The Continuous-Time Fourier Transform

(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 Continuous Fourier Transform
Recall square wave 2 sin 𝑘𝜔0 𝑇1 𝑇 = 4𝑇1
𝑎𝑘 =
𝑘𝜔0 𝑇

2 sin 𝜔𝑇1
 𝑇𝑎𝑘 : Samples of an envelope function 𝑓 𝜔 =
𝜔 𝑇 = 8𝑇1
2 sin 𝜔𝑇1
𝑇𝑎𝑘 = ቚ
𝜔 𝜔=𝑘𝜔0

 𝑇 ↑, 𝜔0 ↓ ⇒ the envelope is sampled with 𝑇 = 16𝑇1

closer spacing
2 sin 𝜔𝑇1
 𝑇 → ∞, ⇒ 𝑇𝑎𝑘 → the envelope
𝜔 1
 Continuous Fourier Transform
Development of FT
 Consider a signal of finite duration, 𝑥 𝑡 = 0 𝑖𝑓 𝑡 > 𝑇1

 Periodic extension of 𝑥 𝑡 with T

 FS representation of 𝑥෤ 𝑡 𝑥(𝑡)
෤ = ෍ 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡
𝑘=−∞
𝑇/2
1
𝑎𝑘 = න 𝑥(𝑡)෤ 𝑒 −𝑗𝑘𝜔0 𝑡 𝑑𝑡
𝑇 −𝑇/2 2
 Continuous Fourier Transform
Development of FT
 FS coefficients of 𝑥෤ 𝑡
1 𝑇/2
𝑎𝑘 = න 𝑥(𝑡)෤ 𝑒 −𝑗𝑘𝜔0 𝑡 𝑑𝑡 ∞
𝑇 −𝑇/2 Define 𝑋(𝑗𝜔) = න 𝑥(𝑡) 𝑒 −𝑗𝜔𝑡 𝑑𝑡
−∞
𝑇/2 ∞
1 −𝑗𝑘𝜔0 𝑡
1 −𝑗𝑘𝜔0 𝑡
1
= න 𝑥(𝑡) 𝑒 𝑑𝑡 = න 𝑥(𝑡) 𝑒 𝑑𝑡 = 𝑋(𝑗𝑘𝜔0 )
𝑇 −𝑇/2 𝑇 −∞ 𝑇
 FS of 𝑥෤ 𝑡 ∞ ∞
1 𝑗𝑘𝜔 𝑡
1
𝑥෤ 𝑡 = ෍ 𝑋(𝑗𝑘𝜔0 ) 𝑒 0 = ෍ 𝑋(𝑗𝑘𝜔0 ) 𝑒 𝑗𝑘𝜔0 𝑡 𝜔0
𝑇 2𝜋
𝑘=−∞ 𝑘=−∞

 𝑇 → ∞, 𝑥෤ 𝑡 →𝑥 𝑡
1 ∞
𝑥(𝑡) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞ 3
 Continuous Fourier Transform
FT pairs
∞
𝑋(𝑗𝜔) = න 𝑥(𝑡) 𝑒 −𝑗𝜔𝑡 𝑑𝑡 Fourier transform (FT)
−∞

1 ∞
𝑥(𝑡) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔 Inverse Fourier transform
2𝜋 −∞

 𝑥(𝑡) is a linear combination (specifically, an integral) of sinusoidal

signals at different frequencies
 𝑋(𝑗𝜔)(𝑑𝜔/2𝜋) is the weight for different frequencies

 𝑋(𝑗𝜔) is called the spectrum

4
 Continuous Fourier Transform
FT vs. FS
Fourier transform (FT) Fourier series (FS)

∞
1 ∞ 𝑥෤ 𝑡 = ෍ 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡
𝑥(𝑡) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞ 𝑘=−∞

∞ 1 𝑇/2
𝑋(𝑗𝜔) = න 𝑥(𝑡) 𝑒 −𝑗𝜔𝑡 𝑑𝑡 𝑎𝑘 = න 𝑥෤ 𝑡 𝑒 −𝑗𝑘𝜔0 𝑡 𝑑𝑡
𝑇 −𝑇/2
−∞
1
𝑎𝑘 = 𝑋(𝑗𝜔) with 𝜔 = 𝑘𝜔0
𝑇
5
 Continuous Fourier Transform
Convergence of FT
 Condition 1: Finite energy condition
∞
න |𝑥(𝑡)|2 𝑑𝑡 < ∞
−∞
 Condition 2: Dirichlet condition
∞
(1) Absolutely integrable ‫׬‬−∞ |𝑥(𝑡)|𝑑𝑡 <∞
(2) Finite maxima and minima in one period with in any finite interval
(3) Finite number of finite discontinuities in any finite interval

