Fourier Transform

Rachaen H.
EEE, BRACU.
Intro to Fourier Transform
Recall: Fourier Series represents a periodic signal as a sum of sinusoids
or complex sinusoids e jk0t
Note: Because the FS uses “harmonically related” frequencies k0, it can only create
periodic signals

Q: Can we modify the FS idea to handle non-periodic signals?

A: Yes!!

What about x(t )  c e
k  
k
j k t
?
With arbitrary discrete frequencies…
NOT harmonically related
This will give some non-periodic signals but
not all signals of interest!!
The problem with this is that it cannot include all possible frequencies!
No matter how close we try to choose the discrete frequencies k there are
always some left out of the sum!!!
We need some way to include ALL frequencies!!
2/14
How about: 
1
x(t ) 
2 

X ( )e jt d Yes… this will work for any
practical non-periodic signal!!

Called the “Fourier Plays the Plays the role of

Integral” also, more jk0t
commonly, called the
role of ck e
“Inverse Fourier Integral replaces sum because it can “add up
Transform” over the continuum of frequencies”!

Okay… given x(t) how do we get X()?


X ( )   x(t )e  jt dt Called the
 “Fourier Transform”
of x(t)

Note: X() is complex-valued function of   (-, )

Need to use two plots

|X()| X ( ) to show it
3/14
Comparison of FT and FS
Fourier Series: Used for periodic signals
Fourier Transform: Used for non-periodic signals (although we
will see later that it can also be used for periodic signals)
Synthesis Analysis

Fourier
Series
x(t )  c e
n  
k
jk0t 1 t 0 T
ck   x(t )e  jk0t dt
T t0
Fourier Series Fourier Coefficients
Fourier 1  
x (t )  
jt
X ( )e d X ( )   x(t )e  jt dt
Transform 2  
Inverse Fourier Transform Fourier Transform

FS coefficients ck are a complex-valued function of integer k

FT X() is a complex-valued function of the variable   (-, )
4/14
Synthesis Viewpoints:

FS: x(t )  k
c e jk0t

n  

|ck| shows how much there is of the signal at frequency k0

ck shows how much phase shift is needed at frequency k0
We need two plots to show these

1 
FT: x(t ) 
2 

X ( )e jt d

|X()| shows how much there is in the signal at frequency 

 X() shows how much phase shift is needed at frequency 

We need two plots to show these

5/14
 Alternate Notations
1. x(t)  X()

2. X() = F{x(t)}

 F{ } is an “operator on”
x(t) to give X()

3. x(t)= F
–1{X()}

 F –1{}is an “operator on”

X() to give x(t)

Analogy: Looking at X() is “like” looking at an x-ray of the signal- in the sense that an
x-ray lets you see what is inside the object… shows what stuff it is made from.
In this sense: X() shows what is “inside” the signal – it shows how much of each complex
sinusoid is “inside” the signal

Note: x(t) completely determines X() There are some advanced mathematical issues
that can be hurled at these comments… we’ll not
X() completely determines x(t) worry about them 6/14
Example: FT of a Rectangular pulse
 = pulse width
p (t )
t
Given: a rectangular pulse signal p(t)
 

2 2

Find: P()… the FT of p(t)

Recall: we use this symbol to
indicate a rectangular pulse
Note the Notational Convention: lower- with width 
case for time signal and corresponding
upper-case for its FT

Solution: (Here we’ll directly do the integral… but later we’ll use the “FT Table”)
Note that
  
1 ,   t 
p ( t )   2 2
 0 , otherwise


7/14
Now apply the definition of the FT: Limit integral to
 /2 where p(t) is non-

P ( )   p (t )e  jt dt   dt
e  jt
zero… and use the
 fact that it is 1 over
 / 2
that region

  j 2 j



j
e 
 1  jt
2
 
2 e  e 2
 j2


Artificially
inserted 2 in
2
  numerator and
denominator

  
 sin   Use Euler’s
 2  Formula

   sin goes up and down

2 sin  between -1 and 1

P ( )   2 
1/ decays down as || gets
 big… this causes the overall
function to decay down
8/14
For this case the FT is real valued so we can plot it using a single plot
(shown in solid blue here):

 = 1/2

2/
2/

-2/
-2/

  
2 sin 
P ( )   2  The
Thesine
sinwiggles
wigglesup
up&down
“between ±2/”
down


9/14
Now… let’s think about how to make a magnitude/phase plot…
Even though this FT is real-valued we can still plot it using magnitude
and phase plots:
We can view any real number as a complex
number that has zero as its imaginary part

A positive real number R will have:

Im |R| = R R = 0

R Re

A negative real number R will have:

Im |R| = -R R = ±
+
Can use
R - Re either one!!

