Signals and System

Chapter 5: The Fourier Transform

1
Objectives
By the end of this chapter, the student should be able to:
• Recall the formula for Fourier Transform and Inverse Fourier Transform
• Compute the Fourier Transform and Inverse Fourier Transform using the integration formula
• Recall the Fourier transforms of popular functions
• Recall the Fourier transform properties
• Utilize the Fourier transform tables (Tables 5.1 and 5.2) to compute the Fourier transform of
various functions

2
Revision
Example 4.6
Find the Fourier Series of the function depicted below:
𝑓𝑓(𝑡𝑡)
𝑉𝑉

𝑇𝑇/2
1 −𝑗𝑗𝜔𝜔0 𝑘𝑘𝑘𝑘
𝑉𝑉 𝑗𝑗𝜔𝜔0 𝑘𝑘𝑘𝑘 𝑗𝑗𝜔𝜔 𝑘𝑘𝑘𝑘
− 0 𝑉𝑉 𝑘𝑘𝑘𝑘𝜔𝜔0
𝐶𝐶𝑘𝑘 = � 𝑉𝑉𝑒𝑒 𝑑𝑑𝑑𝑑 = ⋅ 𝑒𝑒 2 − 𝑒𝑒 2 = sin
𝑇𝑇0 𝑗𝑗𝑗𝑗𝑗𝑗𝑗 𝜋𝜋𝜋𝜋 2
−𝑇𝑇/2
𝑉𝑉 𝑘𝑘𝑘𝑘𝜔𝜔0 𝑘𝑘𝜔𝜔0
sin 𝑥𝑥 sin ⋅ 𝑉𝑉𝑉𝑉 𝑘𝑘𝑘𝑘𝜔𝜔0
Define: sinc 𝑥𝑥 ≜ , then 𝐶𝐶𝑘𝑘 = 𝜋𝜋𝜋𝜋 2
𝑘𝑘𝜔𝜔0
2
= sinc( )
𝑥𝑥 𝑇𝑇0 2
2
3
sinc Function
• The sinc function appears often in EE classes.
• In some books it is defined as

sin 𝑥𝑥
sinc( 𝑥𝑥) ≜
𝑥𝑥

• Next, we will learn how to plot the sinc function.

• At 𝑥𝑥 = 0,
sin 𝑥𝑥
sinc 0 = lim = lim cos(𝑥𝑥) = 1 Note: in some books, the sinc
𝑥𝑥→0 𝑥𝑥 𝑥𝑥→0
• Clearly, the zero crossings appear at function is defined as
𝑥𝑥 = ±𝜋𝜋, ±2𝜋𝜋, ±3𝜋𝜋 … etc sin 𝜋𝜋𝜋𝜋
sinc 𝑥𝑥 ≜
𝜋𝜋𝜋𝜋 4
Fourier Transform
Introduction
• Fourier transform is a method of mathematical representation of signals and systems in the
frequency domain

• Engineers use the Fourier transform to simplify the mathematical analysis of signals and
systems and for explaining physical phenomena mathematically

• It is widely used in the field of electrical engineering, especially in the study of

communication signals and systems

• For this reason, every student of electrical engineering should become familiar with the
Fourier transform and its applications

• Fourier series can be applied to periodic signals. The Fourier transform is derived to
generalize the concepts of frequency analysis for non-periodic signals.
5
Fourier Transform
Derivation
• The Fourier transform is derived from the Fourier series by relaxing the period of the periodic function to reach
∞, so the signal never repeats itself as illustrated in the figure below.

Periodic signal with period 𝑇𝑇1

Periodic signal with period 𝑇𝑇2 = 4𝑇𝑇1

Periodic signal with period 𝑇𝑇0 = ∞ (non-periodic)

• For detailed mathematical derivation, refer to the textbook pages 201 – 203. We will focus instead on the
physical meaning, computation of the c of Fourier Transform, and computation of the inverse Fourier Transform.6
Fourier Transform
Definition
• The Fourier Transform (FT) of aperiodic signal 𝑓𝑓(𝑡𝑡) is denoted ℱ{𝑓𝑓(𝑡𝑡)}, and is
defined as • Negative exponent
• Integration with respect to d𝑡𝑡
∞ • 𝑡𝑡 will be integrated out; i.e., the result does not
ℱ[𝑓𝑓(𝑡𝑡)] = 𝐹𝐹(𝜔𝜔) = � 𝑓𝑓(𝑡𝑡)𝑒𝑒 −𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑 contain any 𝑡𝑡
−∞

