The Fourier Transform

As we have seen, any (suciently smooth) function f(t) that is periodic can be built out of sins and
coss. We have also seen that complex exponentials may be used in place of sins and coss. We shall now
use complex exponentials because they lead to less writing and simpler computations, but yet can easily be
converted into sins and coss. If f(t) has period 2, its (complex) Fourier series expansion is
f(t) =

k=
c
k
e
ik

t
with c
k
=
1
2
_

f(t)e
ik

t
dt (1)
Not surprisingly, each term c
k
e
ik

t
in this expansion also has period 2, because c
k
e
ik

(t+2)
=
c
k
e
ik

t
e
i2k
= c
k
e
ik

t
. We now develop an expansion for non-periodic functions, by allowing complex
exponentials (or equivalently sins and coss) of all possible periods, not just 2, for some xed . So, from
now on, do not assume that f(t) is periodic.
For simplicity well only develop the expansions for functions that are zero for all suciently large |t|.
With a little more work, one can show that our conclusions apply to a much broader class of functions. Let
L > 0 be suciently large that f(t) = 0 for all |t| L. We can get a Fourier series expansion for the part
of f(t) with L < t < L by using the periodic extension trick. Dene F
L
(t) to be the unique function
determined by the requirements that
i) F
L
(t) = f(t) for L < t L
ii) F
L
(t) is periodic of period 2L
Then, for L < t < L,
f(t) = F
L
(t) =

k=
c
k
(L)e
ik

L
t
where c
k
(L) =
1
2L
_
L
L
f(t)e
ik

L
t
dt (2)
If we can somehow take the limit L , we will get a representation of f that is is valid for all ts , not
just those in some nite interval L < t < L. This is exactly what we shall do, by the simple expedient
of interpreting the sum in (2) as a Riemann sum approximation to a certain integral. For each integer k,
dene the k
th
frequency to be
k
= k

L
and denote by =

L
the spacing,
k+1

k
, between any two
successive frequencies. Also dene

f() =
_

f(t)e
it
dt. Since f(t) = 0 for all |t| L
c
k
(L) =
1
2L
_
L
L
f(t)e
ik

L
t
dt =
1
2L
_

f(t)e
i(k

L
)t
dt =
1
2L
_

f(t)e
i
k
t
dt =
1
2L

f(
k
) =
1
2

f(
k
)
In this notation,
f(t) = F
L
(t) =
1
2

k=

f(
k
)e
i
k
t
(3)
for any L < t < L. As we let L , the restriction L < t < L disappears, and the right hand
side converges exactly to the integral
1
2
_

f()e
it
d. To see this, cut the domain of integration into

y
y =
1
2

f()e
it
0

1

2

3

February 25, 2007 The Fourier Transform 1
small slices of width and approximate, as in the above gure, the area under the part of the graph of
1
2

f()e
it
with between
k
and
k
+ by the area of a rectangle of base and height
1
2

f(
k
)e
i
k
t
.
This approximates the integral
1
2
_

f()e
it
d by the sum
1
2

k=

f(
k
)e
i
k
t
. As L the
approximation gets better and better so that the sum approaches the integral. So taking the limit of (3) as
L gives
f(t) =
1
2
_

f()e
it
d where

f() =
_

f(t)e
it
dt (4)
The function

f is called the Fourier transform of f. It is to be thought of as the frequency prole of the
signal f(t).
Example 1 Suppose that a signal gets turned on at t = 0 and then decays exponentially, so that
f(t) =
_
e
at
if t 0
0 if t < 0
for some a > 0. The Fourier transform of this signal is

f() =
_

f(t)e
it
dt =
_

0
e
at
e
it
dt =
_

0
e
t(a+i)
dt =
e
t(a+i)
(a + i)

