ENGR 3324: Signals and Systems

Ch6
Continuous-Time Signal Analysis
Engineering and Physics
University of Central Oklahoma
Dr. Mohamed Bingabr

Outline
Introduction
Fourier Series (FS) representation of
Periodic Signals.
Trigonometric and Exponential Form of FS.
Gibbs Phenomenon.
Parsevals Theorem.
Simplifications Through Signal Symmetry.
LTIC System Response to Periodic Inputs.

Sinusoidal Wave and phase

x(t) = Asin(t) = Asin(250t)
x(t)
A
t
T0 = 20 msec

x(t-0.0025)= Asin(250[t-0.0025])
= Asin(250t-0.25)= Asin(250t-45o)
A
t
td = 2.5 msec

Time delay td = 25 msec correspond to phase shift =45o

Representation of Quantity using Basis

Any number can be represented as a
linear sum of the basis number {1, 10,
100, 1000}
Ex: 10437 =10(1000) + 4(100) + 3(10) +7(1)
Any 3-D vector can be represented as a
linear sum of the basis vectors {[1 0 0],
[0 1 0], [0 0 1]}
Ex: [2 4 5]= 2 [1 0 0] + 4[0 1 0]+ 5[0 0 1]

Basis Functions for Time Signal

Any periodic signal x(t) with fundamental frequency
0 can be represented by a linear sum of the basis
functions {1, cos( 0t), cos(2 0t),, cos(n 0t),
sin( 0t), sin(2 0t),, sin(n 0t)}
Ex:
x(t) =1+ cos(2t)+ 2cos(2 2t)+ 0.5sin(23t)+ 3sin(2t)

x(t) =1+ cos(2t)+ 2cos(2 2t)+ 3sin(2t)+ 0.5sin(23t)

Purpose of the Fourier Series (FS)

FS is used to find the frequency components and
their strengths for a given periodic signal x(t).

The Three forms of Fourier Series

Trigonometric Form
Compact Trigonometric (Polar) Form.
Complex Exponential Form.

Trigonometric Form
It is simply a linear combination of sines and
cosines at multiples of its fundamental frequency,
f0=1/T.

n 1

n 1

x t a0 an cos 2f 0 nt bn sin 2f 0 nt
a0 counts for any dc offset in x(t).
a0, an, and bn are called the trigonometric Fourier
Series Coefficients.
The nth harmonic frequency is nf0.

Trigonometric Form
How to evaluate the Fourier Series Coefficients
(FSC) of x(t)?

n 1

n 1

x t a0 an cos 2nf 0t bn sin 2nf 0t

To find a0 integrate both side of the equation over a full period

1
a0 x t dt
T0 T0

Trigonometric Form

n 1

n 1

x t a0 an cos 2nf 0t bn sin 2nf 0t

To find an multiply both side by cos(2mf0t) and then integrate
over a full period, m =1,2,,n,

2
an x t cos 2nf 0t dt
T0 T0
To find bn multiply both side by sin(2mf0t) and then integrate
over a full period, m =1,2,,n,

2
bn x t sin 2nf 0t dt
T0 T0

Example

f(t)

f t a0 an cos 2nt bn sin 2nt

n 1

e-t/2

1
a0

Fundamental period
T0 =

Fundamental
frequency
f0 = 1/T0 = 1/ Hz

t
2

e dt

1 0.504

2 2t
2

an e cos 2nt dt 0.504

2
0

1 16n

0 = 2/T0 = 2 rad/s

2 2t
8n

bn e sin 2nt dt 0.504

2
0

1 16n
an and bn decrease in amplitude as n .

cos 2nt 4n sin 2nt

f t 0.504 1
2
n 1 1 16n

To what value does the FS converge at the point of discontinuity?

Dirichlet Conditions

A periodic signal x(t), has a Fourier series if

it satisfies the following conditions:
1. x(t) is absolutely integrable over any period,
namely

x(t ) dt

T0

2. x(t) has only a finite number of maxima and

minima over any period
3. x(t) has only a finite number of
discontinuities over any period

Compact Trigonometric Form

Using single sinusoid,

x t

C0

dc component

Cn cos 2nf 0t n

n 1
nth harmonic

C 0 a0
Cn , and n are related to the trigonometric coefficients an

and bn as:
2

Cn an bn

and

bn

n tan
an
1

The above relationships are obtained from the

trigonometric identity

a cos(x) + b sin(x) = c cos(x + )

