) 𝑥[𝑛 ] ∗ 𝑥[𝑛 − 1]
d)
Solution:
= 𝑋(𝑒𝑗𝜔 )(𝑒 −𝑗𝜔 𝑋(𝑒𝑗𝜔 )
= 𝑒 −𝑗𝜔 𝑋 2 (𝑒𝑗𝜔 )
6. Let x[n] and h[n] be signals with the following Fourier transforms:

Determine 𝑦 [𝑛] = 𝑥[𝑛 ] ∗ ℎ [𝑛] .

Solution:
𝑌 (𝑒𝑗𝜔 ) = 𝑋(𝑒𝑗𝜔 )𝐻 (𝑒𝑗𝜔 )
= −3𝑒 −2𝑗𝜔 − 𝑒𝑗𝜔 + 1 − 2𝑒 −𝑗2𝜔 + 6𝑒 −𝑗𝜔 + 2𝑒 −2𝑗𝜔 − 2𝑒 −𝑗3𝜔 + 4𝑒 −𝑗5𝜔 + 3𝑒𝑗5𝜔 + 𝑒𝑗4𝜔
− 𝑒𝑗3𝜔 + 2𝑒𝑗𝜔
∴
𝑦 [𝑛] = 3𝛿 [𝑛 + 5] + 𝛿[𝑛 + 4] − 𝛿[𝑛 + 3] − 3𝛿 [𝑛 + 2] + 𝛿 [𝑛 + 1] + 𝛿 [𝑛] + 6𝛿 [𝑛 − 1]
− 2𝛿 [𝑛 − 3] + 4𝛿[𝑛 − 5]
e

Page 10 of 10