6
 Continuous Fourier Transform
Examples
Consider the signal 𝑥 𝑡 = 𝑒 −𝑎𝑡 𝑢 𝑡 , 𝑎 > 0
Determine its FT
∞
𝑋 𝑗𝜔 = න 𝑒 −𝑎𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡
0
1 − 𝑎+𝑗𝜔 𝑡
∞
=− 𝑒 ቚ
𝑎 + 𝑗𝜔 0
1
= ,𝑎 > 0
𝑎 + 𝑗𝜔
1 𝜔
𝑋 𝑗𝜔 = ∢𝑋 𝑗𝜔 = − tan−1
2
𝑎 +𝜔 2 𝑎 7
 Continuous Fourier Transform
Examples
𝑥 𝑡 = 𝑒 −𝑎|𝑡| , 𝑎 > 0 𝑋 𝑗𝜔 =?
Solution
∞
𝑋 𝑗𝜔 = න 𝑒 −𝑎|𝑡| 𝑒 −𝑗𝜔𝑡 𝑑𝑡
−∞
0 ∞
= න 𝑒 𝑎𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡 + න 𝑒 −𝑎𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡
−∞ 0
1 1
= +
𝑎 − 𝑗𝜔 𝑎 + 𝑗𝜔
2𝑎
= 2
𝑎 + 𝜔2 8
 Continuous Fourier Transform
Examples
𝑥1 𝑡 𝑋1 𝑗𝜔
 𝑥1 𝑡 = 𝛿 𝑡 𝑋1 𝑗𝜔 =?

∞
𝑋1 𝑗𝜔 = න 𝛿 𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡 = 1
−∞

𝑥2 𝑡 𝑋2 𝑗𝜔
 𝑥2 𝑡 = 1 𝑋2 𝑗𝜔 =?
∞
𝑋2 𝑗𝜔 = න 𝑒 −𝑗𝜔𝑡 𝑑𝑡 = 2𝜋𝛿 𝜔
−∞
Hints: 1 ∞ 1 ∞
𝛿 𝑡 = ‫׬‬−∞
1 ∙ 𝑒 𝑗𝜔𝑡 𝑑𝜔 ⇒ 𝛿 −𝜔 = ‫׬‬−∞
1 ∙ 𝑒 −𝑗𝜔𝑡 𝑑𝑡
2𝜋 2𝜋 9
 Continuous Fourier Transform
Examples
1, 𝑡 < 𝑇1
𝑥 𝑡 =ቊ 𝑋 𝑗𝜔 =?
0, 𝑡 > 𝑇1

Solution
𝑇1
−𝑗𝜔𝑡
sin 𝜔𝑇1
𝑋 𝑗𝜔 = න 𝑒 𝑑𝑡 = 2
−𝑇1 𝜔

10
 Continuous Fourier Transform
Examples
1, 𝜔 < 𝑊
𝑋 𝑗𝜔 = ቊ 𝑥 𝑡 =?
0, 𝜔 > 𝑊
Solution
1 ∞
𝑥 𝑡 = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞
1 𝑊 𝑗𝜔𝑡 sin 𝑊𝑡
= න 𝑒 𝑑𝜔 =
2𝜋 −𝑊 𝜋𝑡
sin 𝜋𝜃
sinc(𝜃) =
𝜋𝜃
sin 𝑊𝑡 𝑊 sin 𝑊𝑡 𝑊 𝑊𝑡
= = sinc( )
𝜋𝑡 𝜋 𝑊𝑡 𝜋 𝜋 11
 Continuous Fourier Transform
Examples

12
 The Continuous-Time Fourier Transform
(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 Fourier transform for periodic signals
 A period signal can be represented by a FS, but also a FT
∞
1 ∞
𝑥 𝑡 = ෍ 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡 𝑥 𝑡 = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞
𝐾=−∞
 The relationship between 𝑎𝑘 and 𝑋 𝑗𝜔 ?
 Consider 𝑥1 𝑡 = 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡 , whose FT is
1 ∞
𝑥1 𝑡 = න 𝑋1 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔 = 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡 𝑋1 𝑗𝜔 = 2𝜋𝑎𝑘 𝛿(𝜔 − 𝑘𝜔0 )
2𝜋 −∞
∞ ∞

For 𝑥 𝑡 = ෍ 𝑎𝑘 𝑒 𝑗𝑘𝜔0 𝑡 𝑋 𝑗𝜔 = ෍ 𝑎𝑘 2𝜋𝛿(𝜔 − 𝑘𝜔0 )

𝐾=−∞ 𝐾=−∞
14
 Fourier transform for periodic signals
Examples
𝑋 𝑗𝜔 ?

Solution ∞
𝑗𝑘𝜔0 𝑡
sin 𝑘𝜔0 𝑇1
𝑥 𝑡 = ෍ 𝑎𝑘 𝑒 𝑎𝑘 =
𝜋𝑘
𝐾=−∞
∞

𝑋 𝑗𝜔 = ෍ 𝑎𝑘 2𝜋𝛿(𝜔 − 𝑘𝜔0 )
𝐾=−∞
∞
2 sin 𝑘𝜔0 𝑇1
= ෍ 𝛿(𝜔 − 𝑘𝜔0 )
𝑘 15
𝐾=−∞
 Fourier transform for periodic signals
Examples
𝑥1 𝑡 = sin 𝜔0 𝑡 𝑎1 = 1/2𝑗 𝑎−1 = −1/2𝑗 𝑎𝑘 = 0, 𝑘 ≠ ±1
∞
𝜋 𝜋 for sin 𝜔0 𝑡
𝑋1 𝑗𝜔 = ෍ 𝑎𝑘 2𝜋𝛿(𝜔 − 𝑘𝜔0 ) = 𝛿 𝜔 − 𝜔0 − 𝛿 𝜔 + 𝜔0
𝑗 𝑗
𝐾=−∞

