EE1 and ISE1 Communications I

Pier Luigi Dragotti

Lecture four
Lecture Aims
Trigonometric Fourier series
Fourier spectrum
Exponential Fourier series
1
Trigonometric Fourier series
Consider a signal set
{1, cos
0
t, cos 2
0
t, ..., cos n
0
t, ..., sin
0
t, sin2
0
t, ..., sinn
0
t, ...}
A sinusoid of frequency n
0
is called the n
th
harmonic of the sinusoid, where
n is an integer.
The sinusoid of frequency
0
is called the fundamental harmonic.
This set is orthogonal over an interval of duration T
0
= 2/
0
, which is the
period of the fundamental.
2
Trigonometric Fourier series
The components of the set {1, cos
0
t, cos 2
0
t, ..., cos n
0
t, ..., sin
0
t, sin 2
0
t, ..., sin n
0
t, ...}
are orthogonal as

T
0
cos n
0
t cos m
0
tdt =

0 m = n
T
0
2
m = n = 0

T
0
sinn
0
t sinm
0
tdt =

0 m = n
T
0
2
m = n = 0

T
0
sinn
0
t cos m
0
tdt = 0 for all m and n

T
0
means integral over an interval from t = t
1
to t = t
1
+T
0
for any value of t
1
.
3
Trigonometric Fourier series
This set is also complete in T
0
. That is, any signal in an interval t
1
t t
1
+T
0
can be written as the sum of sinusoids. Or
g(t) = a
0
+ a
1
cos
0
t + a
2
cos 2
0
t + ... + b
1
sin
0
t + b
2
sin2
0
t + ...
= a
0
+

n=1
a
n
cos n
0
t + b
n
sinn
0
t
Series coecients
a
n
=
g(t), cos n
0
t
cos n
0
t, cos n
0
t
b
n
=
g(t), sinn
0
t
sinn
0
t, sinn
0
t
4
Trigonometric Fourier Coecients
Therefore
a
n
=

t
1
+T
0
t
1
g(t) cos n
0
tdt

t
1
+T
0
t
1
cos
2
n
0
tdt
As

t
1
+T
0
t
1
cos
2
n
0
tdt = T
0
/2,

t
1
+T
0
t
1
sin
2
n
0
tdt = T
0
/2.
We get
a
0
=
1
T
0

t
1
+T
0
t
1
g(t)dt
a
n
=
2
T
0

t
1
+T
0
t
1
g(t) cos n
0
tdt n = 1, 2, 3, ...
b
n
=
2
T
0

t
1
+T
0
t
1
g(t) sinn
0
tdt n = 1, 2, 3, ...
5
Compact Fourier series
Using the identity
a
n
cos n
0
t + b
n
sinn
0
t = C
n
cos(n
0
t +
n
)
where
C
n
=

a
2
n
+ b
2
n

n
= tan
1
(b
n
/a
n
).
The trigonometric Fourier series can be expressed in compact form as
g(t) = C
0
+

n=1
C
n
cos(n
0
t +
n
) t
1
t t
1
+ T
0
.
For consistency, we have denoted a
0
by C
0
.
6
Periodicity of the Trigonometric series
We have seen that an arbitrary signal g(t) may be expressed as a trigonometric Fourier series
over any interval of T
0
seconds.
What happens to the Trigonometric Fourier series outside this interval?
Answer: The Fourier series is periodic of period T
0
(the period of the fundamental harmonic).
Proof:
(t) = C
0
+

n=1
C
n
cos(n
0
t +
n
) for all t
and
(t + T
0
) = C
0
+

n=1
C
n
cos[n
0
(t + T
0
) +
n
]
= C
0
+

n=1
C
n
cos(n
0
t + 2n +
n
)
= C
0
+

n=1
C
n
cos(n
0
t +
n
)
= (t) for all t
7
Properties of trigonometric series
The trigonometric Fourier series is a periodic function of period T
0
= 2/
0
.
If the function g(t) is periodic with period T
0
, then a Fourier series representing
g(t) over an interval T
0
will also represent g(t) for all t.
8
Example
g(t)=e
(t)
t/2

2
0
0 t
9
Example

0
= 2/T
0
= 2 rad/s.
g(t) = C
0
+

n=1
C
n
cos(2nt +
n
)
n 0 1 2 3 4
C
n
0.504 0.244 0.125 0.084 0.063

n
0 -75.96 -82.87 -85.84 -86.42
We can plot
the amplitude C
n
versus this gives us the amplitude spectrum
the phase
n
versus (phase spectrum).
This two plots together are the frequency spectra of g(t).
10
Amplitude and phase spectra

0 2 4 6 8 10
12
0.504
0.244
C
n

0.125
0.084

/2

n

0 2 4 6

11
Exponential Fourier Series
Consider a set of exponentials
e
jn
0
t
n = 0, 1, 2, ...
The components of this set are orthogonal.
A signal g(t) can be expressed as an exponential series over an interval T
0
:
g(t) =

n=
D
n
e
jn
0
t
D
n
=
1
T
0

T
0
g(t)e
jn
0
t
dt
12
Trigonometric and exponential Fourier series
Trigonometric and exponential Fourier series are related. In fact, a sinusoid in the
trigonometric series can be expressed as a sum of two exponentials using Eulers
formula.
C
n
cos(n
0
t +
n
) =
C
n
2
[e
j(n
0
t+
n
)
+ e
j(n
0
t+
n
)
]
=

C
n
2
e
j
n

e
jn
0
t
+

C
n
2
e
j
n

e
jn
0
t
= D
n
e
jn
0
t
+ D
n
e
jn
0
t
D
n
=
1
2
C
n
e
j
n
D
n
=
1
2
C
n
e
j
n
13
Amplitude and phase spectra. Exponential case

|D
n
|
0.504
0.122
0 2 2
0.122
0.0625

D
n

14
Parsevals Theorem
Trigonometric Fourier series representation g(t) = C
0
+

n=1
C
n
cos(n
0
t +
n
).
The power is given by
P
g
= C
2
0
+
1
2

n=1
C
2
n
.
Exponential Fourier series representation g(t) =

n=
D
n
e
jn
0
t
.
Power for the exponential representation
P
g
=

n=
|D
n
|
2
15
Conclusions
Trigonometric Fourier series
Exponential Fourier series
Amplitude and phase spectra
Parsevals theorem
16