10/14
 Applying these Ideas to the Real-valued FT P()
Phase = 0

Phase = ±

Here I have chosen - to display odd symmetry 11/14

Definition of “Sinc” Function   
2 sin 
The result we just found had this mathematical form: P ( )   2 

This structure shows up enough that we define a special function to capture it:

sin(x )
Define: sinc ( x ) 
x
Plot of sinc(x)
Note that sinc(0) = 0/0.
So… Why is sinc(0) = 1?

It follows from
L’Hopital’s Rule

12/14
With a little manipulation we can re-write the FT result for a pulse in terms of the
sinc function:
Recall:
sin(x ) Need  times
sinc ( x ) 
x something…

          Now we need the

2 sin  2 sin   2 sin   same thing down
 2    2   2 
P ( )  here as inside the
   sine…

     
2 sin   sin  
  2    2     sinc  
  
2    
  2 
2 2

  
P ( )   sinc  
 2 
13/14
FT of Rect. Pulse = Sinc Function

 = pulse width p (t )

Time-Domain View
  t

2 2

Frequency-Domain View

Tells what amplitude is

needed at each frequency

Tells what phase is

needed at each frequency

14/14
FT Example: Decaying Exponential
Given a signal x(t) = e-btu(t) find X(ω) if b > 0

Solution: First see what x(t) looks like:

x(t )
1
b controls decay rate
t What does this look
like if b < 0???

The u(t) part forces this to zero

Now…apply the definition of the Fourier transform. Recall the general

form:
∞
X (ω ) = ∫ x (t )e − jωt dt
−∞

9/27
Now plug in for our signal:
∞ ∞ ∞
X (ω ) = ∫ e u (t )e
−bt − jωt
dt = ∫ e e − bt − jωt
dt = ∫ e −(b + jω ) t dt
−∞ 0 0

integrand = 0 for t < 0 Set lower limit to 0 Easy

due to the u(t) and then u(t) = 1 over
integral!
integration range

t =∞
⎡ − 1 −( b+ jω ) t ⎤
=⎢
ω
e ⎥ =
−1
e [
−( b + jω ) ∞
− e −( b + jω ) 0
]
⎣ b + j ⎦ t =0 b + jω

− 1 ⎡ −b∞ − jω ∞ 0 ⎤ −1
= ⎢eN e
− eN ⎥ = [0 − 1]
b + jω ⎢⎣ =0 mag =1 =1 ⎥⎦ b + jω

1 Only if b>0… what

= happens if b<0
b + jω
10/27
Summary of FT Result for Decaying Exponential

1
x (t ) = e u (t )− bt
X (ω ) =
b + jω
For b > 0
(Complex Valued)

1
X (ω ) = Magnitude
b +ω
2 2

⎛ω ⎞
∠X (ω ) = − tan ⎜ ⎟ Phase
−1

⎝b⎠

x (t ) = e − bt u(t ) X (ω )
1 b > 0 controls
decay rate

t ω
11/27
 1
Time Signal 10
Fourier Transform
Exploring
b=0.1 b=0.1 Effect of

|X( ω)|
decay rate b
x(t)

0.5 5

on the
0 0
Fourier
-10 0 10 20 30 40 -100 -50 0 50 100 Transform’s
t (sec) ω (rad/sec)
1 1 Shape
b=1 b=1

|X(ω )|
x(t)

0.5 0.5

0 0
-10 0 10 20 30 40 -100 -50 0 50 100
t (sec) ω (rad/sec)
1 0.1

b=10 b=10
|X( ω )|
x(t)

0.5 0.05

0 0
-10 0 10 20 30 40 -100 -50 0 50 100
t (sec) ω (rad/sec)

Note: As b increases… Short Signals have FTs that spread

1. Decay rate in time signal increases more into High Frequencies!!!
2. High frequencies in Fourier transform are more prominent. 13/27
Effect of Pulse Width on the FT Pτ(ω)
2
1