𝑥2 𝑡 = cos 𝜔0 𝑡 𝑎𝑘 = 1/2, 𝑘 = ±1, 𝑎𝑘 = 0, 𝑘 ≠ ±1

𝑋1 𝑗𝜔 = 𝜋𝛿 𝜔 − 𝜔0 + 𝜋𝛿 𝜔 + 𝜔0 for cos 𝜔0 𝑡

16
 Fourier transform for periodic signals
Examples ∞

𝑥 𝑡 = ෍ 𝛿(𝑡 − 𝑘𝑇)
𝐾=−∞

1 𝑇/2 −𝑗𝑘𝜔 𝑡
1
𝑎𝑘 = න 𝛿(𝑡)𝑒 0 𝑑𝑡 =
𝑇 −𝑇/2 𝑇
∞
2𝜋
𝑋 𝑗𝜔 = ෍ 𝛿(𝜔 − 𝑘𝜔0 )
𝑇
𝐾=−∞
∞
2𝜋 2𝑘𝜋
= ෍ 𝛿(𝜔 − )
𝑇 𝑇
𝐾=−∞
17
 The Continuous-Time Fourier Transform
(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 Properties of continuous-time Fourier Transform
Short notation for FT pairs
∞ ∞
1
𝑥(𝑡) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔 𝑋 𝑗𝜔 = න 𝑥(𝑡)𝑒 −𝑗𝜔𝑡 𝑑𝑡
2𝜋 −∞ −∞

ℱ
𝑥 𝑡 𝑋 𝑗𝜔

𝑋 𝑗𝜔 = ℱ 𝑥 𝑡

𝑥 𝑡 = ℱ −1 𝑋 𝑗𝜔

19
 Properties of continuous-time Fourier Transform
Linearity ℱ
𝑥 𝑡 𝑋 𝑗𝜔
ℱ
𝑦 𝑡 𝑌 𝑗𝜔
⇓
ℱ
𝑎𝑥 𝑡 + 𝑏𝑦 𝑡 𝑎𝑋 𝑗𝜔 + 𝑏𝑌 𝑗𝜔

20
 Properties of continuous-time Fourier Transform
Time shifting ℱ ℱ
𝑥 𝑡 𝑋 𝑗𝜔 ⟹ 𝑥 𝑡 − 𝑡0 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔
 proof
1 ∞
𝑥(𝑡) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞

1 ∞ 1 ∞
𝑥(𝑡 − 𝑡0 ) = න 𝑋 𝑗𝜔 𝑒 𝑗𝜔(𝑡−𝑡0 ) 𝑑𝜔 = න 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞ 2𝜋 −∞

ℱ 𝑥 𝑡 = 𝑋 𝑗𝜔 = 𝑋 𝑗𝜔 𝑒 𝑗∢𝑋 𝑗𝜔

ℱ 𝑥 𝑡 − 𝑡0 = 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔 = 𝑋 𝑗𝜔 𝑒 𝑗∢𝑋 𝑗𝜔 −𝜔𝑡0

 A time shift on a signal introduces a phase shift into its FT, −𝜔𝑡0 ,
which is a linear function of 𝜔. 21
 Properties of continuous-time Fourier Transform
Examples
 𝑥(𝑡) can be expressed as
1
𝑥 𝑡 = 𝑥1 𝑡 − 2.5 + 𝑥2 (𝑡 − 2.5)
2
sin 𝜔𝑇1 sin 𝜔/2
𝑋1 𝑗𝜔 = 2 =2
𝜔 𝜔
sin 3𝜔/2
𝑋2 𝑗𝜔 = 2
𝜔
sin 𝜔/2 + 2 sin 3𝜔/2
𝑋 𝑗𝜔 = 𝑒 −𝑗5𝜔/2
𝜔
22
 Properties of continuous-time Fourier Transform
Conjugation and Conjugate Symmetry
ℱ ℱ
 Conjugation property 𝑥 𝑡 𝑋 𝑗𝜔 ⟹ 𝑥∗ 𝑡 𝑋 ∗ −𝑗𝜔
∞ ∗ ∞
𝑋 ∗ 𝑗𝜔 = න 𝑥(𝑡)𝑒 −𝑗𝜔𝑡 𝑑𝑡 = න 𝑥(𝑡)∗ 𝑒 𝑗𝜔𝑡 𝑑𝑡
−∞ −∞
∞
𝑋 ∗ −𝑗𝜔 = න 𝑥(𝑡)∗ 𝑒 −𝑗𝜔𝑡 𝑑𝑡 = ℱ 𝑥 ∗ 𝑡
−∞

 Conjugation Symmetry

𝑋 −𝑗𝜔 = 𝑋 ∗ 𝑗𝜔 𝑥 𝑡 real .
For a real-valued signal, the FT need only to be specified for positive frequencies
23
 Properties of continuous-time Fourier Transform
Time reversing ℱ ℱ
𝑥 𝑡 𝑋 𝑗𝜔 ⟹ 𝑥 −𝑡 𝑋 −𝑗𝜔