• The Inverse Fourier Transform (IFT ) of 𝐹𝐹(𝜔𝜔) is denoted as ℱ −1 [𝐹𝐹(𝜔𝜔)], and is

defined as • Positive exponent
• Integration with respect to d𝜔𝜔
∞
−1
1 • 𝜔𝜔 will be integrated out; i.e., the does not
ℱ 𝐹𝐹 𝜔𝜔 = 𝑓𝑓(𝑡𝑡) = � 𝐹𝐹(𝜔𝜔)𝑒𝑒 𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑 contain any 𝜔𝜔
2𝜋𝜋 −∞

• We say 𝑓𝑓(𝑡𝑡) and 𝐹𝐹(𝜔𝜔) is a Fourier Transform Pair. We write:

ℱ
𝑓𝑓(𝑡𝑡) 𝐹𝐹(𝜔𝜔) or 𝑓𝑓 𝑡𝑡 𝐹𝐹(𝜔𝜔) 7
Fourier Transform
Notes on the Definition
Notes:

• In some communication systems books, authors prefer to express the Fourier transform as a function 𝑓𝑓,
were 𝜔𝜔 = 2𝜋𝜋𝜋𝜋. Then:
∞
• 𝐹𝐹 𝑓𝑓 = ∫−∞ 𝑓𝑓 𝑡𝑡 𝑒𝑒 −𝑗𝑗𝑗𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑, and

• the normalization factor will disappear in the inverse Fourier definition because of the substitution
∞
𝜔𝜔 = 2𝜋𝜋𝜋𝜋 ⇒ 𝑑𝑑𝑑𝑑 = 2𝜋𝜋𝜋𝜋𝜋𝜋. Which leads to 𝑓𝑓 𝑡𝑡 = ∫−∞ 𝐹𝐹 𝑓𝑓 𝑒𝑒 𝑗𝑗𝑗𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑

• Some books use the notation 𝐹𝐹(𝑗𝑗𝜔𝜔) which is the same as 𝐹𝐹(𝜔𝜔)

• The Fourier transform exists if and only if the Dirichlet conditions are satisfied. Please refer to the pages
206 – 207 for more information.

• For this class, you do not have to check if the Fourier transform exists every time you compute the Fourier
the transform.
8
Fourier Transform
Example 5.1
𝑡𝑡
Find the Fourier transform of the function 𝑓𝑓 𝑡𝑡 = 𝑉𝑉 ⋅ rect( ) (Note the similarity to Example
𝑇𝑇
4.6)
Solution:
𝑇𝑇/2 𝜔𝜔𝜔𝜔
𝑉𝑉 𝑗𝑗𝑗𝑗𝑗𝑗 𝑗𝑗𝑗𝑗𝑗𝑗 2𝑉𝑉 sin
−𝑗𝑗𝑗𝑗𝑗𝑗 − 2
𝐹𝐹 𝜔𝜔 = � 𝑉𝑉𝑒𝑒 𝑑𝑑𝑑𝑑 = ⋅ 𝑒𝑒 2 − 𝑒𝑒 2 =
𝑗𝑗𝑗𝑗 𝜔𝜔
−𝑇𝑇/2
𝜔𝜔𝜔𝜔
𝑉𝑉 sin 2 𝜔𝜔𝜔𝜔
= ⋅ 𝑇𝑇 = 𝑇𝑇𝑇𝑇 sinc
𝜔𝜔𝜔𝜔 2
2
• Note: the form of the Fourier transform is similar to the
result obtained in Example 4.6
9
Fourier Transform
Magnitude Spectrum and Phase Spectrum
• In general, the Fourier transform 𝐹𝐹 𝜔𝜔 is a complex function. As a result, 𝐹𝐹(𝜔𝜔) has a
magnitude |𝐹𝐹 𝜔𝜔 | and a phase 𝜃𝜃 𝜔𝜔 = ∠𝐹𝐹(𝜔𝜔) such that
𝐹𝐹 𝜔𝜔 = 𝐹𝐹 𝜔𝜔 𝑒𝑒 𝑗𝑗𝑗𝑗(𝜔𝜔) = 𝐹𝐹 𝜔𝜔 ∠𝐹𝐹(𝜔𝜔)