τ=2 τ=2

|X( ω )|
x(t)

0.5 1

0 0
-4 -3 -2 -1 0 1 2 3 4 -100 -50 0 50 100
t (sec) 1 ω (rad/sec)
1

τ=1 τ=1

|X( ω )|
x(t)

0.5 0.5

0 0
-4 -3 -2 -1 0 1 2 3 4 -100 -50 0 50 100
t (sec) 1 ω (rad/sec)
1

τ = 1/2 τ = 1/2
|X( ω )|
x(t)

0.5 0.5

0 0
-4 -3 -2 -1 0 1 2 3 4 -100 -50 0 50 100
t (sec) ω (rad/sec)

Note: As width decreases, FT is more widely spread

Î Narrow pulses “take up more frequency range”
19/27
For your FT Table you should spend time making sketches of the entries
… like this:

X (ω)

t
ω

Pτ (ω)
pτ (t)
t
τ τ
−
2 2
ω

23/27
Bandlimited and Timelimited Signals
Now that we have the FT as a tool to analyze signals, we can use it to identify
certain characteristics that many practical signals have.
A signal x(t) is timelimited (or of finite duration) if there are 2 numbers T1 & T2
such that:
x(t ) = 0 ∀t ∉ [T1 , T2 ]

T1 T2 t
A (real-valued) signal x(t) is bandlimited if there is a number B such that

X (ω ) = 0 ∀ω > 2πB

X (ω ) 2πB is in rad/sec
B is in Hz
− 2πB 2πB ω
Recall: If x(t) is real-valued then |X(ω)| has “even symmetry”
24/27
FACT: A signal can not be both timelimited and bandlimited
⇒ Any timelimited signal is not bandlimited
⇒ Any bandlimited signal is not timelimited
Note: All practical signals must “start” & “stop”
⇒ timelimited ⇒ Practical signals are not bandlimited!

But… engineers say practical signals are effectively bandlimited

because for almost all practical signals |X(ω)| decays to zero as
ω gets large
Recall: sinc decays as 1/ω

FT of pulse X (ω ) Some application-specific level

ω that specifies “small enough to
be negligible”
− 2πB 2πB

This signal is effectively bandlimited to B Hz because |X(ω)| falls

below (and stays below) the specified level for all ω above 2πB
25/27
Bandwidth (Effective Bandwidth) Abbreviate Bandwidth as “BW”
For a lot of signals – like audio – they fill up the lower frequencies but then decay
as ω gets large:
X (ω )
ω Signals like this are
called “lowpass” signals
− 2πB 2πB

We say the signal’s BW = B in Hz if there is “negligible” content for |ω| > 2πB

Must specify what

“negligible” means
For Example:
1. High-Fidelity Audio signals have an accepted BW of about 20 kHz
2. A speech signal on a phone line has a BW of about 4 kHz

Early telephone engineers determined that limiting speech to a BW

of 4kHz still allowed listeners to understand the speech
26/27
For other kinds of signals – like “radio frequency (RF)” signals – they
are concentrated at high frequencies
Signals like this are
X (ω ) called “bandpass” signals
ω
− ω2 − ω1 ω1 = 2πf1 ω2 = 2πf 2

If the signal’s FT has negligible content for |ω| ∉ [ω1, ω2]

then we say the signals BW = f2 - f1 in Hz

For Example:
1. The signal transmitted by an FM station has a BW of 200 kHz = 0.2 MHz
a. The station at 90.5 MHz on the “FM Dial” must ensure that its signal
does not extend outside the range [90.4, 90.6] MHz
b. Note that: FM stations all have an odd digit after the decimal point.
This ensures that adjacent bands don’t overlap:
i. FM90.5 covers [90.4, 90.6]
ii. FM90.7 covers [90.6, 90.8], etc.
2. The signal transmitted by an AM station has a BW of 20 kHz
a. A station at 1640 kHz must keep its signal in [1630, 1650] kHz
b. AM stations have an even digit in the tens place and a zero in the ones
27/27
Table of Common Fourier Transform Results
We have just found the FT for a common signal…
  
1,  2  t  2   
p ( t )   P (  )   sinc  
 0 , otherwise  2 


We derived that result by directly applying the integral form of the FT to the given
signal equation.
For the common “textbook” signals this has already been done… and the results are
available in tables published in books and on-line
You should study the table provided…
• If you encounter a time signal or FT that is on this table you should recognize that it
is on the table without being told that it is there.
• You should be able to recognize entries in graphical form as well as in equation form.
• Later we’ll learn about some “FT properties” that will expand your ability to apply
these entries on the FT Table
In the real-world, engineers use these table results to understand basic ideas and
concepts and to think through how things work in principle!