 𝑥 𝑡 even ⟹ 𝑋 𝑗𝜔 = 𝑋 −𝑗𝜔 , 𝑥 𝑡 real ⟹ 𝑋 −𝑗𝜔 = 𝑋 ∗ 𝑗𝜔

 𝑥 𝑡 real and even ⟹ 𝑋 𝑗𝜔 real and even
 𝑥 𝑡 real and odd ⟹ 𝑋 𝑗𝜔 purely imaginary and odd
 If 𝑥 𝑡 real
𝑥 𝑡 = 𝑥𝑒 𝑡 + 𝑥𝑜 (𝑡) ℱ
ℰ𝓋 𝑥 𝑡 ℛℯ 𝑋 𝑗𝜔
⇔ ℱ
ℱ 𝑥 𝑡 = ℱ 𝑥𝑒 𝑡 + ℱ 𝑥𝑜 (𝑡) 𝒪𝒹 𝑥 𝑡 ℐ𝓂 𝑋 𝑗𝜔
Real Imaginary
24
 Properties of continuous-time Fourier Transform
Example ℱ
𝑒 −𝑎𝑡 𝑢 𝑡 1/(𝑎 + 𝑗𝜔)
 For 𝑎 > 0
ℱ
𝑒 −𝑎|𝑡| 2𝑎/(𝑎2 + 𝜔2 )
 use FT properties

𝑒 −𝑎|𝑡| = 𝑒 −𝑎𝑡 𝑢 𝑡 + 𝑒 𝑎𝑡 𝑢 −𝑡 = 2ℰ𝓋 𝑒 −𝑎𝑡 𝑢 𝑡

ℱ 1
ℰ𝓋 𝑒 −𝑎𝑡 𝑢 𝑡 ℛℯ
𝑎 + 𝑗𝜔

ℱ 1 2𝑎
𝑒 −𝑎|𝑡| 2ℛℯ = 2
𝑎 + 𝑗𝜔 𝑎 + 𝜔2
25
 Properties of continuous-time Fourier Transform
Differential and integration
𝑡 1
ℱ 𝑑𝑥(𝑡) ℱ ℱ
𝑥 𝑡 𝑋 𝑗𝜔 ⟹ 𝑗𝜔𝑋 𝑗𝜔 න 𝑥 𝜏 𝑑𝜏 𝑋 𝑗𝜔 + 𝜋𝑋 0 𝛿(𝜔)
𝑑𝑡 −∞
𝑗𝜔

 Proof
𝑑𝑥(𝑡) 1 ∞ 𝑑(𝑒 𝑗𝜔𝑡 ) 1 ∞
= න 𝑋 𝑗𝜔 𝑑𝜔 = න 𝑋 𝑗𝜔 ∙ 𝑗𝜔 ∙ 𝑒 𝑗𝜔𝑡 𝑑𝜔
𝑑𝑡 2𝜋 −∞ 𝑑𝑡 2𝜋 −∞
𝑡 𝑡
1 ∞ 1 ∞ 𝑡
න 𝑥 𝜏 𝑑𝜏 = න න 𝑋 𝑗𝜔 𝑒 𝑗𝜔𝜏 𝑑𝜔 𝑑𝜏 = න 𝑋 𝑗𝜔 න 𝑒 𝑗𝜔𝜏 𝑑𝜏 𝑑𝜔
−∞ −∞ 2𝜋 −∞ 2𝜋 −∞ −∞
1 ∞ 𝑒 𝑗𝜔𝑡 𝑒 𝑗𝜔𝜏
= න 𝑋 𝑗𝜔 − lim 𝑑𝜔
2𝜋 −∞ 𝑗𝜔 𝜏→−∞ 𝑗𝜔
1 ∞ 𝑋 𝑗𝜔 𝑗𝜔𝑡 𝑒 𝑗𝜔𝜏
= න 𝑒 𝑑𝜔 , 𝜔 ≠ 0 lim ? DC component
𝜏→−∞ 𝑗𝜔
2𝜋 −∞ 𝑗𝜔 26
 Properties of continuous-time Fourier Transform
Differential and integration −𝑒 𝛼𝑡 , 𝑡 < 0
Let 𝑧(𝑡) = ൜ −𝛼𝑡 ,𝛼 > 0 sgn(𝑡) = lim 𝑧(𝑡)
𝑡
𝑒 ,𝑡 > 0 𝛼→0
ℱ 1
න 𝑥 𝜏 𝑑𝜏 𝑋 𝑗𝜔 + 𝜋𝑋 0 𝛿(𝜔) 0 ∞
𝑗𝜔
−∞ ℱ 𝑧(𝑡) = න −𝑒 𝛼𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡 + න 𝑒 −𝛼𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡
 Proof −∞
0
0
∞
𝑡 ∞ 𝑡
1 = − න 𝑒 (𝛼−𝑗𝜔)𝑡 𝑑𝑡 + න 𝑒 −(𝛼+𝑗𝜔)𝑡 𝑑𝑡
න 𝑥 𝜏 𝑑𝜏 = න 𝑋 𝑗𝜔 න 𝑒 𝑗𝜔𝜏 𝑑𝜏 𝑑𝜔 −∞ 0
−∞ 2𝜋 −∞ −∞
∞ ∞ 1 1 −2𝑗𝜔
1 = − = 2
= න 𝑋 𝑗𝜔 න 𝑢(𝑡 − 𝜏)𝑒 𝑗𝜔𝜏 𝑑𝜏 𝑑𝜔 𝛼 + 𝑗𝜔 𝛼 − 𝑗𝜔 𝛼 + 𝜔 2
2𝜋 −∞ −∞
1 ∞ ∞ −2𝑗𝜔 2
= න 𝑋 𝑗𝜔 න 𝑢(𝑝)𝑒 𝑗𝜔(𝑡−𝑝) 𝑑𝑝 𝑑𝜔 ℱ sgn(𝑡) = lim ℱ 𝑧(𝑡) = lim 2
𝛼→0 𝛼 + 𝜔 2
=
2𝜋 −∞ −∞
𝛼→0 𝑗𝜔
1 ∞ ∞
1 1
= න 𝑋 𝑗𝜔 න 𝑢(𝑝)𝑒 −𝑗𝜔𝑝 𝑑𝑝 𝑒 𝑗𝜔𝑡 𝑑𝜔 𝑢 𝑡 = sgn 𝑡 +
2𝜋 −∞ −∞ 2 2
𝑡 ∞
1 1 1
න 𝑥 𝜏 𝑑𝜏 = න 𝑋 𝑗𝜔 + 𝜋𝛿(𝜔) 𝑒 𝑗𝜔𝑡 𝑑𝜔 ℱ 𝑢(𝑡) = + 𝜋𝛿(𝜔)
−∞ 2𝜋 −∞ 𝑗𝜔 𝑗𝜔 27
 Properties of continuous-time Fourier Transform
Example FT of unit sept 𝑥 𝑡 = 𝑢(𝑡)
𝑡
ℱ
𝑔 𝑡 =𝛿 𝑡 𝐺 𝑗𝜔 = 1 𝑥 𝑡 = 𝑢 𝑡 = න 𝛿 𝜏 𝑑𝜏
−∞

