ENGINEERING ANALYSIS

CHAPTER 2
FOURIER TRANSFORM “ FT ”

A. Prof. Dr. Yazen A K 1

INTRODUCTION:
 The word Fourier refers to the scientist JOSEPH FOURIER.
He was a French mathematician and physicist.

 The Fourier Transform (FT) decomposes

a function of time (signal) into the sinusoid
frequencies that make it up. It is a technique
that transforms between two different set of
information for the same function.

 The Fourier Transform is a mathematical technique that transforms a

function of time, x(t), to a complex function of frequency X(ejw) or simply
X(w). (Transforming from Time domain to Frequency domain).
2
 The “PRISM” is a simple and famous example of this process.

Frequency domain
Time Domain
SPECTRUM

3
4
F( ω) is just another way of looking at a function or wave f(t).
5
The Rectangular function can be simulated with sinusoidal functions

6
Every time domain function can be represented as a sum of (infinite) sine waves

7
8
 Time domain

Frequency domain

9
*
As the function
has sharp
transitions in
time domain
leads to more
frequency
components
(Harmonics) in
freq. Domain

10
 Fourier Transform Operator

F [f(t)] = F(w)

F -1 [F(w)] = f(t)

⇄
Fourier Transform Pairs “Definition”:
For a continuous-time function x(t)
∞

𝐹 𝜔 = න 𝑓 𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡 Fourier Transform (FT)

−∞
∞
1
𝑓 𝑡 = න 𝐹 𝜔 𝑒 𝑗𝜔𝑡 𝑑𝜔 Inverse Fourier Transform (IFT)
2𝜋
−∞ 11
The Fourier Transform function is a complex function as shown below:

𝑭 𝝎 = 𝑭(𝝎) 𝒆𝒋𝜽(𝝎) or 𝑭 𝝎 = 𝑭(𝝎) ∟F(w) or 𝑭 𝝎 = 𝑭(𝝎) ∟q(w)

Where :

w = 2pf

𝐅(𝛚) =magnitude c/cs, or called absolute value or Magnitude spectrum.

∟F(w) = q (w) is the angle (Phase) c/cs, or Phase spectrum this is related

with the imaginary part of the function, so if F(w) appears as real, then

q(w)=0 (there is No phase c/cs).

12
Fourier Transform of the basic and some common functions:

 Fourier Transform of Delta Function f(t) = d(t)

F [d(t)] = 1

13
 Fourier Transform of constant function f(t) = 1

14
 Fourier Transform of the +ve unit step function:
∞
𝟏
𝑭 𝝎 = න 𝒆−𝒋𝝎𝒕 𝒅𝒕 = 𝝅𝜹 𝝎 +
𝒋𝝎
𝟎

𝟏
𝒖 𝒕 ⇆ 𝝅𝜹 𝝎 +
𝒋𝝎

 Fourier Transform of the –ve unit step function:

tutorialspoint.com/fourier-transform-of-unit-step-function

𝟎
𝟏
𝑭 𝝎 = න 𝒆−𝒋𝝎𝒕 𝒅𝒕 = 𝝅𝜹 𝝎 +
𝒋𝝎
−∞

𝟏
𝒖 −𝒕 ⇆ 𝝅𝜹 𝝎 −
𝒋𝝎
15
 Fourier Transform of the real exponential function:

16
Example: By using definition, Find the Fourier Transform function of the
exponential function shown below:

f(t) = e-at t≥ 0 or e-at u(t)

Sol:

1 𝑎 − 𝑗𝜔 𝑎 − 𝑗𝜔 𝑎 𝜔
𝐹 𝜔 = × = 2 = − 𝑗
𝑎 + 𝑗𝜔 𝑎 − 𝑗𝜔 𝑎 + 𝜔 2 𝑎2 + 𝜔 2 𝑎2 + 𝜔 2

𝑎 2 +(
𝜔 2 =
1
𝐹 𝜔 = ( 2 ) )
𝑎 + 𝜔2 𝑎2 + 𝜔 2 𝑎2 + 𝜔 2

−𝜔
𝜃 𝜔 = tan−1 ( )
𝑎
17
Example: Sol:
By using definition,
Find the Fourier
Transform function of
the double sided
decaying exponential
function shown below:

f(t) = e-a|t| a> 0

18
Any number divided by infinity is equal to 0

19
 Fourier Transform of Sign (Signum) function:

20
Example: By using definition, Find the Fourier Transform function of the pulse
function (rectangular) shown below:
Sol: f(t) can be written as following:

𝛕 𝛕
𝐀 − ≤𝐭≤
𝐟(𝐭) = ቐ 𝟐 𝟐
𝟎 𝐞𝐥𝐬𝐞𝐰𝐡𝐞𝐫
or can be written as follow:

f(t) = 𝑨 𝒓𝒆𝒄𝒕(
𝒕
)
𝝉
A : Amplitude
By using the definition, we can find the Fourier Transform F(w)
t: Duration (Time width)

𝑡 𝑒 −𝑗𝜔𝑡 𝑑𝑡 =
∞
F ω = ‫׬‬−∞ 𝑓

21
22
 𝒕 𝝉
𝑨 𝒓𝒆𝒄𝒕 ↔ 𝑨𝝉𝒔𝒊𝒏𝒄( 𝝎)
𝝉 𝟐

23
Note:
Since f(t) is even function leads that F(w) is a real function, then there is only magnitude c/cs
shown above and there is no phase c/cs, q(w) = 0

24
The effects of A and t on the Frequency domain (spectrum):
𝑡 𝑡 𝑡
𝑟𝑒𝑐𝑡( ) 𝑟𝑒𝑐𝑡( ) 𝑟𝑒𝑐𝑡( )
6 4 2

𝟔𝒔𝒊𝒏𝒄(𝟑𝝎) 𝟒𝒔𝒊𝒏𝒄(𝟐𝝎) 𝟐𝒔𝒊𝒏𝒄(𝝎)

Wider rectangular time pulse leads to narrow spectrum and vise versa 25
Fourier Transform Properties: There are many properties of FT, we will
deal with the most important of them, these properties can simplify the
process of calculating FT rather than using definition.
The following are some of them:
1. Linearity (superposition).
2. Shift in time domain (Delay – Advance).
3. Time scaling.
4. Shift in Frequency domain.
5. Differentiation.
6. Convolution in Time domain.

In next slides, we will use the following notation to explain the properties
f(t) ⇌ F(w), and h(t) ⇌ H(w) 26
1. Linearity (superposition):

Af(t) ± Bh(t) ⇌ AF(w) ± BH(w)

2. Shift in time domain (Delay / Advance):

f(t ± to) ⇌ F(w) 𝒆±𝒋𝝎𝒕𝒐 [ shift in time leads to complex exp in Freq ]

shift in Time ⇄ exp in Freq

3. Scale in time domain:

𝟏 𝝎
f(at ) ⇌ 𝑭( ) 27
𝒂 𝒂
Example: if the FT of x(t) is X(w), then find the FT of y(t) = x(2t-3)

Solution: Shift krawa pashan scale...

yasai shift pashan

yasai scale bakardenin.
x(t) ⇄ X(w)

y(t) = x(2t-3) ⇄ Y(w) = ?

𝟏 𝝎 −𝒋𝟑𝝎 𝟏 𝝎 −𝒋𝟏.𝟓𝝎
𝑿 𝝎 → 𝑿 𝝎 𝒆−𝒋𝟑𝝎 → 𝑿 𝒆 𝟐 → 𝒀 𝝎 = 𝑿 𝒆
𝟐 𝟐 𝟐 𝟐
we gotta divide this
as well.

28
4. Shift in Frequency domain:

𝒇 𝒕 𝒆±𝒋𝝎𝒐𝒕 ⇄ 𝑭(𝝎 ∓ 𝝎𝒐 )

Complex exp in Time Shift in Freq.