So… next we’ll look at some of the more important entries in the table provided…
2/8
Decaying Exponential
As we’ll see later… this signal naturally occurs in lots of real-world places!

1
x (t )  e u (t ) bt
X ( ) 
b  j
For b > 0
(Complex Valued)

1
X ( )  Magnitude
b 
2 2

 
X ( )   tan   Phase
1

b

x (t )  e  bt u(t ) X ( )
1 b > 0 controls
decay rate

t 
3/8
 Can Use Matlab to Make Plots of FT Results
FT of e-btu(t)
(volts)

Technically V/Hz Note that

for b = 10 magnitude
plot has even
symmetry

Note that
phase plot has
odd symmetry

True for every

real-valued signal
MATLAB Commands to Compute FT Plotting Commands
w=-100:0.2:100; subplot(2,1,1); plot(w,abs(X))
b=10; xlabel('Frequency \omega (rad/sec)')
X=1./(b+j*w); ylabel('|X(\omega|) (volts)'); grid
subplot(2,1,2); plot(w,angle(X))
xlabel('Frequency \omega (rad/sec)')
ylabel('<X(\omega) (rad)'); grid 4/8
 Effect of Exp. Decay Rate b on FT Magnitude
Time Signal FT Magnitude
 bt 1
x (t )  e u (t ) X ( ) 
( ) ( b 2)   2
1 1

b=1 b=1

|X( )|
x(t)

0.5 0.5

0 0
-10 0 10 20 30 40 -100 -50 0 50 100
t (sec)  (rad/sec)
1 0.1

b=10 b=10
|X(  )|
x(t)

0.5 0.05

0 0
-10 0 10 20 30 40 -100 -50 0 50 100
t (sec)  (rad/sec)

Note: As b increases… Short Signals have FTs that spread

1. Decay rate in time signal increases more into High Frequencies!!!
2. High frequencies in Fourier transform are more prominent. 5/8
Some Important Signals & Their FTs (see Table for More!)

1,    t   2 ( )

u(t )  ( )  1 / j

 0.5  u(t ) 1 / j

 (t ) 1,      

cos(o t )   (  o )   (  o )

sin(o t ) j  (  o )   (  o )

e jot 2 (  o ), o real

6/8
FT of Periodic Signal
Note that we have now used the FT to analyze cosine and sine… which are
PERIODIC signals!!! Before we used the Fourier Series to analyze
periodic signals… Now we see that we can also use the Fourier Transform!

If x(t) is periodic then we can write the FS of it as: x(t )  k
c e jk0t

k  

  jk0t 
Now we can take the FT of both sides of this: Fx (t )  F   ck e 
k   

 

FT of a Periodic Signal  k
c F e jk0t

 k  

X ( )   2c  (  k )
k  
k 0 2 (  k0 )

Note: the FT is a bunch of delta functions

with “weights” given by the FS coefficients!

7/8
 X ( ) / 2
c2 c1 c0 c1 c2
c4 c3 c3 c4 
 40  30  20  0 0 20 30 40
So the FT of a periodic signal is zero except at multiples of the
fundamental frequency 0, where you get impulses.

We call these spikes “Spectral Lines”

Note that if we start with the Amplitude-Phase Trig form we end up

with the same result for the FT

x (t )  A0   Ak cos  k0t   k 
k 1

For each cosine term we get two deltas (a positive frequency & negative
frequency):
cos(ot   )   e  j  (  o )  e j  (  o ) 

8/8
 Fourier Transform Properties
These properties are useful for two main things:
1. They help you apply the table to a wider class of signals
2. They are often the key to understanding how the FT can be used in a given
application.
So… even though these results may at first seem like “just boring math” they are
important tools that let signal processing engineers understand how to build
things like cell phones, radars, mp3 processing, etc.
Here… we will only cover the most important properties.
See the available table for the complete list of properties!
In this note set we simply learn these most-important properties… in the next note
set we’ll see how to use them.