0
=
1
a + i
Since a + i has modulus

a
2
+
2
and phase tan
1
a
, we have that

f() = A()e
i()
with A() =
1
|a+i|
=
1

a
2
+
2
and () = tan
1
a
. The amplitude A() and phase () are plotted below.
A()

a a
()

a a
Example 2 Suppose that a signal consists of a single rectangular pulse of width 1 and height 1. Lets
say that it gets turned on at t =
1
2
and turned o at t =
1
2
. The standard name for this normalized
rectangular pulse is
rect(t) =
_
1 if
1
2
< t <
1
2
0 otherwise
t 1
2

1
2
1
It is also called a normalized boxcar function. The Fourier transform of this signal is

rect() =
_

rect(t) e
it
dt =
_
1/2
1/2
e
it
dt =
e
it
i

1/2
1/2
=
e
i/2
e
i/2
i
=
2

sin

2
February 25, 2007 The Fourier Transform 2
when = 0. When = 0,

rect(0) =
_
1/2
1/2
dt = 1. By lH opitals rule
lim
0

rect() = lim
0
2
sin

2

= lim
0
2
1
2
cos

2
1
= 1 =

rect(0)
so

rect() is continuous at = 0. There is a standard function called sinc that is dened
(1)
by sinc =
sin

. In this notation

rect() = sinc

2
. Here is a graph of

rect().

rect()

2 2
1
Properties of the Fourier Transform
Linearity
If and are any constants and we build a new function h(t) = f(t) +g(t) as a linear combination
of two old functions f(t) and g(t), then the Fourier transform of h is

h() =
_

h(t)e
it
dt =
_

_
f(t) + g(t)

e
it
dt =
_

f(t)e
it
dt +
_

g(t)e
it
dt
=

f() + g()
(L)
Time Shifting
Suppose that we build a new function h(t) = f(t t
0
) by time shifting a function f(t) by t
0
. The easy
way to check the direction of the shift is to note that if the original signal f(t) has a jump when its argument
t = a, then the new signal h(t) = f(t t
0
) has a jump when t t
0
= a, which is at t = a + t
0
, t
0
units to
the right of the original jump.
t
0
f(t) h(t) = f(t t
0
)
t
a
a + t
0
The Fourier transform of h is

h() =
_

h(t)e
it
dt =
_

f(t t
0
)e
it
dt =
_

f(t

)e
i(t

+t0)
dt

where t = t

+ t
0
, dt = dt

= e
it0
_

f(t

)e
it

dt

= e
it0
f()
(T)
(1)
There are actually two dierent commonly used denitions. The rst, which we shall use, is sinc =
sin

. The second is
sinc =
sin

. It is sometimes called the normalized sinc function.

February 25, 2007 The Fourier Transform 3
Scaling
If we build a new function h(t) = f
_
t

_
by scaling time by a factor of > 0, then the Fourier transform
of h is

h() =
_

h(t)e
it
dt =
_

f
_
t

_
e
it
dt =
_

f(t

)e
it

dt

where t = t

, dt = dt

=

f()
(S)
Example 3 Now consider a signal that consists of a single rectangular pulse of height H, width W and
centre C.
H
C
W
2
C +
W
2
y
y = r
HWC
(t)
t
The function r
HWC
(t) = H rect
_
tC
W
_
(with rect the normalized rectangular pulse of Example 2) has height
H and jumps when
tC
W
=
1
2
, i.e. t = C
1
2
W and so is the specied signal. By combining properties (L),
(T) and (S), we can determine the Fourier transform of r
HWC
(t) = H rect
_
tC
W
_
for any H, C and W. We
build it up in three steps.
First we consider the special case in which H = 1 and C = 0. Then we have R
1
(t) = rect
_
t
W
_
. So,
according to the scaling property (S) with
f(t) = rect(t), h(t) = R
1
(t) = rect
_
t
W
_
= f
_
t
W
_
= f
_
t

_
with = W
we have

R
1
() =

h() =

f() = W

rect(W) = W
2
W
sin
W
2
=
2

sin
W
2
Next we allow a nonzero C too and consider R
2
(t) = rect
_
tC
W
_
= R
1
(t C). By (T), with
f(t) = R
1
(t), h(t) = R
2
(t) = R
1
(t C) = f(t t
0
), with t
0
= C
the Fourier transform is

