300 Signals and Systems

The magnitude of X(0) is given by

|X(o) =x¡ (o)' +X, (o?

and the phase of X(0) is given by

X(0) = tan -1 X,(w)

Xp (0)
The plot of |X(0)| versus o is known as amplitude spectrum, and the plot of X(@)
versus ois known as phase spectrum. The amplitude spectrum and phase spectrum together
is called frequency spectrum.

5.4 EXISTENCE OF FOURIER TRANSFORMS

The Fourier transform does not exist for all aperiodic functions. The conditions for a function
x() to have Fourier transform, called Dirichlet's conditions, are:
1. x() is absolutely integrable over the interval -oo to oo, that is
Tx() dt <o
2. x() has a finite number of discontinuities in every finite time interval. Further, each
of these discontinuities must be finite.
3. x() has a finite number of maxima and minima in every finite time interval.
Almost all the signals that we come across in physical problems satisfy all the above
conditions except possibly the absolute integrability condition.
Dirichlet' s condition is a sufficient condition but not necessary condition. This means,
Fourier transform will definitely exist for functions which satisfy these conditions. On the
other hand, in some cases, Fourier transform can be found with the use of impulses even for
functions like step function, sinusoidal function, etc. which do not satisfy the convergence
condition.

5.5 FOURIER TRANSFORMS OF STANDARD SIGNALS

5.5.1 Impulse Function (t)
|1 for t=0
Given x() = 8(), S() =0 for t#0

/Then X(0)=() e jo dt = 8) e jan d:

e jOt
-0 =1
 Fourier Transforns 301

F[()] 1 ot) FT’1

Hence, the Fourier transform of a unit impulse function is unity.

|X(0)) =1 for all O

X
(@) =0 for all w

The impulse function with its magnitude and phase spectra are shown in Figure 5.2.
4 |X(@) |
4 x() 1
4|X(0)

t |0
(a) (b) (c)
Figure 5.2 (a) Impulse function, (b) Its magnitude spectrum, (c) Its phase spectrum.

Similarly, F[`t -,))= 8t -to) eio dt =e ) i.e. 8(-t) Fe

5.5.2 Single-sided Real Exponential Function eu(t)
1 for t>0
Given x() = e u(t), 0 for t<0

Then X(0) = | x()ea dt =e u(t) e j dt

e-(atjw) e0
-a ejor di=eetjo" dt = <at jo) <a+ jo)
0-1
<a+ jo) at jo
1
F[e"ut)]= Or -at ut) "
a+ jo a+ jo
1 a-jo
Now, X(0) =
a+ jo (a+ jo)(a-j0)
a- jo 1 -1 0
2 tan
a + a'

|X(o)) =Ja+ o X(0) =-tan

a
for all o
 302 Signals and Systems

Figure 5.3 shows the single-sided exponential function with its magnitúde and phase spectra.
4x(1) = eu(t)
4|X(0)| 4|X(0)
1/a

+/2

-/2

(a) (b) (c)

Figure 5.3 (a) One-sided exponential function, (b) Its amplitude spectrum, and (c) Its phase spectrum.
5.5.3 Double-sided Real Exponential Function ealt|
Given x(t) = e-|
lera-) =e for t<0
x()=e-al|
for t>0

=ea-) ul-t) + e u(t)

=eu(-t) +e"ult)

X(0) = x() e-i dt

0

= [ee jar dt + ea ejor dt = ela-jo dt+etatjo)" dt

e(a-jo); o0 ,-(a+jo):
-[ee-jo)" dt +e -(a+j@)N dt = -(a- jo) +
-<a+ jo)
Jo
e -
-p e-e 1 1 2a
+
-(a- ja) -(a + jo) a- jo a+ jo a+o
2a 2a
F(eal)= or
a2 + o 2

2a
|X(0) =a +o'
for all O

and |X(0) = 0 for all w

Atwo-sided exponential function and its amplitude and phase spectra are shown in
Figure 5.4.
 Fourier Transforms 303

4 x()=eal| 4LX(0)
4|X(0) |
2/a

0
(a) (b) (c)
Figure 5.4 (a) Two-sided exponential function, (b) Its magnitude spectrum, and (c) Its phase spectrum.