2/15
1. Linearity (Supremely Important) Gets used virtually all the time!!

If x(t )  X ( ) & y (t )  Y ( )

then ax (t )  by (t )  aX ( )  bY ( )

Another way to write this property:

Fax (t )  by (t )  aFx (t ) bFy (t )

Use Defn
 

To see why: F ax (t )  by (t ) 

 jt
ax ( t )  by ( t ) e dt of FT

 
 a  x(t )e  j t
dt  b  y (t )e  jt dt
 
By standard
Property of  X ( )  Y ( )
Integral of sum
of functions By Defn of
FT
3/15
2. Time Shift (Really Important!) Used often to understand practical
issues that arise in audio,
communications, radar, etc.

If x (t )  X ( ) then x (t  c )  X ( )e  jc
Note: If c > 0 then x(t – c) is a delay of x(t)

So… what does this mean??

 jc
First… it does nothing to the magnitude of the FT: X ( )e  X ( )
That means that a shift doesn’t change “how much” we need of each of
the sinusoids we build with
 jc
Second… it does change the phase of the FT: { X ( )e }  X ( )  e  jc

 X ( )  c
Line of slope –c This gets added
Phase shift increases linearly to original phase
as the frequency increases

Shift of Time Signal  “Linear” Phase Shift of Frequency Components

4/15
3. Time Scaling (Important)
Q: If x(t )  X ( ) , then x ( at )  ??? for a0

1  
A: x(at )  X 
a a

If the time signal is Then… The FT is

Time Scaled by a Freq. Scaled by 1/a

An interesting “duality”!!!

5/15
To explore this FT property…first, what does x(at) look like?

x(t )
Original
Signal t
1 2 3 4 5 6 7

x(2t )
Time-Scaled
|a| > 1 “squishes” horizontally
w/ a = 2 3.5 t

1 
x t  |a| < 1 “stretches” horizontally
Time-Scaled 2 
w/ a = 1/2 t

1 2 3 4 5 6 7 8 9 10 11 12 13 14

|a| > 1 makes it “wiggle” faster  need more high frequencies

|a| < 1 makes it “wiggle” slower  need less high frequencies
6/15
 1  
When |a| > 1  |1/a| < 1 x(at )  X 
a a
Time Signal is Squished FT is Stretched Horizontally
and Reduced Vertically

Original Signal & Its FT

x(t )
A X  
t 

Squished Signal & Its FT

x(2t )
1  
X 
2 2

.5A
t

7/15
When |a| < 1  |1/a| > 1 1  
x(at )  X  
a a
Time Signal is Stretched
FT is Squished Horizontally
and Increased Vertically

x(t ) Original Signal & Its FT

A X  
t 
1 
x t  Stretched Signal & Its FT 2A 2X 2 
2 
t 

Rough Rule of Thumb we can extract from this property:

 Duration   Bandwidth
 Duration   Bandwidth
Very Short Signals tend to take up Wide Bandwidth 8/15
4. Time Reversal (Special case of time scaling: a = –1)

x(t )  X ( )
double conjugate
 
 dt   x(t )e  jt dt
 j (  ) t
Note: X ( )  x(t )e = “No Change”
 

 Conjugate changes to –j
  x(t )e  jt dt


= x(t) if x(t) is real


  x (t )e  jt dt  X ( ) Recall: conjugation
 doesn’t change abs.
value but negates the
angle
So if x(t) is real, then we get the special case:
X ( )  X ( )
x ( t )  X ( )  X ( )  X ( )
9/15
 Essential for understanding
5. Modulation Property Super important!!! practical issues that arise in
communications, radar, etc.
There are two forms of the modulation property…
1. Complex Exponential Modulation … simpler mathematics, doesn’t
directly describe real-world cases
2. Real Sinusoid Modulation… mathematics a bit more complicated,
directly describes real-world cases

Euler’s formula connects the two… so you often can use the Complex
Exponential form to analyze real-world cases

Complex Exponential Modulation Property: x(t )e j0t  X (  0 )