 use integration property

1 1
𝑋 𝑗𝜔 = 𝐺 𝑗𝜔 + 𝜋𝐺 0 𝛿 𝜔 = + 𝜋𝛿(𝜔)
𝑗𝜔 𝑗𝜔

 Recover 𝐺 𝑗𝜔 by differential property

𝑑𝑢(𝑡) ℱ 1
𝛿 𝑡 = 𝑗𝜔 + 𝜋𝛿(𝜔) = 1
𝑑𝑡 𝑗𝜔
28
 Properties of continuous-time Fourier Transform
Example Determine the FT of 𝑥 𝑡

 Solution 𝑑
𝑔 𝑡 = 𝑥(𝑡)
𝑑𝑡 𝑑
𝑔 𝑡 = 𝑥(𝑡)
𝑑𝑡
2 sin 𝜔
𝐺 𝑗𝜔 = − 𝑒 𝑗𝜔 − 𝑒 −𝑗𝜔
𝜔

 use FT properties
1
𝑋 𝑗𝜔 = 𝐺 𝑗𝜔 + 𝜋𝐺 0 𝛿 𝜔
𝑗𝜔
2 sin 𝜔 2 cos 𝜔
𝑋 𝑗𝜔 = 2
−
𝑗𝜔 𝑗𝜔 29
 Properties of continuous-time Fourier Transform
Time and frequency scaling
ℱ ℱ 1 𝑗𝜔
𝑥 𝑡 𝑋 𝑗𝜔 ⟹ 𝑥 𝑎𝑡 𝑋
𝑎 𝑎
𝑎≠0

 Proof
∞
ℱ 𝑥 𝑎𝑡 = න 𝑥 𝑎𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡
−∞

1 ∞
න 𝑥 𝜏 𝑒 −𝑗 𝜔/𝑎 𝜏 𝑑𝜏 , 𝑎>0
𝑎 −∞
ℱ 𝑥 𝑎𝑡 =
1 ∞
− න 𝑥 𝜏 𝑒 −𝑗 𝜔/𝑎 𝜏 𝑑𝜏 , 𝑎<0
𝑎 −∞
30
 Properties of continuous-time Fourier Transform
Duality

31
 Properties of continuous-time Fourier Transform
Example 𝑋 𝑗𝜔
𝑥 𝑡

𝑥 𝑡 =𝛿 𝑡 𝑋 𝑗𝜔 = 1

𝑥 𝑡 =1 𝑋 𝑗𝜔 = 2𝜋𝛿 𝜔
Principle 𝑥 𝑡 𝑋 𝑗𝜔

1 ∞
𝑥(𝑡) = න 𝑋 𝑗𝜔 ∙ 𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞
1 ∞ ∞
𝑥(𝑗𝜔) = න 𝑋 𝑡 ∙ 𝑒 𝑗𝜔𝑡 𝑑𝑡 2𝜋 ∙ 𝑥(−𝑗𝜔) = න 𝑋 𝑡 ∙ 𝑒 −𝑗𝜔𝑡 𝑑𝑡
2𝜋 −∞
−∞ 32
 Properties of continuous-time Fourier Transform
2
Example 𝑔 𝑡 = 𝐺 𝑗𝜔 =?
1 + 𝑡2

Solution: calculate 𝐺 𝑗𝜔 is difficult; use duality property

ℱ
𝑒 −𝑎|𝑡| 2𝑎/(𝑎2 + 𝜔2 )

1 ∞ 2
𝑒 −|𝑡| = න 2
∙ 𝑒 𝑗𝜔𝑡
𝑑𝜔
2𝜋 −∞ 1 + 𝜔

∞
−|𝜔|
2 𝑗𝜔𝑡
2𝜋𝑒 =න 2
∙ 𝑒 𝑑𝑡 ∴ 𝐺 𝑗𝜔 = 2𝜋𝑒 −|𝜔|
−∞ 1 + 𝑡
33
 Properties of continuous-time Fourier Transform
Example
Duality property can determine or suggest other FT properties