R
2
() =

h() = e
it0
f() = e
iC

R
1
() = We
iC

rect
_
W
_
Finally, by (L), with = H and = 0, the Fourier transform of r
HWC
= H rect
_
tC
W
_
= HR
2
(t) is
r
HWC
() = H

R
2
() = HWe
iC

rect(W) = HWe
iC 2
W
sin
W
2
= e
iC 2H

sin
W
2
Example 4
Now suppose that we have a signal that consists of a number of rectangular pulses of various heights
and widths. Here is an example
February 25, 2007 The Fourier Transform 4
2 1 0 1 2 3 4
1
2
y
y = s(t)
t
We can write this signal as a sum of rectangular pulses. As we saw in the last example, r
HWC
(t) = H rect
_
tC
W
_
is a single rectangular signal with height H, centre C and width W. So
s(t) = s
1
(t) + s
2
(t) + s
3
(t) where s
n
(t) = r
HWC
(t) with
_
H = 2, W = 1, C = 1.5 for n = 1
H = 1, W = 2, C = 1 for n = 2
H = 0.5, W = 2, C = 3 for n = 3
So, using (L) and r
HWC
() = e
iC 2H

sin
W
2
, which we derived in Example 3,
s() = s
1
() + s
2
() + s
3
() =
4

e
i
3
2

sin

2
+
2

e
i
sin +
1

e
i3
sin
Dierentiation
If we build a new function h(t) = f

(t) by dierentiating an old function f(t), then the Fourier transform

of h is

h() =
_

h(t)e
it
dt =
_

(t)e
it
dt
Now integrate by parts with u = e
it
and dv = f

(t) dt so that du = ie
it
dt and v = f(t). Assuming
that f() = 0, this gives

h() =
_

u dv = uv

v du =
_

f(t) (i)e
it
dt = i

f() (D)
Example 5 The dierentiation property is going to be useful when we use the Fourier transform to solve
dierential equations. As an example, lets take another look at the RLC circuit
+

x(t)
R L
C
+

y(t)
thinking of the voltage x(t) as an input signal and the voltage y(t) as an output signal. If we feed any input
signal x(t) into an RLC circuit, we get an output y(t) which obeys the dierential equation
LCy

(t) + RCy

(t) + y(t) = x(t) (5)

February 25, 2007 The Fourier Transform 5
You should be able to quickly derive this equation, which is also (16) in the notes Fourier Series, on your
own. Take the Fourier transform of this whole equation and use that
the Fourier transform of y

(t) is i y() and

the Fourier transform of y

(t) is i times the Fourier transform of y

(t) and so is
2
y()
So the Fourier transform of (5) is
LC
2
y() + iRC y() + y() = x()
This is trivially solved for
y() =
x()
LC
2
+iRC+1
(6)
which exhibits classic resonant behaviour. The circuit acts independently on each frequency component
of the signal. The amplitude | y()| of the frequency part of the output signal is the amplitude | x()| of
the frequency part of the input signal multiplied by A() =
1
|LC
2
+iRC+1|
=
1

(1LC
2
)
2
+R
2
C
2

2
.

A
We shall shortly learn how to convert (6) into an equation that expresses the time domain output signal y(t)
in terms of the time domain input signal x(t).
Example 6 The Fourier transform,

rect(), of the rectangular pulse function of Example 2 decays rather
slowly, like
1

for large . We can try suppressing large frequencies by eliminating the discontinuities at
t =
1
2
in rect(t). For example
g(t) =
_

_
0 if t
5
8
4(t +
5
8
) if
5
8
t
3
8
1 if
3
8
t
3
8
4(
5
8
t) if
3
8
t
5
8
0 if t
5
8
y = g(t)
t
y
3
8

3
8
5
8

5
8
1
It is not very dicult to evaluate the Fourier transform g() directly. But it easier to use properties (L)(D),
since
g