Multiply signal by a Shift the FT
complex sinusoid in frequency

X   F{x (t )e jot }  X   o 
A A

 
o 10/15
Real Sinusoid Modulation
Based on Euler, Linearity property, & the Complex Exp. Modulation Property
Euler’s Formula
1

Fx (t ) cos(0t )  F  x (t )e  x (t )e
j0t  j0t 
 
 2  Linearity of FT
1
   
 F x (t )e j0t  F x (t )e  j0t
2

Comp. Exp. Mod.
 X (  o )  X (  o )
1
2

The Result: x (t ) cos(0t ) 

1
X (  0 )  X (  0 )
2
Shift Down Shift Up

Related Result: x (t ) sin(0t ) 

j
X (  0 )  X (  0 )
2

Exercise: x (t ) cos(0t  0 )  ?? 11/15

Visualizing the Result x (t ) cos(0t ) 
1
X (  0 )  X (  0 )
2
Shift up Shift down

X ( )


Shift Down Shift Up

Fx (t ) cos(0t )


 0 0
Interesting… This tells us how to move a signal’s spectrum up to higher
frequencies without changing the shape of the spectrum!!!
What is that good for??? Well… only high frequencies will radiate from an
antenna and propagate as electromagnetic waves and then induce a signal in a
receiving antenna…. 12/15
6. Convolution Property (The Most Important FT Property!!!)

y (t )   x ( )h(t   )d  Y ( )  X ( ) H ( )


We will not yet discuss the “Convolution” aspect of this now… but we
will talk about it in depth later.
In the next Note Set we will explore the real-world use of the right side
of this result!

7. Parseval’s Theorem (Recall Parseval’s Theorem for FS!)

 1 
 x(t ) dt   X ( ) d
2 2
 2 

Energy computed in time domain Energy computed in frequency domain

2 d
x(t ) dt X ( )
2

2
= energy at time t = energy at freq. 
13/15
 
8. Duality: X ( )   x(t )e  jt dt


x (t ) X ( )
1 
x(t ) 
2 

X ( )e jt d

Both FT & IFT are pretty much the “same machine”: c  f ( )e  j d


So if there is a “time-to-frequency” property we would expect a

virtually similar “frequency-to-time” property

Illustration: Delay Property: x(t  c)  X ( )e  jc

j 0 t
Modulation Property: x(t )e  X (  0 )
Other Dual Properties: (Multiply by tn) vs. (Diff. in time domain)
(Convolution) vs. (Mult. of signals)
14/15
Also, this duality structure gives FT pairs that show duality.
Suppose we have a FT table that a FT Pair A… we can get the dual
Pair B using the general Duality Property:
1. Take the FT side of (known) Pair A and replace  by t and move it
to the time-domain side of the table of the (unknown) Pair B.
2. Take the time-domain side of the (known) Pair A and replace t by
–, multiply by 2, and then move it to the FT side of the table of
the (unknown) Pair B.

Here is an example… We found the FT pair for the pulse signal:

  
p (t )  sinc  
Pair A
 2  
Here we have used the
Step 1 Step 2 fact that p(-) = p()

 t 
Pair B  sinc   2 p ( )
 2 

15/15
 FT Examples
Example:
Find FT of x(t) given below: x(t )
A
2 t
-2
−A
Solution:
Note: Ap2 (t ) So : x (t ) = Ap2 (t + 1) − Ap2 (t − 1)
A

Use Linearity : F{x (t )} = AF{p2 (t + 1)}− AF{p2 (t − 1)}

Use Time Shift : F{p2 (t + 1)} = P2 (ω )e jω
F{p2 (t − 1)} = P2 (ω )e − jω
 [
So... X (ω ) = AP2 (ω ) e jω − e − jω ]
⎡ e jω − e − jω ⎤
= AP2 (ω )2 j ⎢ ⎥
⎣ 2j ⎦
= A2 j sin(ω ) P2 (ω )
⎛ω ⎞
From Table : P2 (ω ) = 2sinc⎜ ⎟
⎝π ⎠
⎛ω ⎞
X (ω ) = A4 j sin(ω )sinc⎜ ⎟
⎝π ⎠
Example B x(t )
Find FT of x(t) t
−2 2
Solution #1