5.5.4 Complex Exponential Function

To find the Fourier transform of complex exponential function e, consider finding the
inverse Fourier transform of S(@ - an). Let
X(0) = S(o - a)
1
x(t) =F[X(0)]=F[S(O- =X(O)eja
)) dw
2x

1
do =
27 2

F[ö0- )]= or F(2roo- a)]=el

21

F[e= 2ro(W- w)
ja l2r8(0- @)

5.5.5 Constant Amplitude (1)

So
Since x() = 1is not absolutely integrable, we cannot find its Fourier transform directly.
8(o).
the Fourier transform of x(t) = 1is determined thrOugh inverse Fourier transform of
We know that
1
x(t)=
277
X
(0)e do
1 for w=0
Consider X(0) = Ô(o) S 0) =
|0 for 00

x() =FX(0)]=F[B(O))
1
dw=1)s
21 2 21
304 Signals and Systems

F6(0) =2n Or F|2rö 0)]=1

F|I]= 2Ö( 0)
or

Similarly. F[A]= 2nAl 0)

Alternate method
Let x(t) = 1

X(0) =FI]=()ejan,dt

We have e dt=2nð(x)
-o

F[1]= 2zÑ(0)
Another alternate method
Let x(t) = 1
The waveform of a constant function is shown in Figure 5.5. Let us consider a small
section of the constant function, say, of duration t. If we extend the small duration to infinity,
we will get back the original function. Therefore,
x(t) = Lt

|0 -T/2 t/2
(a) (b)
Figure 5.5 (a) Constant function, (b) Small section of the constant function.

where
for
ss
elsewhere
 Fourier Transforms
By definition, the Fourier transform of x(t) is:

X(@) =F[x()] = F| Lt rect

t/2
e
-ju |t/2
= Lt | (1)eja dt = Lt
T’oo
-r/2 t ’ - j o J-/2
eJOtI2)jo(/2) 2 sin[oTl2)1|
= Lt = Lt
-jo
sin[ot/2)1| cl2
= Lt
t/2) =227 L sa o
t/2
Using the sampling property of the delta function Lt Sa we get

X(0)= F Lt rect l=2ralo)

5.5.6 Signum Function sgn(t)
The signum function is denoted by sgn (t) and is defined by
for t> 0
sgn()
for t <0

This function is not absolutely integrable. So we cannot directly find its Fourier transform.
Therefore, let us consider the function e sgn(t) and substitute the limit a ’0 to obtain the
above sgn ().
Given x(t) = sgn(t) = a-’0
Lt esgn(t) = a-’0
Lt [e"u(t)- eu(-)]

X(0) = F[sgn()]= || |a-’0

o

=L
a-’0
e e jal u(t) dt-e"e jar'u-) dt
-o0

= Lt
a’0
e-tatjo)r dt-e- jo dt= Lt eta* jax,dt-ela- ja)t dt
e-(atjo)r
= Lt -
e0 e0
a-’0 -(a+ jo) Jo |-(a- jo) a’o-(a+ jo) -{a- j0) |
1 1 1 1 2
= Lt
a’0 a+ jo a- jo jo -jaO jo
 306 Signals and Systems
2
F[sgn()]=
jo
2
or |Sgn(1) T
jw

2 and X(@)= for @<0 and

for o>0
2

Figure 5.6 shows the signum function and its magnitude and phase spectra.
4sgn (1) 4| X(@)| 4X(@)

-r/2

(a) (b) (c)

Figure 5.6 (a) Signum function, (b) Its amplitude spectrunm, (c) Its phase spectrum.

5.5.7 Unit Step Function u(t)

The unit step function is defined by
for >0
u1)
for t <0

Since the unit step function is not absolutely integrable, we cannot directly find its Fourier
transform. So express the unit step function in terms of signum function as:

u(?)= ,Sgní)
Given x() =u() =l+sgn()]
X(e) =Fla0)|=Fl+sgnt1
1
-;{FI]+F[sgn(c)1}
We know that

2
F[1]= 2zÑ(o) and F[sgn(t)]=
jo
2
= TÒl0) +
jo jo
 Fourier Transforms 307

F [u()] = NÐ(0) +
jo
FT 1
or u(t) ’ nÒlO) +
jo

X(0)l = o at o = 0 and is equal to 0 at w = -oo and o = o,

Figure 5.7 shows the unit step function and its spectrum.
4 x(t) = u(t) 4 X(0)
t0.5 sgn(()
0.5 + 0.5 sgn(1)
1

0.5

-0.5

(a) (b) (c)

Figure 5.7 (a) 0.5 sgn (), (b) u(t), (c) Spectrum of w(t).