(t) =
_

_
0 if t
5
8
4 if
5
8
t
3
8
0 if
3
8
t
3
8
4 if
3
8
t
5
8
0 if t
5
8
_

_
= s
4
(t) + s
5
(t)
y = g

(t)
t
y
3
8

3
8
5
8

5
8
where s
n
(t) = r
HWC
(t) with
_
H = 4, W =
1
4
, C =
1
2
for n = 4
H = 4, W =
1
4
, C =
1
2
for n = 5
By Example 3, the Fourier transform of
s
4
() + s
5
() =
8

e
i/2
sin

8

8

e
i/2
sin

8
= 2i
8

sin

2
sin

8
February 25, 2007 The Fourier Transform 6
By Property (D), the Fourier transform of g

(t) is i times g(). So the Fourier transform of g(t) is

1
i
times
the Fourier transform of g

(t):
g() =
1
i
_
2i
8

sin

2
sin

8
_
=
16

2
sin

2
sin

8
This looks somewhat like the the Fourier transform in Example 2 but exhibits faster decay for large .
g()

2 2
1
Parsevals Relation
The energy carried by a signal f(t) is
_

|f(t)|
2
dt =
_

f(t)f(t) dt
Subbing in that
f(t) =
1
2
_

f()e
it
d =
1
2
_

f() e
it
d =
1
2
_

f()e
it
d
we have that
_

|f(t)|
2
dt =
1
2
_

f(t)

f()e
it
d dt =
1
2
_

f()
_ _

f(t)e
it
dt
_
d
=
1
2
_

f()

f() d =
1
2
_

f()|
2
d
This formula,
_

|f(t)|
2
dt =
1
2
_

f()|
2
d, is called Parsevals relation.
Duality
The duality property says that if we build a new timedomain function g(t) =

f(t) by exchanging the
roles of time and frequency, then the Fourier transform of g is
g() = 2f() (Du)
To verify this, just write down just the denition of g() and the Fourier inversion formula (4) for f(t) and,
in both integrals, make a change of variables so that the integration variable is s:
g() =
_

g(t)e
it
dt =
_

f(t)e
it
dt
t=s
=
_

f(s) e
is
ds
f(t) =
1
2
_

f()e
it
d
=s
=
1
2
_

f(s) e
ist
ds
So g(), which is given by the last integral of the rst line is exactly 2 times the last integral of the second
line with t replaced by , which is 2f().
February 25, 2007 The Fourier Transform 7
Example 7 In this example, we shall compute the Fourier transform of sinc(t) =
sin t
t
. Our starting point
is Example 2, where we saw that the Fourier transform of the rectangular pulse rect(t) of height one and
width one is

rect() =
2

sin

2
= sinc

2
. So, by duality (Du), with f(t) = rect(t) and g(t) =

f(t) =

rect(t),
the Fourier transform of g(t) = sinc
t
2
is g() = 2f() = 2 rect() = 2 rect(), since rect(t) is
even. So we now know that the Fourier transform of sinc
t
2
is 2 rect(t). To nd the Fourier transform
of sinc(t), we just need to scale the
1
2
out of
t
2
. So we apply the scaling property (S) with f(t) = sinc
t
2
and h(t) = sinc(t) = f(2t) = f
_
t

_
where =
1
2
. By (S), the Fourier transform of h(t) = sinc(t) is

f() = 2 rect() = rect
_

2
_
.
Multiplication and Convolution
A very common operation in signal processing is that of ltering. It is used to eliminate high frequency
noise and also to eliminate sixty cycle hum, arising from the ordinary household AC current. It is also
used to extract the signal from any one desired radio or TV station. In general, the lter is described
by a function

H() of frequency. For example you might have

H() = 1 for desired frequencies and

H() = 0 for undesirable frequencies. When a signal f(t) is fed into the lter, an output signal g(t), whose
Fourier transform is

f()

H() is produced. For example, the RLC circuit of Example 5 is a lter with

H() =
1
LC
2
+iRC+1
.
The question what is g(t)? remains. Of course it is the inverse Fourier transform
g(t) =
1
2
_

f()

H()e
it
d
of

f()