𝑑𝑥(𝑡) ℱ ℱ 𝑑𝑋(𝑗𝜔)
𝑗𝜔𝑋 𝑗𝜔 ⇔ −𝑗𝑡𝑥 𝑡
𝑑𝜔
𝑑𝑡

𝑡 𝜔
ℱ 1 1 ℱ න 𝑥 𝜂 𝑑𝜂
න 𝑥 𝜏 𝑑𝜏
𝑗𝜔
𝑋 𝑗𝜔 + 𝜋𝑋 0 𝛿(𝜔) ⇔ − 𝑥(𝑡) + 𝜋𝑥 0 𝛿(𝑡)
𝑗𝑡
−∞ −∞

ℱ ℱ
𝑥(𝑡 − 𝑡0 ) 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔 ⇔ 𝑒 𝑗𝜔0 𝑡 𝑥 𝑡 𝑋 𝑗(𝜔 − 𝜔0 )

34
 Properties of continuous-time Fourier Transform
∞ ∞
Parseval’s relation 1
න 𝑥(𝑡) 2 𝑑𝑡 = න 𝑋 𝑗𝜔 2 𝑑𝜔
−∞ 2𝜋 −∞
 Proof
∞ ∞
න 𝑥(𝑡) 2 𝑑𝑡 = න 𝑥 𝑡 𝑥 ∗ (𝑡) 𝑑𝑡
−∞ −∞
∞
1 ∞ ∗
= න 𝑥(𝑡) න 𝑋 (𝑗𝜔)𝑒 −𝑗𝜔𝑡 𝑑𝜔 𝑑𝑡
−∞ 2𝜋 −∞
1 ∞ ∗ ∞
= න 𝑋 (𝑗𝜔) න 𝑥(𝑡)𝑒 −𝑗𝜔𝑡 𝑑𝑡 𝑑𝜔
2𝜋 −∞ −∞

1 ∞ 2 𝑑𝜔
= න 𝑋 𝑗𝜔
2𝜋 −∞
35
 The Continuous-Time Fourier Transform
(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 The convolution property
ℱ
𝑦 𝑡 =ℎ 𝑡 ∗𝑥 𝑡 𝑌 𝑗𝜔 = 𝐻 𝑗𝜔 𝑋 𝑗𝜔

 proof +∞ +∞
𝑌 𝑗𝜔 = ℱ 𝑦(𝑡) = න න 𝑥 𝜏 ℎ 𝑡 − 𝜏 𝑑𝜏 𝑒 −𝑗𝜔𝑡 𝑑𝑡
−∞ −∞
+∞ +∞
=න 𝑥 𝜏 න ℎ 𝑡 − 𝜏 𝑒 −𝑗𝜔𝑡 𝑑𝑡 𝑑𝜏
−∞ −∞
+∞ +∞
=න 𝑥 𝜏 𝑒 −𝑗𝜔𝜏 𝐻 𝑗𝜔 𝑑𝜏 = 𝐻 𝑗𝜔 න 𝑥 𝜏 𝑒 −𝑗𝜔𝜏 𝑑𝜏
−∞ −∞

= 𝐻 𝑗𝜔 𝑋 𝑗𝜔
 𝐻 𝑗𝜔 : Frequency response; important for analyzing LTI systems
 Only stable continuous-time LTI systems have 𝐻 𝑗𝜔
 Non-stable continuous-time LTI system: Laplace transform 37
 The convolution property
Example
𝑥(𝑡) ℎ 𝑡 𝑦(𝑡)

 Assume ℎ 𝑡 = 𝛿(𝑡 − 𝑡0 ), ℱ 𝑥 𝑡 = 𝑋(𝑗𝜔), determine 𝑌 𝑗𝜔

 Solution 1
𝐻 𝑗𝜔 = 𝑒 −𝑗𝜔𝑡0 𝑌 𝑗𝜔 = 𝐻 𝑗𝜔 𝑋 𝑗𝜔 = 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔
 Solution 2
𝑦 𝑡 = 𝑥(𝑡 − 𝑡0 ) 𝑌 𝑗𝜔 = 𝑒 −𝑗𝜔𝑡0 𝑋 𝑗𝜔

38
 The convolution property
Example 𝑑𝑥(𝑡)
𝑥(𝑡) ℎ 𝑡 𝑦(𝑡) =
𝑑𝑡

 Differentiation property ⇒ 𝑌 𝑗𝜔 = 𝑗𝜔𝑋 𝑗𝜔

 Convolution property ⇒ 𝑌 𝑗𝜔 = 𝐻(𝑗𝜔)𝑋 𝑗𝜔

 Therefore, 𝐻 𝑗𝜔 = 𝑗𝜔

39
 The convolution property
Example 𝑡
𝑥(𝑡) ℎ 𝑡 𝑦 𝑡 = න 𝑥 𝜏 𝑑𝜏 𝑌 𝑗𝜔 =?
−∞
𝑡
ℎ 𝑡 = න 𝛿 𝜏 𝑑𝜏 = 𝑢(𝑡)
−∞
1
 Frequency response 𝐻 𝑗𝜔 = + 𝜋𝛿 𝜔
𝑗𝜔
 Convolution property 𝑌 𝑗𝜔 = 𝐻(𝑗𝜔)𝑋 𝑗𝜔
1
𝑌 𝑗𝜔 = 𝑋 𝑗𝜔 + 𝜋𝑋 0 𝛿 𝜔
𝑗𝜔
 Consistent with integration property 40
 The convolution property
Example
𝑥(𝑡) ℎ 𝑡 𝑦(𝑡)

ℎ 𝑡 = 𝑒 −𝑎𝑡 𝑢 𝑡 , a > 0 𝑥 𝑡 = 𝑒 −𝑏𝑡 𝑢 𝑡 , b > 0 𝑦 𝑡 =?