Role of Amplitude in Shaping Waveform

x t C0 Cn cos 2nf 0t n
n 1

Role of the Phase in Shaping a

Periodic Signal

x t C0 Cn cos 2nf 0t n
n 1

Compact Trigonometric

f t C0 Cn cos 2nt n

f(t)

n 1

a0 0.504

e-t/2

2
2
1 16n
8n

bn 0.504
2
1 16n
C0 ao 0.504
an 0.504

Fundamental period
T0 =
Fundamental frequency
f0 = 1/T0 = 1/ Hz

Cn a b 0.504
2
1 16n
1 bn
tan 1 4n
n tan
an
2
n

0 = 2/T0 = 2 rad/s

n 1

1 16n

f t 0.504 0.504

2
n

cos 2nt tan 1 4n

Line Spectra of x(t)

The amplitude spectrum of x(t) is defined
as the plot of the magnitudes |Cn| versus

The phase spectrum of x(t) is defined as

the plot of the angles Cn phase(Cn )
versus
This results in line spectra
Bandwidth the difference between the
highest and lowest frequencies of the
spectral components of a signal.

Line Spectra
f(t)

Cn 0.504

C0 0.504

1
e-t/2

1 16n 2

n tan 1 4n

f t 0.504 0.504
n 1

2
1 16n 2

cos 2nt tan 1 4n

f(t)=0.504 + 0.244 cos(2t-75.96o) + 0.125 cos(4t-82.87o) +

o
o
0.084
cos(6t-85.24
)
+
0.063
cos(8t-86.24
)+
C
n

0.504
0.244
0.125

0.084

0.063

10

-/2

Line Spectra

n 1

1 16n

f t 0.504 0.504

cos 2nt tan 1 4n

f(t)=0.504 + 0.244 cos(2t-75.96o) + 0.125 cos(4t-82.87o) +

o
o
0.084
cos(6t-85.24
)
+
0.063
cos(8t-86.24
)+
C
n

0.504
0.244
0.125

0.084

0.063

10

-/2

HW8_Ch6: 6.1-1 (a,d), 6.1-3, 6.1-7(a, b, c)

Exponential Form
x(t) can be expressed as

x t

D e

j 2f 0 nt

j 2f 0 nt

To find Dn multiply both side by e

over a full period, m =1,2,,n,

1
Dn
To

and then integrate

j 2f 0 nt

x
t
e
dt , n 0, 1, 2,....

To

Dn is a complex quantity in general Dn=|Dn|ej

D-n = Dn*

|Dn|=|D-n|
Even

Dn = -

D-n

Odd

D0 is called the constant or dc component of x(t)

Line Spectra of x(t) in the Exponential

Form
The line spectra for the exponential form has
negative frequencies because of the
mathematical nature of the complex exponent.
x(t ) ... | D 2 | e j 2 e j 20t | D1 | e j1 e j0t D0
| D1 | e j1 e j0t | D2 | e j 2 e j 20t ...
x(t ) C0 C1 cos(0t 1 ) C2 cos(20t 2 ) ...
|Dn| = 0.5 Cn

Dn =

Cn

Example
Find the exponential Fourier Series for the squarepulse periodic signal.
f(t)
/2
1
1
jnt
Dn
e
dt

2 / 2
sin n / 2

0.5sinc(n / 2)
n
1
D0
2
n even
0
Dn
1 / n n odd
0
n

for all n 3,7,11,15,

n 3,7,11,15,

Fundamental period
T0 =
Fundamental frequency
f0 = 1/T0 = 1/ Hz

0 = 2/T0 = 1 rad/s

Exponential Line Spectra

|Dn|

Dn

Example
The compact trigonometric Fourier Series
coefficients for the square-pulse periodic signal.
f(t)

1
C0
2
0 n even
Cn 2
n odd
n
0 for all n 3,7,11,15,
n
n 3,7,11,15,

Relationships between the Coefficients

of the Different Forms

Dn 0.5 an jbn
D n D

0.5 an jbn

Dn 0.5Cn n 0.5Cn e
D0 a0 C0

j n

Relationships between the Coefficients

of the Different Forms

an Dn D n 2 Re Dn

bk j Dn D n 2 Im Dn
an Cn cos n

bn Cn sin n
a0 D0 c0

Relationships between the Coefficients

of the Different Forms
2

Cn an bn

bn

n tan
an
1

Cn 2 Dn

n Dn
C0 a0 D0

Example
Find the exponential Fourier Series and sketch the
corresponding spectra for the impulse train shown
below. From this result sketch the trigonometric
spectrum and write the trigonometric Fourier Series.
T0 (t )