H(), but we would like to express it more directly in terms of the original timedomain signal f(t).
So lets substitute

f() =
_

f()e
i
d (which expresses

f() in terms f(t)) and see if we can simplify
the result.
g(t) =
1
2
_

f()e
i

H()e
it
dd
=
_

f()
_
1
2
_

H()e
i(t)
d
_
d
=
_

f()H(t ) d
The last integral is called a convolution integral and is denoted
(f H)(t) =
_

f()H(t ) d
If we make the change of variables = t

, d = d

we see that we can also express

(f H)(t) =
_

f(t

)H(

) (d

) =
_

f(t

)H(

) d

We conclude that
g() =

f()

H() =g(t) = (f H)(t) (C)
Example 8 The convolution integral gives a sort of moving weighted average, with the weighting determined
by H. Its value at time t involves the values of f for times near t. I say sort of because there is no
requirement that
_

H(t) dt = 1 or even that H(t) 0.

February 25, 2007 The Fourier Transform 8
Suppose for example, that we use the lter with

H() =
2

sin

2
The graph of this function was given in Example 2. It is a low pass lter in the sense that it lets through
most small frequencies and suppresses high frequencies. It is not really a practical lter, but I have chosen
it anyway because it is relatively easy to see the eect of this lter in tspace. From Example 2 we know
that
H(t) =
_
1 if
1
2
< t <
1
2
0 otherwise
So when this lter is applied to a signal f(t), the output at time t is
(f H)(t) =
_

f(t )H() d =
_ 1
2

1
2
f(t ) d =
_
t
1
2
t+
1
2
f(

) d

where

= t , d

= d
=
_
t+
1
2
t
1
2
f(

) d

which is the area under the part of the graph of f(

) from

= t
1
2
to

= t +
1
2
. To be concrete, suppose
that the input signal is
f(

) =
_
1 if

0
0 if

< 0

y
Then the output signal at time t is the area under the part of this graph from

= t
1
2
to

= t +
1
2
.
if t +
1
2
< 0, that is t <
1
2
, then f(

) = 0 for all t
1
2
<

< t +
1
2
(see the gure below) so that
(f H)(t) = 0
t
1
2
t+
1
2

y
y = f(

)
if t +
1
2
0 but t
1
2
0, that is
1
2
t
1
2
, then f(

) = 0 for t
1
2
<

< 0 and f(

) = 1 for
0 <

< t +
1
2
(see the gure below) so that (f H)(t) =
_
t+
1
2
t
1
2
f(

) d

=
_
t+
1
2
0
1 d

= t +
1
2
t
1
2
t+
1
2

y
y = f(

)
if t
1
2
> 0, that is t >
1
2
, then f(

) = 1 for all t
1
2
<

< t +
1
2
and (f H)(t) =
_
t+
1
2
t
1
2
f(

) d

=
_
t+
1
2
t
1
2
1 d

= 1.
t
1
2
t+
1
2

y
y = f(

)
All together, the output signal is
February 25, 2007 The Fourier Transform 9
(f H)(t) =
_

_
0 if t <
1
2
t +
1
2
if
1
2
t
1
2
1 if t >
1
2
y = (f H)(t)
t
y

1
2
1
2
Example 9
For a similar, but more complicated example, we can through the procedure of Example 8 with H(t)
still being rect(t) but with f(t) replaced by the more complicated signal s(t) of Example 4. This time the
output signal at time t is
(s H)(t) =
_

s(t )H() d =
_ 1
2

1
2
s(t ) d =
_
t+
1
2
t
1
2
s(

) d

where

= t
which is the area under the part of the graph of s(

) from

= t
1
2
to

= t +
1
2
. We can read o this
area from the gures below. In each gure, the left hand vertical dotted line is

= t
1
2
and the right hand
vertical dotted line is

= t +
1
2
. The rst gure has t so small that the entire interval t
1
2
<

< t +
1
2
is to
the left of

= 2, where the graph of s(

) starts. Then, in each successive gure, we increase t, moving

the interval to the right. In the nal gure t is large enough that the entire interval t
1
2
<