1 p2 (t )
Note: x (t ) = ( p2 (t ) ∗ p2 (t ) )
B
Verify it!!
t 2
−1 1
B 2
Using Convolution Property: X (ω ) = P2 (ω )
2
⎛ 2ω ⎞ ⎛ω ⎞
From Table: P2 (ω ) = 2sinc⎜ ⎟ = 2sinc⎜ ⎟
⎝ 2π ⎠ ⎝π ⎠
⎛ω ⎞
So... X (ω ) = 2 Bsinc ⎜ ⎟
2

⎝π ⎠
 dx(t )
Solution #2 dt B
dx(t ) 2
Take to get 2 t
dt −2 B
−
2
dx 2 (t )
Take another derivative to get: B B
dt
2 2 t
−2 2
−B

Now by Linearity and F{δ (t )} & time shift:

⎧ dx 2 (t ) ⎫ B
F⎨ ⎬ = F {δ ( t + 2 )} − B F {δ ( t )} +
B
F{δ (t − 2)}
⎩ dt ⎭ 2 2

⎡ 1 jω 2 1 − jω 2 ⎤
= B⎢ e −1+ e ⎥ = B (cos( 2ω ) − 1) = −2 B sin 2
(ω )
⎣2 2 ⎦
Euler! ⇒ cos(ω2)
Now by derivative property:

⎧ d 2 x (t ) ⎫ − ⎧ 2
x (t ) ⎫
⎬ = (1
jω ) X (ω ) ⇒ X (ω ) = 2 F ⎨
2 1 d
F⎨ ⎬
23 ω ⎩ dt ⎭
⎩ dt ⎭ = −ω 2

⇒ X (ω ) =
−1
ω2
[− 2 B sin (ω )]
2

2
⎡ ⎛ ω ⎞⎤
⎢ sin⎜ π ⎟ ⎥
⎡ sin(ω ) ⎤
2
⎝ π⎠
= 2B⎢ ⎥ = 2 B ⎢ ⎥
⎣ ω ⎦ ⎢ π⋅ ω ⎥
⎢⎣ π ⎥⎦
⎛ω ⎞
X (ω ) = 2 Bsinc ⎜ ⎟ Same Result!!
2

⎝π ⎠
Recall that FT Properties can be used for:
1. Expanding use of the FT table
2. Understanding real-world concepts

Example Application of “Linearity of FT”: Suppose we need to find the FT

of the following signal…
2 x(t )
1 t
 2 1 1 2
We don’t see this on our table… so we should think brainstorm ways to use FT
properties to tackle it…

• One way is to break x(t) down into a sum of signals on our table!!!

2/12
 Break a complicated signal down into simple signals before finding FT:

p4 (t )
1
t
2 x(t )
2 2 Add 1
p2 (t ) to get t
1 2 2
t
1 1

Mathematically we write: x(t )  p4 (t )  p2 (t ) X ( )  P4 ( )  P2 ( )

From FT Table we have a known result for the FT of a pulse, so…

 2   
X ( )  4 sinc    2 sinc 
    
3/12
Application of Modulation Property to Radio Communication
FT theory tells us what we need to do to make a simple radio system… then
electronics can be built to perform the operations that the FT theory calls for:

Sound and Transmitter (Tx) antenna

microphone Modulator
x (t )
amp multiply amp
cos(0t )
oscillator

FT of Message Signal
Fx (t ) cos(0t )
X ( )

 0 0 
Choose f0 > 10 kHz to enable efficient radiation (with 0 = 2f0 )
AM Radio: around 1 MHz FM Radio: around 100 MHz
Cell Phones: around 900 MHz, around 1.8 GHz, around 1.9 GHz etc.
4/12
 The next several slides show how these ideas are used to make a receiver:

Receiver (Rx) Speaker

De-Modulator
Amp & Amp &
multiply
Filter Filter
Signals from Other cos(0t )
Transmitters
oscillator
Signals from Other
Transmitters

 0 0 

The “Filter” removes the Other signals Here we “imagine” the

(We’ll learn about filters later) use of some RLC circuit
to suppress the undesired
frequency bands:
Y() = H() X()
 0 0 
5/12
 Receiver Speaker
De-Modulator
Amp &
multiply
 Filter
cos(0t )
 0 0
oscillator

By the Real-Sinusoid Modulation Property… the De-Modulator shifts up & down:

Shifted Up

 2 0  0 0 2 0
Shifted Down

 2 0  0 0 2 0

Add… gives double


 2 0  0 0 2 0
6/12
 Receiver Speaker
De-Modulator
Amp & Amp &
multiply
Filter Filter
cos(0t )
oscillator
Extra Stuff we don’t want


 2 0  0 0 2 0
The “Filter” removes the Extra Stuff


 2 0  0 0 2 0

Speaker is driven by desired message signal!!!

