Lecture 05 Anotated

Download as pdf or txt
Download as pdf or txt
You are on page 1of 34

Lecture 5

Signals and Systems

Prof. Young-Seok Choi

Dept. Electronics and Communications Eng.


Kwangwoon University

1 / 31
Table of Contents

1. Continuous-time Fourier transform (CTFT)


2. from CTFS to CTFT
3. relation to Fourier series
4. examples of CTFT
5. properties of CTFT

2 / 31
From CTFS

consider a rectangular pulse train

Z T1 /2
1 jk!0 t 2 sin(k!0 T1 ) sin(k!0 T1 )
ak = e dt = = , k 6= 0
T T1 /2 k!0 T k⇡
2T1
while a0 = T

3 / 31
Tak =

To
ZinkwoT)

=/who
a function of w
From CTFS (cont,)
-
im =
X (jn)

Tak
X(jw) /w
=

kno
=

4 / 31
From CTFS (cont.)

T= D


=> O
wo-F
=

5 / 31
Conditions for CTFT

continuous-time Fourier transform: infinite-length (or aperiodic) and


continuous-time signals can be represented as an infinite-length discrete-frequency
signals via continuous-time Fourier transform (CTFT)
1. aperiodic signal x(t) ! periodic extension of x(t)
2. assuming that x(t) is repeated every T seconds and denoting it as x̃(t)
3. if we take the limit as T ! 1, we obtain a precise model of an aperiodic signal
for which all rules that govern periodic signals can be applied

6 / 31
Development of CTFT
the derivation of Fourier transform consists of 3 steps:
step 1: We assume that an aperiodic signal x(t) has finite duration, i.e., x(t) = 0 for
|t| > T /2, for some T . Since x(t) is aperiodic, we first construct a periodic signal x̃(t):
x̃(t) = x(t)
for T /2 < t < T /2, and x̃(t + T ) = x̃(t)

7 / 31
Development of CTFT (Cont.)

step 2: Since x̃(t) is periodic, we may express x̃(t) using Fourier series:
1
X
x̃(t) = ak ejk!0 t
k= 1

where Z
1 jk!0 t
ak = x̃(t)e dt
T T

8 / 31
Development of CTFT (cont.)
The Fourier series coefficients ak can further be calculated as
Z
1 T /2
ak = x̃(t)e jk!0 t dt
T T /2
Z
1 T /2
= x(t)e jk!0 t dt, * x̃(t) = x(t), for T /2 < t < T /2
T T /2
Z

3
1 1
= x(t)e jk!0 t dt, * x(t) = 0, for |t| > T /2
T 1

if we define
=
X(jkwo) Z 1
j!t
X(j!) = x(t)e dt
1
then it holds that
1
ak = X(jk!0 )
T
9 / 31
Development of CTFT (cont.)
P1 jk!0 t
consequently, substituting this into x̃(t) = k= 1 ak e
1
X 1
X
1 jk!0 t 1
x̃(t) = X(jk!0 )e = X(jk!0 )ejk!0 t !0
T 2⇡
k= 1 k= 1
step 3: Now, note that x̃(t) is the periodic padded version of x(t). When the period
T ! 1, the periodic signal x̃(t) approaches x(t). Therefore, x̃(t) ! x(t) as T ! 1
Moreover, when T ! 1, or equivalently !0 ! 0, the limit of the sum becomes an
integral as
1
X Z 1
1 1
x(t) = lim x̃(t) = lim X(jk!0 )ejk!0 t !0 = X(j!)ej!t d!
!0 !0 !0 !0 2⇡ 1 2⇡
k= 1
by combining above two equations, we have
Z 1
1
x(t) = X(j!)ej!t d!
2⇡ 1
10 / 31
Definition of CTFT

Theorem
The Fourier transform X(j!) of a signal x(t) is given by
Z 1
Yeset)] = X(j!) =
1
x(t)e j!t dt (analysis equation)

and the inverse Fourier transform is given by

Xigni
Z 1
1
= x(t) = X(j!)ej!t d! (synthesis equation)
2⇡ 1

the two equations are known as the Fourier Transform pair. The first is called the
analysis equation (because we are analyzing the time signal in the Fourier domain) and
the second is called the synthesis equation (because we are gathering the Fourier
domain information and reconstruct the time signal).

11 / 31
Idea of Fourier Transform

12 / 31
Relation to CTFS
What is the di↵erence between Fourier Series and Fourier Transform? Let us apply
Fourier Transform to the following two types of signals.
I aperiodic signal: the Fourier transform of an aperiodic signal is the limiting case
(when !0 ! 0) of applying Fourier series analysis on the periodically padded
version of the aperiodic signal. Fourier transform can be applied to both periodic
and aperiodic signals, whereas Fourier series analysis can be applied th periodic
signals.
I periodic signal: if the signal x(t) is periodic, then we do not need to construct
x̃(t) and set !0 ! 0. In fact, !0 is fixed by the period of the signal: If the period
of x(t) is T0 , then !0 = 2⇡/T . Now since x(t) is periodic, we can apply Fourier
series analysis to x(t) and get
1
X
x(t) = ak ejk!0 t
k= 1

13 / 31
Relation to CTFS (cont.)
if we further apply Fourier transform to the previous equation, then we have
Z 1" X 1
# 1
X Z 1
jk!0 t j!t
X(j!) = ak e e dt = ak ejk!0 t e j!t dt
1 k= 1 k= 1 1
-
1
X
= ak (2⇡) (! k!0 )
=
etf(n-kwo)
k= 1
here,
Z 1
1 1
F {2⇡ (! k!0 )} = 2⇡ (! k!0 )ej!t d!
2⇡ 1
Z 1
= (! k!0 )ej!t d! = ejk!0 t
1
therefore, we showed that the Fourier Transform of a periodic signal is a train of
impulses with amplitude defined by the Fourier Series coefficient (and scaled by a
factor of 2⇡).
14 / 31
=
Examples of CTFT

I x(t) = e at u(t) a>0


EFT exists
10 pltdt<
sol) X(jw) =?

x(r)
State
=

15 / 31
Xij =

a =
(x(jn))ejkx(n)

=
X

4 xijn)
=

-tan()
=> (xigo)) (x ( -jw))
=
kX() = tan" ()
-
=
fant
even
symmetry
=
=

(X(-ja)
odd
symmetry
Examples of CTFT (Cont.)