< t +
1
2
is to
the right of

= 4, where the graph of s(

) ends.
2 1 0 1 2 3 4
1
2
y
y = s(

Case: t +
1
2
< 2, that is t <
5
2
area = 0
t
1
2
t+
1
2 2 1 0 1 2 3 4
1
2
y
y = s(

Case: 2 t +
1
2
1, that is
5
2
t
3
2
area = 2((t +
1
2
) (2)) = 2t + 5
2 1 0 1 2 3 4
1
2
y
y = s(

Case: 1 t +
1
2
0, that is
3
2
t
1
2
area = 2((1) (t
1
2
)) = 1 2t
2 1 0 1 2 3 4
1
2
y
y = s(

Case: 0 t +
1
2
1, that is
1
2
t
1
2
area = t +
1
2
2 1 0 1 2 3 4
1
2
y
y = s(

Case: 1 t +
1
2
2, that is
1
2
t
3
2
area = 1
2 1 0 1 2 3 4
1
2
y

Case: 2 t +
1
2
3, that is
3
2
t
5
2
area = [2 (t
1
2
) +
1
2
[(t +
1
2
) 2] =
7
4

1
2
t
February 25, 2007 The Fourier Transform 10
2 1 0 1 2 3 4
1
2
y

Case: 3 t +
1
2
4, that is
5
2
t
7
2
area =
1
2
2 1 0 1 2 3 4
1
2
y

Case: 4 t +
1
2
5, that is
7
2
t
9
2
area =
1
2
[4 (t
1
2
)] =
9
4

1
2
t
2 1 0 1 2 3 4
1
2
y
y = s(

)
Case: t
1
2
> 4, that is t >
9
2
area = 0
t
1
2
t+
1
2
All together, the graph of the output signal is
2 1 0 1 2 3 4
1
2
(s H)(t)
t
Impulses
The inverse Fourier transform H(t) of

H() is called the impulse response function of the lter, because
it is the output generated when the input is an impulse at time 0. An impulse, usually denoted (t) (and
called a delta function) takes the value 0 for all times t = 0 and the value at time t = 0. In fact it is
so innite at time 0 that the area under its graph is exactly 1. Of course there isnt any such function, in
t
y
y = (t)

the usual sense of the word. But it is possible to generalize the concept of a function (to something called
a distribution or a generalized function) so as to accommodate delta functions. One generalization involves
February 25, 2007 The Fourier Transform 11
taking the limit as 0 of approximate delta functions like

(t) =
_
1
2
if < t <
0 otherwise
t
y
y =

(t)

1
2
A treatment of these generalization procedures is well beyond the scope of this course. Fortunately, in practice
it suces to be able to compute the value of the integral
_

(t)f(t) dt for any continuous function f(t)

and that is easy. Because (t) = 0 for all t = 0, (t)f(t) is the same as (t)f(0). (Both are zero for t = 0.)
Because f(0) is a constant, the area under (t)f(0) is f(0) times the area under (t), which we already said
is 1. So
_

(t)f(t) dt = f(0)
In particular, choosing f(t) = e
it
gives the Fourier transform of (t):

() =
_

(t)e
it
dt = e
it

t=0
= 1
By the time shifting property (T), the Fourier transform of (t t
0
) (where t
0
is a constant) is e
it0
. We
may come to the same conclusion by rst making the change of variables = t t
0
to give
_

(t t
0
)e
it
dt
=tt0
=
_

()e
i(+t0)
d
and then applying
_

()f() d = f(0) with f() = e

i(+t0)
.
Returning to the impulse response function, we can now verify that the output generated by a lter

H() in response to the impulse input signal (t) is indeed

( H)(t) =
_

()H(t ) d = H(t)
where we have applied
_

()f() d = f(0) with f() = H(t ).