• The Magnitude Spectrum is the plot of the 𝐹𝐹 𝜔𝜔 vs 𝜔𝜔 (or |𝐹𝐹 𝑓𝑓 | vs 𝑓𝑓); the Phase Spectrum
is the 𝜃𝜃(𝜔𝜔) vs 𝜔𝜔 (or 𝜃𝜃(𝑓𝑓) vs 𝑓𝑓)

• To plot the magnitude and phase spectra, students must find the magnitude and the phase of
a complex function. This skill is illustrated by examples later.

10
Fourier Transform
Example
Find the Fourier transform, and magnitude and phase spectra of: Magnitude Spectrum
a) 𝑔𝑔 𝑡𝑡 = 𝛿𝛿(𝑡𝑡). 𝑍𝑍 𝜔𝜔

b) 𝑧𝑧 𝑡𝑡 = 𝐴𝐴𝐴𝐴(𝑡𝑡 − 𝑡𝑡0 )
Solution:
∞
a) 𝐺𝐺 𝜔𝜔 = ∫−∞ 𝛿𝛿 𝑡𝑡 𝑒𝑒 −𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑 = 𝑒𝑒 0 = 1 ⇒ 𝐺𝐺 𝜔𝜔 =
1& ∠𝐺𝐺(𝜔𝜔) = 0 (draw the spectra by yourself)
ℱ
𝛿𝛿 𝑡𝑡 1 Phase Spectrum
∞ ∞
b) 𝑍𝑍 𝜔𝜔 = ∫−∞ 𝐴𝐴𝐴𝐴 𝑡𝑡 − 𝑡𝑡0 𝑒𝑒 −𝑗𝑗𝑗𝑗𝑗𝑗 𝑑𝑑𝑑𝑑 = ∫−∞ 𝐴𝐴𝐴𝐴 𝜏𝜏 𝑒𝑒 𝑗𝑗𝑗𝑗(𝜏𝜏+𝑡𝑡0 ) 𝑑𝑑𝑑𝑑 = ∠𝑍𝑍(𝜔𝜔)
𝐴𝐴𝑒𝑒 −𝑗𝑗𝑗𝑗𝑡𝑡0
⇒ 𝑍𝑍 𝜔𝜔 = 𝐴𝐴 & ∠𝑍𝑍 𝜔𝜔 = −𝜔𝜔𝑡𝑡0
ℱ
𝐴𝐴𝐴𝐴 𝑡𝑡 − 𝑡𝑡0 𝐴𝐴𝑒𝑒 −𝑗𝑗𝑗𝑗𝑡𝑡0
11
Fourier Transform
Example
Find the inverse Fourier transform of 𝐹𝐹 𝜔𝜔 = 𝛿𝛿(𝜔𝜔 − 𝜔𝜔0 ).
Solution
∞ ∞
1 𝑗𝑗𝑗𝑗𝑗𝑗
1 1 𝑗𝑗𝜔𝜔 𝑡𝑡
𝑓𝑓 𝑡𝑡 = � 𝛿𝛿 𝜔𝜔 − 𝜔𝜔0 𝑒𝑒 𝑑𝑑𝑑𝑑 = � 𝛿𝛿 𝛼𝛼 𝑒𝑒 𝑗𝑗 𝛼𝛼+𝜔𝜔0 𝑡𝑡
𝑑𝑑𝑑𝑑 = 𝑒𝑒 0
2𝜋𝜋 2𝜋𝜋 2𝜋𝜋
−∞ −∞

ℱ
𝑒𝑒 𝑗𝑗𝜔𝜔0 𝑡𝑡 2𝜋𝜋𝜋𝜋(𝜔𝜔 − 𝜔𝜔0 )

Note the duality!

12