Solution
Dn 1 / T0
1
T0 (t )
T0

jn0 t
e

-2T0 -T0

Cn 2 | Dn | 2 / T0
C0 | D0 | 1 / T0
1
T0 (t )
T0

n 1

1 2 cos(n0t )

T0

2T0

Rectangular Pulse Train Example

Clearly x(t) satisfies the Dirichlet conditions.
x(t)
1

The compact trigonometric form is

1
2

( n 1) / 2
x(t ) cos nt (1)
1
2 n 1 n
2

n odd

Does the Fourier series converge to x(t) at every point?

Gibbs Phenomenon
Given an odd positive integer N, define the
N-th partial sum of the previous series

1
2

( n 1) / 2
x N (t ) cos nt (1)
1
2 n 1 n
2

n odd

According to Fouriers theorem,

theorem it should be

lim | xN (t ) x(t ) | 0

Gibbs Phenomenon Contd

x3 (t )

x9 (t )

Gibbs Phenomenon Contd

x21 (t )

x45 (t )

overshoot:
overshoot about 9 % of the signal magnitude
(present even if N )

Parsevals Theorem
Let x(t) be a periodic signal with period T
The average power P of the signal is defined as

1
P
T

T /2

T / 2

x(t ) dt

Expressing the signal as

x t C0 Cn cos(n0t n )
n 1

it is also

P C0 0.5Cn
2

n 1

P D 2 Dn
2
0

n 1

Simplifications Through Signal Symmetry

If x (t) is EVEN: It must contain DC and
Cosine Terms. Hence bn = 0, and Dn = an/2.
If x(t) is ODD: It must contain ONLY
Sines Terms. Hence a0 = an = 0, and
Dn=-jbn/2.

LTIC System Response to Periodic

Inputs

j 0 t

H(s)
H(j)

H ( j 0 ) e j 0 t

A periodic signal x(t) with period T0 can be expressed as

x(t )

jn0t
D
e
n

For a linear system

x(t )

jn0t
D
e
n

H(s)
H(j)

y (t )

D H ( jn )e

jn0t

Fourier Series Analysis of DC Power

Supply
A full-wave rectifier is used to obtain a dc signal from a
sinusoid sin(t). The rectified signal x(t) is applied to the
input of a lowpass RC filter, which suppress the timevarying component and yields a dc component with
some residual ripple. Find the filter output y(t). Find
also the dc output and the rms value of the ripple
voltage.
R=15

sint

Full-wave
rectifier

x(t)

C=1/5 F

y(t)

Fourier Series Analysis of Full-Wave

Rectifier
2

Dn

(1 4n 2 )

2
j 2 nt
x(t )
e
2

(
1

4
n
)
n
1
H ( j )
3 j 1
y (t )

D H ( jn )e

jn0 t

D0 2 /

PDC 4 / 2

Pripple 2 | Dn |2
n 1

Dn
Pripple

(1 4n ) 36n 1
0.0025
2

ripple rms Pripple 0.05

2
j 2 nt
y (t )
e
2
Ripple rms is only 5%

(
1

4
n
)( j 6n 1)
n
of the input amplitude

HW9_Ch6: 6.3-1(a,d), 6.3-5, 6.3-7, 6.3-11, 6.4-1, 6.4-3

Fourier Series Analysis of Full-Wave

Rectifier- Matlab Code
clear all
t=0:1/1000:3*pi;
for i=1:100
n=i;
yp=(2*exp(j*2*n*t))/(pi*(1-4*n^2)*(j*6*n+1));
This Matlab code will
n=-i;
plot y(t) for -100 n
yn=(2*exp(j*2*n*t))/(pi*(1-4*n^2)*(j*6*n+1));
100 and find the ripple
y(i,:)=yp+yn;
power according to the
end
equations below

yf = 2/pi + sum(y);
2
j 2 nt
y
(
t
)

e
plot(t,yf, t, (2/pi)*ones(1,length(yf)))
2

(
1

4
n
)( j 6n 1)
n

axis([0 3*pi 0 1]);

Pripple 2 | Dn |2 0.0025

Power=0;
for n=1:50
Power(n) = abs(2/(pi*(1-4*n^2)*(j*6*n+1)));
end
TotalPower = 2*sum((Power.^2));
figure; stem( Power(1,1:20));

n 1