Example 10 We saw in (6) that for an RLC circuit

H() =
1
LC
2
+iRC+1
To make the numbers work out cleanly, lets choose R = 1, L = 6 and C = 5. Then

H() =
1
6
2
+i5+1
=
1
(3i+1)(2i+1)
Recall from Example 1 that
f(t) =
_
e
at
if t 0
0 if t < 0


f() =
1
a+i
(7)
So we can determine the impulse response function H(t) for the RLC lter just by using partial fractions to
write H() as a linear combination of
1
a+i
s. Since

H() =
1
(3i + 1)(2i + 1)
=
a
3i + 1
+
b
2i+1
=
a(2i + 1) + b(3i + 1)
(3i + 1)(2i + 1)
=
(2a + 3b)i + (a + b)
(3i + 1)(2i + 1)
2a + 3b = 0, a + b = 1 b = 1 a, 2a + 3(1 a) = 0 a = 3, b = 2
February 25, 2007 The Fourier Transform 12
we have, using (7) with a =
1
3
and a =
1
2
,

H() =
3
3i + 1

2
2i + 1
=
1
1
3
+ i

1
1
2
+ i
H(t) =
_
e
t/3
e
t/2
if t 0
0 if t < 0
which has graph
t
y
y = H(t)
Example 10 (again) Here is a sneakier way to do the partial fraction expansion of Example 10. We know
that

H() has a partial fraction expansion of the form

H() =
1
(3i + 1)(2i + 1)
=
a
3i + 1
+
b
2i + 1
The fast way to determine a is to multiply both sides of this equation by (3i + 1)
1
2i + 1
= a +
b(3i + 1)
2i + 1
and then evaluate both sides at 3i +1 = 0. Then the b term becomes zero, because of the factor (3i +1)
in the numerator and we get
a =
1
2i + 1

i=
1
3
=
1
1
3
= 3
Similarly, multiplying by (2i + 1) rather than (3i + 1),
b =
1
3i + 1

2i+1=0
=
1
3i + 1

i=
1
2
=
1

1
2
= 2
February 25, 2007 The Fourier Transform 13
Properties of the Fourier Transform
Property Signal Fourier Transform
x(t) =
1
2
_

x()e
it
d x() =
_

x(t)e
it
dt
y(t) =
1
2
_

y()e
it
d y() =
_

y(t)e
it
dt
Linearity Ax(t) + By(t) A x() + B y()
Time shifting x(t t
0
) e
it0
x()
Frequency shifting e
i0t
x(t) x(
0
)
Scaling x
_
t

_
|| x()
Time shift & scaling x
_
tt0

) ||e
it0
x()
Frequency shift & scaling ||e
i0t
x(t) x
_
0

_
Conjugation x(t) x()
Time reversal x(t) x()
tDierentiation x

(t) i x()
x
(n)
(t) (i)
n
x()
Dierentiation tx(t) i
d
d
x()
t
n
x(t)
_
i
d
d
_
n
x()
Convolution
_

x()y(t ) d x() y()

Multiplication x(t)y(t)
1
2
_

x() y( ) d
Duality x(t) 2x()
Parseval
_

|x(t)|
2
dt =
1
2
_

| x()|
2
d
e
at
u(t) =
_
0 if t < 0,
e
at
if t > 0
1
a + i
(a constant, Re a > 0)
e
at
u(t) =
_
e
at
if t < 0,
0 if t > 0

1
a + i
(a constant, Re a < 0)
e
a|t|
2a
a
2
+
2
(a constant, Re a > 0)
Boxcar in time rect(t) =
_
1 if |t| <
1
2
0 if |t| >
1
2
sinc
_

2
_
=
2

sin
_

2
_
General boxcar r
HWC
(t) =
_
H if |t C| <
W
2
0 if |t C| >
W
2
HWe
iC
sinc
_
W
2
_
= e
iC 2H

sin
W
2
Boxcar in frequency
1
2
sinc
_
t
2
_
=
1
t
sin
_
t
2
_
rect() =
_
1 if || < 1/2
0 if || > 1/2
Impulse in time (t t
0
) e
it0
(t t
0
) x(t) e
it0
x(t
0
)
Single frequency e
i0t
2 (
0
)
February 25, 2007 The Fourier Transform 14