7/12
So… what have we seen in this example:
Using the Modulation property of the FT we saw…

1. Key Operation at Transmitter is up-shifting the message spectrum:

a) FT Modulation Property tells the theory then we can build…
b) “modulator” = oscillator and a multiplier circuit
2. Key Operation at Transmitter is down-shifting the received spectrum
a) FT Modulation Property tells the theory then we can build…
b) “de-modulator” = oscillator and a multiplier circuit
c) But… the FT modulation property theory also shows that we need
filters to get rid of “extra spectrum” stuff
i. So… one thing we still need to figure out is how to deal with
these filters…
ii. Filters are a specific “system” and we still have a lot to learn
about Systems…
iii. That is the subject of much of the rest of this course!!!
8/12
Fourier Transform Table
Time Signal Fourier Transform
1, − ∞ < t < ∞ 2πδ (ω )
− 0.5 + u(t ) 1 / jω
u(t ) πδ (ω ) + 1 / jω
δ (t ) 1, − ∞ < ω < ∞
δ (t − c), c real e − jωc , c real
e − bt u(t ), b > 0 1
, b>0
jω + b
e jωot , ωo real 2πδ (ω − ωo ), ωo real
pτ (t ) τ sinc[τω / 2π ]
τ sinc[τ t / 2π ] 2πpτ (ω )
[1 − ] p (t )
2t
τ τ
τ
2 sinc2 [τ ω / 4π ]
τ
2 sinc 2 [τ t / 4π ] [
2π 1 −
2ω
τ ] p (ω )
τ

cos(ωo t ) π [δ (ω + ωo ) + δ (ω − ωo )]
cos(ωo t + θ ) π [e − jθ δ (ω + ωo ) + e jθ δ (ω − ωo )]
sin(ωo t ) jπ [δ (ω + ωo ) − δ (ω − ωo )]
sin(ωo t + θ ) jπ ⎡⎣ e − jθ δ (ω + ωo ) − e jθ δ (ω − ωo ) ⎤⎦
Fourier Transform Properties
Property Name Property
Linearity ax(t ) + bv(t ) aX (ω ) + bV (ω )
Time Shift x (t − c ) e − jωc X (ω )
Time Scaling x(at ), a ≠ 0 a X (ω / a ),
1
a≠0
Time Reversal x ( −t ) X ( −ω )

X (ω ) if x (t ) is real
n
Multiply by t t x (t ), n = 1, 2, 3, …
n
dn
j n
X (ω ), n = 1, 2, 3, …
dω n
Multiply by Complex Exponential e jωo t x(t ), ωo real X (ω − ωo ), ωo real
sin(ωo t ) x (t )
Multiply by Sine j
[X (ω + ωo ) − X (ω − ωo )]
2
cos(ωo t ) x (t )
Multiply by Cosine 1
[X (ω + ωo ) + X (ω − ωo )]
2
Time Differentiation dn ( jω ) n X (ω ), n = 1, 2, 3, …
x (t ), n = 1, 2, 3, …
dt n
Time Integration t 1
X (ω ) + πX (0)δ (ω )
∫ x ( λ ) dλ
−∞
jω
Convolution in Time x (t ) * h (t ) X (ω ) H (ω )
Multiplication in Time x (t ) w(t ) 1
X (ω ) *W (ω )
2π
Parseval’s Theorem (General) ∞ ∞
1
∫ x(t )v(t )dt =
−∞
2π ∫ X (ω )V (ω )dω
−∞

Parseval’s Theorem (Energy) ∞ ∞

1
∫−∞x (t )dt = 2π ∫ X (ω ) dω if x (t ) is real
2 2

−∞

∞ ∞
1
∫−∞ x(t ) dt = 2π ∫ X (ω ) dω
2 2

−∞

Duality: If x(t) ↔ X(ω) X (t ) 2πx ( −ω )