5.5.8 Rectangular Pulse (Gate pulse)

is
Consider a rectangular pulse as shown in Figure 5.8. This is called a unit gate function and
defined as:
4 II(t/)

1.0

-TI2 tI2 t

Figure 5.8 Unit gate function.

x() = rect
for st/2
otherwise

T/2 e j a t/2
= ()e j dt =
-t/2 -jo
J-l2
308 Signals and Systems

e J0(T2) jo(t/2) elo(th2)-pj0(tl 2)

-jo jo
ejo(t/2) -p jo(r/2) sin o(t/2)
o(tI2) 2j oltI2)
=t sinc
2

= TSa
2

: that is

rect T sinc
2

Figure 5.9 shows the spectra of the gate function.

| X(0)|

2 27

4T
Main lobe
(a
4X(0)

2r 4n 6 87

8r 6 4

(b)

Figure 5.9 (a) Amplitude spectrum, and (b) Phase spectrum of II(/).
The amplitude spectrum is obtained as follows:
At o =0, sinc (wt/2) = 1. Therefore. X(@)l at o =0 is equal to t. At (or2) = tnt,
i.e. at
2nn
W=t -,n=1, 2,..., sinc
 Fourier Transforms 309

The phase spectrum is:

X(0) =0 if sinc(ot/2) > 0
=tr if sinc(otl2) <0
The amplitude response between the first two zero crossings is known as main lobe and the
portions of the response for o <{2t) and > (2/ are known as side lobes. From the
amplitude spectrum, we can find that majority of the energy of the signal is contained in the
main lobe. The first zero crossing occurs at o = (2r/) or at f= (1/) Hz. As the width of the
rectangular pulse is made longer, the main lobe becomes narrower. The phase spectrum is
odd function of 0. If the amplitude spectrum is positive, then phase is zero, and if the
amplitude spectrum is negative, then the phase is -t or n.

5.5.9 Triangular Pulse A

Consider the triangular pulse as shown in Figure 5.10. It is defined as:

-T/2 t/2

Figure 5.10 Triangular function.

1
-e-| t/2 for

xt)= - 1
-1-4) for
0<t<
elsewhere

2||
i.e. as
otherwise

X() =F[x0)] =F -ja dt

0 T/2
2t

-t/2
1+=e -jot dt + 0
dt
310 Signals and Systems

t/2 t/2
dt
0

t/2 t/2 t/2

T/2 21
-fema-faje dt + ejo dt -
0

t/2 T/2
|[e+e Ja] dt- | [eo +e jdt
t/2
=|2cos ot dt 2tcos at dt
0 0
7/2 t/2
sin ot Sin ot sin ot
2 dt
Jo Jo

sin t Sin cOS Ot

=2 +
Jo

4 4
=
si SIn COS
ot 2 2
4 2sin20r
1- cos
2 4
2 sin? (orI4)
Sinc?
(ori4) 2 4

sinc2
4

or A
2 4

Figure 5.11 shows the ampitude spectrum of a triangular pulse.

4|X(@)|

-8r/t 4rlt 0 4t/t 8r/t

Figure 5.11 Amplitude spectrum of a triangular pulse.

 Fourier Transforms 311

5.5.10 Cosine Wave cOS aht

Given x(t) = COS h!

Then X(@)= F[x(t)]=F(cos agi]=F(e te a)

-F(elay+ F(e ay=2nô(0-a,)+ 2nô(w +a,))
2
= n[ö(0+ )+ o(0- a,)]
FT
F[cos ob = T[o(o + a)+ o(o- a)] or cos W,tt
Figure 5.12 shows the cosine wave and its amplitude and phase spectra.
4| X(0) | 4LX(0)
A

(b) (c)
(a)
phase spectrum.
Figure 5.12 (a) Cosine wave, (b) Its magnitude speetrum, (c) Its

5.5.11 Sine Wave sin aht

x() =sin h!
Given
eJop-jy!
Then F[x()] =X(@) =F[sin a]=F 2j

-Fe-F[e =(2nöa
2j
- )

=-jn[ö(o- a)- o+,)]

-jn[S (o- a) + o(0+ ab) or sin Wi
FT
’-jn[öo- a,)- w+ on)
F[sin oh] =
amplitude and phase spectra.
Figure 5.13 shows the sine wave signal and its
4 x(1)= sin o,i |X(0)| 4LX(0)

(a) (b) (c)

spectrum.
Figure 5.13 (a) Sinusoidal signal, (b) Its amplitude spectrum, (c) Its phase