I x(t) = (t)
50
%Set t = -1
.

see)
Xljn) =

:
sifting property

>
-
si
t
t
16 / 31
Examples of CTFT (cont.)

I
Let's find( X(jw) .

1, |t| < T1
x(t) =
0, |t| > T1

soe) Xig =
Wat =

-
-
-
- 17 / 31
Examples of CTFT (cont.)

I
Let's find ((t).
(
1, |!| < W
X(j!) =
0, |!| > W

soIt) =We

=We
=NEW
· 18 / 31
The sinc function

sinc(✓) =
sin ⇡✓
⇡✓
if on kit
,
k=1,
2
,
...

Sinc(O) =

/cros ing
& 000000

19 / 31
20 / 31
CTFT for periodic signals
let us consider a signal x(t) with Fourier transform X(j!) that is a single impulse of
area 2⇡ at ! = !0
X(j!) = 2⇡ (! !0 )
apply the inverse Fourier transform
Z 1
1
x(t) = 2⇡ (! !0 )ej!t d! = ej!0 t
2⇡ 1
generally, if X(j!) is a linear combination of impulses equally spaced in frequency
1
X
X(j!) = 2⇡ak (! k!0 )
k= 1

then
1
X
x(t) = ak ejk!0 t
k= 1

21 / 31
an eskrot zitarf(w-kwo)

↓ linearity
CARS
Examples of CTFT of periodic signals

1. ( 2.
1
X
1, |t| < T1 x(t) = (t kT )
x(t) =
0, T1 < |t| < T /2 k= 1

=w
an-
Ar

22 / 31
Examples of CTFT of periodic signals (Cont.)

S
= jejrotWt
etwot
1. x(t) = sin !0 t 2. x(t) = cos !0 t
I
t
it a di
=
=

=
a, , ,

X(jw) 2π[ *f(w wo) + f(w wo)]


-
+ -

X(jw) ent[ + f(w-wa]


=

= -
S(w +wo) +
=
iS(c + wo) +f (w -wo)
- f(w wo)
+ S( wo)
-

+
=

X(jr) NT

X(yw)
↑ ↑
wo t
I w
-
wo
> W

-
w w

23 / 31
Examples of CTFT of periodic signals (Cont.)

24 / 31
Properties of CTFT

the properties of Fourier Transform is very similar to those of Fourier Series


F F
I linearlity: if x1 (t) ! X1 (j!) and x2 (t) ! X2 (j!), then
F
ax1 (t) + bx2 (t) ! aX1 (j!) + bX2 (j!)

F j!t0 X(j!)
I time shifting: x(t t0 ) !e

F
I conjugation: x⇤ (t) ! X ⇤ ( j!)

25 / 31
Properties of CTFT (cont.)

I di↵erentiation and integration


d F
x(t) ! j!X(j!)
dt
Z t
F 1
x(⌧ )d⌧ ! X(j!) + ⇡X(0) (!)
1 j!

I time scaling ✓ ◆
F 1 j!
x(at) ! X
|a| a

26 / 31
Properties of CTFT (cont.)

((t))2 :
energy per
unit
time :
energy per
unit frequency
I Parseval’s relation
Z 1 Z 1
2 1
|x(t)| dt = |X(j!)|2 d!
1 2⇡ 1

F
I duality: if x(t) ! X(j!), then if another signal y(t) has the shape of X(j!),
we can deduce that Y (j!) will have the shape of x(t)

27 / 31
Properties of CTFT (cont.)

I duality (cont.):

28 / 31
Properties of the CTFT (cont.)
I convolution:
F
y(t) = h(t) ⇤ x(t) ! Y (j!) = H(j!)X(j!)

Proof: R1
y(t) = h(t) ⇤ x(t) = 1 x(⌧ )h(t ⌧ )d⌧ , let us find Y (j!)
Z 1 Z 1
Y (j!) = F{y(t)} = x(⌧ )h(t ⌧ )d⌧ e j!t dt
1 1
Z 1 Z 1
= x(⌧ ) h(t ⌧ )e j!t dt d⌧

=>g[ht-21] :timeshif
1 1
Z 1
= x(⌧ )e j!⌧ H(j!)d⌧
1
Z 1
= H(j!) x(⌧ )e j!⌧ d⌧ = H(j!)X(j!)
1
29 / 31
Properties of CTFT (cont.)
I multiplication (or modulation):
F 1
r(t) = s(t)p(t) ! R(j!) = [S(j!) ⇤ P (j!)]
2⇡ Z
1
1
= S(j✓)P (j(! ✓))d✓
2⇡ 1
* Duality Je
x(t) = h(t) X(ji) H(jc)

Le
s(t)
p(t) #Tsijn) p(jw)]
=

30 / 31
Properties of CTFT (cont.)
Suspi, sign)
Plymout
*

e
ex.) p(t) = cos !0 t
then

P (j!) = ⇡ (! !0 ) + ⇡ (! + !0 )

1
R(j!) = [S(j!) ⇤ P (j!)]
2⇡
1 1
= S(j(! !0 )) + S(j(! + !0 ))
2 2

31 / 31

You might also like