[ complex exp in time leads to shift in Freq. ]

29
Example: Find the FT of the complex exponential functions: 𝒆±𝒋𝝎𝒐 𝒕 , 𝒆−𝒋𝟓𝒕 and 𝒆𝒋𝝅𝒕

Sol: This can depend on property (5), since we have exponential in time domain, then it
will be shift in frequency domain

𝒆±𝒋𝝎𝒐𝒕 ⇄ 𝟐𝝅𝜹(𝝎 ∓ 𝝎𝒐 )

f(t) = 1 ⇄ F(w) = 2pd(w)

𝒆−𝒋𝟓𝒕 ⇄ 𝟐𝝅𝜹(𝝎 + 𝟓)

𝒆𝒋𝝅𝒕 ⇄ 𝟐𝝅𝜹(𝝎 − 𝝅) 30
Example: Find the FT of the sinusoidal functions: sinwot , coswot
like it's theta.

Sol: By Euler’s formula,

𝟏
𝒄𝒐𝒔𝝎𝒐 𝒕 = [ 𝒆𝒋𝝎𝒐𝒕 + 𝒆−𝒋𝝎𝒐𝒕 ] +-
𝟐
𝟏 𝟏 𝟏 𝟏
𝒆𝒋𝝎𝒐𝒕 + 𝒆−𝒋𝝎𝒐𝒕 ⇄ × 𝟐𝝅𝜹 𝝎 − 𝝎𝒐 + × 𝟐𝝅𝜹 𝝎 + 𝝎𝒐
𝟐 𝟐 𝟐 𝟐

= 𝝅𝜹 𝝎 − 𝝎𝒐 + 𝝅𝜹 𝝎 + 𝝎𝒐

Fast Way :D

f (t) F(w)
𝒄𝒐𝒔𝝎𝒐 𝒕 ⇄ 𝝅𝜹 𝝎 − 𝝎𝒐 + 𝝅𝜹 𝝎 + 𝝎𝒐 -++ 31
 𝟏 𝒋𝝎 𝒕
𝒔𝒊𝒏𝝎𝒐 𝒕 = 𝒆 𝒐 − 𝒆−𝒋𝝎𝒐 𝒕 --
𝟐𝒋
𝟏 𝒋𝝎 𝒕 𝟏 𝟏 𝝅 𝝅
= 𝒆 𝒐 − 𝒆−𝒋𝝎𝒐 𝒕 ⇄ × 𝟐𝝅𝜹 𝝎 − 𝝎𝒐 − × 𝟐𝝅𝜹 𝝎 + 𝝎𝒐 = 𝜹 𝝎 − 𝝎𝒐 − 𝜹 𝝎 + 𝝎𝒐
𝟐𝒋 𝟐𝒋 𝟐𝒋 𝒋 𝒋
= −𝒋𝝅𝜹 𝝎 − 𝝎𝒐 + 𝒋𝝅𝜹 𝝎 + 𝝎𝒐

f(t) F(w) +--

𝒔𝒊𝒏𝝎𝒐 𝒕 ⇄ 𝒋𝝅𝜹 𝝎 + 𝝎𝒐 − 𝒋𝝅𝜹 𝝎 − 𝝎𝒐 32
Single (one) tone ⇄

Double (two) tones ⇄

33
Example: Find and plot the FT of the following time domain function. Use the suitable FT
properties (specify the used properties for each case)

f(t) = cos(50pt) + 4cos(200pt)

Sol:
𝐅 𝛚 = 𝜋𝜹 𝜔 − 50𝜋 + 𝜋𝜹 𝜔 + 50𝜋 + 4𝜋𝜹 𝜔 − 200𝜋 + 4𝜋𝜹 𝜔 + 200𝜋 They all divided by two!