 Solution 𝑏 ≠ 𝑎
1 1 1
𝐻 𝑗𝜔 = 𝑋 𝑗𝜔 = 𝑌 𝑗𝜔 =
𝑎 + 𝑗𝜔 𝑏 + 𝑗𝜔 (𝑎 + 𝑗𝜔)(𝑏 + 𝑗𝜔)
𝐴 𝐵 1
𝑌 𝑗𝜔 = + 𝐴= = −𝐵
𝑎 + 𝑗𝜔 𝑏 + 𝑗𝜔 𝑏−𝑎
1 1 1 1
𝑌 𝑗𝜔 = − 𝑦 𝑡 = 𝑒 −𝑎𝑡 − 𝑒 −𝑏𝑡 𝑢 𝑡 , 𝑏 ≠ 𝑎
𝑏 − 𝑎 𝑎 + 𝑗𝜔 𝑏 + 𝑗𝜔 𝑏−𝑎
41
 The convolution property
Example
𝑥(𝑡) ℎ 𝑡 𝑦(𝑡)

ℎ 𝑡 = 𝑒 −𝑎𝑡 𝑢 𝑡 , a > 0 𝑥 𝑡 = 𝑒 −𝑏𝑡 𝑢 𝑡 , b > 0 𝑦 𝑡 =?

 Solution 𝑏 = 𝑎
1 𝑑 1
𝑌 𝑗𝜔 =
(𝑎 + 𝑗𝜔)2 = 𝑗 𝑑𝜔 𝑎 + 𝑗𝜔

𝑒 −𝑎𝑡 𝑢 𝑡 1/(𝑎 + 𝑗𝜔)

𝑑 1
𝑡𝑒 −𝑎𝑡 𝑢 𝑡 𝑗
𝑑𝜔 𝑎 + 𝑗𝜔
∴ 𝑦 𝑡 = 𝑡𝑒 −𝑎𝑡 𝑢 𝑡
42
 The Continuous-Time Fourier Transform
(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 The multiplication property
ℱ 1 ∞
𝑟 𝑡 =𝑠 𝑡 𝑝 𝑡 𝑅 𝑗𝜔 = න 𝑆(𝑗𝜃)𝑃(𝑗(𝜔 − 𝜃))𝑑𝜃
2𝜋 −∞

 multiplication of two signals is often referred to as amplitude modulation

1 ∞ 𝑗𝜃𝑡
1 ∞
′ 𝑗𝜔 ′𝑡
𝑠 𝑡 𝑝 𝑡 = න 𝑆(𝑗𝜃)𝑒 𝑑𝜃 න 𝑃(𝑗𝜔 )𝑒 𝑑𝜔′
2𝜋 −∞ 2𝜋 −∞
1 ∞ 1 ∞ ′ 𝑗(𝜃+𝜔 ′ )𝑡
= න න 𝑆(𝑗𝜃) 𝑃(𝑗𝜔 )𝑒 𝑑𝜃𝑑𝜔′
2𝜋 −∞ 2𝜋 −∞
𝜔′ = 𝜔 − 𝜃
1 ∞ 1 ∞
= න න 𝑆[𝑗(𝜃)] 𝑃(𝑗(𝜔 − 𝜃))𝑒 𝑗𝜔𝑡 𝑑𝜃𝑑𝜔
2𝜋 −∞ 2𝜋 −∞
1 ∞ 1 ∞ 𝑅(𝑗𝜔)
= න න 𝑆(𝑗𝜃)𝑃(𝑗(𝜔 − 𝜃))𝑑𝜃𝑒 𝑗𝜔𝑡 𝑑𝜔
2𝜋 −∞ 2𝜋 −∞ 44
 The multiplication property
Example
Consider a signal 𝑝 𝑡 = cos 𝜔0 𝑡 and a signal 𝑠 𝑡
with spectrum 𝑆(𝑗𝜔), determine the FT of 𝑟 𝑡 =
𝑝 𝑡 𝑠 𝑡
 Solution 𝑃 𝑗𝜔 = 𝜋𝛿 𝜔 − 𝜔0 + 𝜋𝛿(𝜔 + 𝜔0 )

𝑅 𝑗𝜔 = 1/2𝜋 ∙ 𝑆 𝑗𝜔 ∗ 𝑃 𝑗𝜔

= 1/2𝜋 ∙ 𝑆 𝑗𝜔 ∗ [𝜋𝛿 𝜔 − 𝜔0 + 𝜋𝛿(𝜔 + 𝜔0 )]

= 1/2[𝑆[𝑗 𝜔 − 𝜔0 ] + 𝑆[𝑗 𝜔 + 𝜔0 ]
45
 The multiplication property
Example 𝑔(𝑡) = 𝑟 𝑡 𝑝 𝑡 𝐺 𝑗𝜔 =?