34
Example: Find the FT of the function x(t) shown below using rectangular property:
𝒕 𝝉
𝑨 𝒓𝒆𝒄𝒕 ↔ 𝑨𝝉𝒔𝒊𝒏𝒄( 𝝎)
𝝉 𝟐

Solution:

x(t) = x1(t) + x2(t)

x1(t) ⇄ X1(w) x2(t) ⇄ X2(w)
∴ X(w) = X1(w) + X2(w)
𝒕 𝟏𝟎 𝒕 𝟒
𝟐 𝒓𝒆𝒄𝒕 ↔ (𝟐)(𝟏𝟎)𝒔𝒊𝒏𝒄( 𝟐 𝝎) 𝟒 𝒓𝒆𝒄𝒕 ↔ (𝟒)(𝟒)𝒔𝒊𝒏𝒄(𝟐 𝝎)
𝟏𝟎 𝟒

𝑿𝟏 𝝎 = 𝟐𝟎𝒔𝒊𝒏𝒄(𝟓𝝎) 𝑿𝟐 𝝎 = 𝟏𝟔𝒔𝒊𝒏𝒄(𝟐𝝎)

𝑿 𝝎 = 𝟐𝟎𝒔𝒊𝒏𝒄 𝟓𝝎 + 𝟏𝟔𝒔𝒊𝒏𝒄(𝟐𝝎)
35
5. Differentiation:

𝒅𝒏 𝒇(𝒕)
⇌ (𝒋𝝎) 𝒏𝑭 𝝎 Short Notation:
𝒅𝒕𝒏
𝒅𝒇(𝒕)
= 𝒇′ (𝒕)
 1st derivative F 𝑓′ 𝑡 = 𝒋𝝎 𝑭 𝝎 𝒅𝒕

F [𝑓 ′ 𝑡 ] 𝒅𝟐 𝒇(𝒕)
then 𝐹 𝜔 = 𝟐
= 𝒇′′ 𝒕
𝑗𝜔 𝒅𝒕

 2nd derivative F 𝑓 ′′ 𝑡 = −𝝎𝟐 𝑭 𝝎

then F [𝑓 ′′ 𝑡 ]
𝐹 𝜔 =
−𝜔 2

36
6. Convolution in Time:
𝒙 𝒕 ∗𝒉 𝒕 ⇄ 𝑿 𝝎 𝑯(𝝎)

Convolution in Time Multiplication in Freq

Impulse
Response

Time domain x(t) h(t) y(t) = x(t) * h(t)

Frequency domain X(w) H(w) Y(w) = X(w) . H(w)

Frequency
Response (FR)
37
Example: The Fourier Transform of a time function h(t) is H(w)=(2cosw)(sin2w)/w. find h(0).
Solution: To find h(0), first, we have to find h(t) then at t=0 we can get h(0). h(t) ⇄ H(w)

2 cos 𝜔 sin(2𝜔) 𝒄𝒐𝒔𝜽 𝒔𝒊𝒏𝜷 =

𝟏
[𝒔𝒊𝒏 𝜽 + 𝜷 − 𝒔𝒊𝒏 𝜽 − 𝜷 ] 𝑨 𝒓𝒆𝒄𝒕
𝒕 𝝉
↔ 𝑨𝝉𝒔𝒊𝒏𝒄( 𝝎)
𝐻 𝜔 = 𝟐 𝝉 𝟐
𝜔
𝑓𝑜𝑟 𝑡ℎ𝑒 1𝑠𝑡 𝐻1 𝜔 = 𝑠𝑖𝑛𝑐(3𝜔)
sin 𝜔 + 2𝜔 − sin(𝜔 − 2𝜔) 𝜏
𝐻(ω) = 3𝜔 = 𝜔 → 𝜏 = 6
𝜔 2
sin 3𝜔 − sin(−𝜔) 𝒔𝒊𝒏 −𝜽 = −𝒔𝒊𝒏 𝜽 𝐴𝜏 = 3 → 𝐴 = 0.5
𝐻 𝜔 =
𝜔
ℎ1 𝑡 = 0.5 𝑟𝑒𝑐𝑡(𝑡ൗ6)
sin 3𝜔 + sin(𝜔)
𝐻 𝜔 =
𝜔 𝑓𝑜𝑟 𝑡ℎ𝑒 2𝑛𝑑 𝐻2 𝜔 = 𝑠𝑖𝑛𝑐(𝜔)
𝜏
sin 3𝜔 sin(𝜔) 𝜔= 𝜔 → 𝜏=2
2
𝐻 𝜔 = +
𝜔 𝜔 𝐴𝜏 = 1 → 𝐴 = 0.5
3sin 3𝜔 sin(𝜔) ℎ2 𝑡 = 0.5 𝑟𝑒𝑐𝑡(𝑡ൗ2)
𝐻 𝜔 = + = 3𝑠𝑖𝑛𝑐 3𝜔 + 𝑠𝑖𝑛𝑐 𝜔
3𝜔 𝜔
ℎ 0 = ℎ1 0 + ℎ2(0)
𝐻 𝜔 = 𝐻1 𝜔 + 𝐻2(𝜔)
𝒉 𝟎 = 𝟎. 𝟓 + 𝟎. 𝟓 = 𝟏
ℎ 𝑡 = ℎ1 𝑡 + ℎ2(𝑡) ↔ 𝐻1 𝜔 + 𝐻2(𝜔) 38
Example: For the following system, find the output function y(t). Given the mag. and
phase c/cs of its Frequency Response.
it's combinational of both.