46
 The multiplication property
Example sin(𝑡) sin(𝑡/2)
𝑥 𝑡 = 𝑋 𝑗𝜔 =?
𝜋𝑡 2

 Solution sin(𝑡) sin(𝑡/2)

𝑥 𝑡 =𝜋
𝜋𝑡 𝜋𝑡

1 sin(𝑡) sin(𝑡/2)
𝑋 𝑗𝜔 = ℱ ∗ℱ
2 𝜋𝑡 𝜋𝑡

47
 The Continuous-Time Fourier Transform
(ch.4)
 Representation of aperiodic signals- Continuous Fourier Transform
 Fourier transform for periodic signals
 Properties of continuous-time Fourier Transform
 The convolution property
 The multiplication property
 System characterized by differential equations
 System characterized by differential equations
𝑁 𝑀
𝑘
 Differential equation 𝑑 𝑦(𝑡) 𝑑 𝑘 𝑥(𝑡)
෍ 𝑎𝑘 𝑘
= ෍ 𝑏𝑘
𝑑𝑡 𝑑𝑡𝑘
𝐾=0 𝐾=0

𝑌 𝑗𝜔
𝑌 𝑗𝜔 = 𝐻 𝑗𝜔 𝑋 𝑗𝜔 ⟹ 𝐻 𝑗𝜔 =
𝑋 𝑗𝜔
𝑁 𝑀 𝑁 𝑀
𝑑 𝑘 𝑦(𝑡) 𝑑 𝑘 𝑥(𝑡) 𝑑 𝑘 𝑦(𝑡) 𝑑 𝑘 𝑥(𝑡)
ℱ ෍ 𝑎𝑘 = ℱ ෍ 𝑏𝑘 ⟹ ෍ 𝑎𝑘 ℱ = ෍ 𝑏𝑘 ℱ
𝑑𝑡𝑘 𝑑𝑡𝑘 𝑑𝑡𝑘 𝑑𝑡𝑘
𝐾=0 𝐾=0 𝐾=0 𝐾=0

⟹
𝑁 𝑀 𝑁 𝑀
𝑘 𝑘
⟹ ෍ 𝑎𝑘 𝑗𝜔 𝑘 𝑌 𝑗𝜔 = ෍ 𝑏𝑘 𝑗𝜔 𝑘 𝑋 𝑗𝜔
𝑌 𝑗𝜔 ෍ 𝑎𝑘 𝑗𝜔 = 𝑋 𝑗𝜔 ෍ 𝑏𝑘 𝑗𝜔
𝐾=0 𝐾=0 𝐾=0 𝐾=0
⟹

𝑌 𝑗𝜔 σ𝑀
𝑘=0 𝑏𝑘 𝑗𝜔
𝑘
𝐻 𝑗𝜔 = = 𝑁 𝑘
𝑋 𝑗𝜔 σ𝑘=0 𝑎𝑘 𝑗𝜔 49
 System characterized by differential equations
Example 𝑑𝑦(𝑡)
+ 𝑎𝑦(𝑡) = 𝑥(𝑡) a>0
𝑑𝑡

𝑑𝑦(𝑡)
ℱ + 𝑎𝑦(𝑡) = ℱ 𝑥(𝑡)
𝑑𝑡

𝑗𝜔𝑌 𝑗𝜔 + 𝑎𝑌 𝑗𝜔 = 𝑋 𝑗𝜔

1
𝐻 𝑗𝜔 = ⟹ ℎ 𝑡 = 𝑒 −𝑎𝑡 𝑢(𝑡)
𝑗𝜔 + 𝑎

50
 System characterized by differential equations
Example 𝑑2 𝑦(𝑡) 𝑑𝑦(𝑡) 𝑑𝑥(𝑡)
2
+4 + 3𝑦(𝑡) = + 2𝑥(𝑡)
𝑑𝑡 𝑑𝑡 𝑑𝑡

(𝑗𝜔) + 2
𝐻 𝑗𝜔 =
(𝑗𝜔)2 +4(𝑗𝜔) + 3

1 1 1 1
𝐻 𝑗𝜔 = +
2 𝑗𝜔 + 1 2 𝑗𝜔 + 3
1 −𝑡 1 −3𝑡
ℎ 𝑡 = 𝑒 𝑢 𝑡 + 𝑒 𝑢(𝑡)
2 2
51
 System characterized by differential equations
Example
𝑥(𝑡) ℎ 𝑡 𝑦(𝑡)

𝑑 2 𝑦(𝑡) 𝑑𝑦(𝑡) 𝑑𝑥(𝑡)

𝑥 𝑡 = 𝑒 −𝑡 𝑢(𝑡) +4 + 3𝑦(𝑡) = + 2𝑥(𝑡) 𝑦 𝑡 =?
𝑑𝑡 2 𝑑𝑡 𝑑𝑡
 Solution
𝑗𝜔 + 2 1 𝑗𝜔 + 2
𝑌 𝑗𝜔 = 𝐻 𝑗𝜔 𝑋 𝑗𝜔 = =
(𝑗𝜔 + 1)(𝑗𝜔 + 3) 𝑗𝜔 + 1 (𝑗𝜔 + 1)2 (𝑗𝜔 + 3)
𝐴11 𝐴12 𝐴21 1 1 1
= + 2
+ 𝐴11 = 𝐴12 = 𝐴21 =−
𝑗𝜔 + 1 (𝑗𝜔 + 1) 𝑗𝜔 + 3 4 2 4

1 1 1 1 1 1 1 −𝑡 1 −𝑡 1 −𝑡
𝑌 𝑗𝜔 = + − ⟹ 𝑦 𝑡 = 𝑒 + 𝑡𝑒 − 𝑒 𝑢 𝑡
4 𝑗𝜔 + 1 4 (𝑗𝜔 + 1) 2 4 𝑗𝜔 + 3 4 2 4
52