x(t)=6sin(2t) H(w) y(t) = ?

|H(w)| ∠H(w) We have these two.

39
Solution:

Given the input as: x(t)=6sin(2t) → Asin(wot+f)

∴ A=6 , wo=2 , f=0
y(t) = A |H(w)| sin(wot + f + ∠H(w))

x(t)=6sin(2t) H(w) y(t)

|H(w)| ∠H(w)

From the given plots, |H(wo)|=|H(2)|=4 and ∠H(wo) = ∠H(2)=p/2

y(t) = (6) (4) sin(2t + p/2) = 24sin(2t + p/2) = 24cos(2t)

40
41
 Compare with example of slides #12-14

HW:
• Find FT of f(t) = 𝟒 𝒓𝒆𝒄𝒕 𝟐
𝒕−𝟑

• Find the FT of the function below:

42
43
 Remember :
𝟏
𝒔𝒊𝒏𝟐 𝜽 = [𝟏 − 𝒄𝒐𝒔 𝟐𝜽 ]
𝟐

∴ 𝟐𝒔𝒊𝒏𝟐 𝜽 = [𝟏 − 𝒄𝒐𝒔 𝟐𝜽 ]

44
45
46
47
Many problems are in Question Bank 2 (QB2)
48
 Some useful Trigonometric Identities
𝟏
𝒄𝒐𝒔𝜽 𝒄𝒐𝒔𝜷 = [𝒄𝒐𝒔 𝜽 + 𝜷 + 𝒄𝒐 𝒔 𝜽 − 𝜷 ]
𝟐
𝟏
𝒔𝒊𝒏𝜽 𝒔𝒊𝒏𝜷 = [𝒄𝒐𝒔 𝜽 − 𝜷 − 𝒄𝒐 𝒔 𝜽 + 𝜷 ]
𝟐
𝟏
𝒔𝒊𝒏𝜽 𝒄𝒐𝒔𝜷 = [𝒔𝒊𝒏 𝜽 + 𝜷 + 𝒔𝒊𝒏 𝜽 − 𝜷 ]
𝟐

𝟏
𝒄𝒐𝒔𝜽 𝒔𝒊𝒏𝜷 = [𝒔𝒊𝒏 𝜽 + 𝜷 − 𝒔𝒊𝒏 𝜽 − 𝜷 ]
𝟐
𝟏
𝒔𝒊𝒏𝟐 𝜽 = [𝟏 − 𝒄𝒐𝒔 𝟐𝜽 ]
𝟐
𝟏
𝒄𝒐𝒔𝟐 𝜽 = [𝟏 + 𝒄𝒐𝒔 𝟐𝜽 ]
𝟐
𝒔𝒊𝒏 −𝜽 = −𝒔𝒊𝒏 𝜽

𝒄𝒐𝒔 −𝜽 = 𝒄𝒐𝒔 𝜽

𝒔𝒊𝒏 𝟐𝜽 = 𝟐𝒄𝒐𝒔(𝜽)𝒔𝒊𝒏 𝜽

𝒔𝒊𝒏𝟐 𝜽 + 𝒄𝒐𝒔𝟐 𝜽 = 1

GIVEN IN EXAMS
49