Top Banner
Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3
30

Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Dec 16, 2015

Download

Documents

Kelli Drewry
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
Page 1: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Lecture 11:Introduction to Fourier Series

Sections 2.2.3, 2.3

Page 2: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Trigonometric Fourier Series

•Outline▫Introduction▫Visualization▫Theoretical Concepts▫Qualitative Analysis▫Example▫Class Exercise

Page 3: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Introduction• What is Fourier Series?

▫ Representation of a periodic function with a weighted, infinite sum of sinusoids.

• Why Fourier Series?▫ Any arbitrary periodic signal, can be approximated by

using some of the computed weights

▫ These weights are generally easier to manipulate and analyze than the original signal

Page 4: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Periodic Function

• What is a periodic Function?▫ A function which remains unchanged when time-shifted

by one period f(t) = f(t + To) for all values of t

• What is To

To To

Page 5: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties of a periodic function 1

• A periodic function must be everlasting▫ From –∞ to ∞

• Why?

• Periodic or Aperiodic?

Page 6: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties of a periodic function

• You only need one period of the signal to generate the entire signal▫ Why?

• A periodic signal cam be expressed as a sum of sinusoids of frequency F0 = 1/T0 and all its harmonics

Page 7: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

VisualizationCan you represent this simple function using sinusoids?

Single sinusoid representation

Page 8: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Visualization

To obtain the exact signal, an infinite number of sinusoids are required

)cos( 01 ta amplitude Fundamental

frequency

)3cos( 03 ta

New amplitude 2nd Harmonic

)5cos( 05 ta amplitude 4th Harmonic

Page 9: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Theoretical Concepts                                                                   

(6)

,...3,2,1....)sin()(2

,...3,2,1....)cos()(2

2

)sin()cos()(

01

1

01

1

00

00

00

01

01

0

ndttntfT

b

ndttntfT

a

T

tnbtnaatf

Tt

t

n

Tt

t

n

nn

nn

PeriodCosine terms

Sine terms

Page 10: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Theoretical Concepts                                                                   

(6)

n

nn

nnn

nn

n

a

b

bac

ac

tncctf

1

22

00

01

0

tan

)cos()(

Page 11: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

DC Offset

What is the difference between these two functions?

A

0 1 2-1-2

-A

A

0 1 2-1-2

Average Value = 0

Average Value ?

Page 12: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

DC OffsetIf the function has a DC value:

01

1

)(1

)sin()cos(2

1)(

00

01

01

0

Tt

t

nn

nn

dttfT

a

tnbtnaatf

Page 13: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Qualitative Analysis• Is it possible to have an idea of what your solution

should be before actually computing it?

For Sure

Page 14: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – DC Value• If the function has no DC value, then a0 = ?

-1 1 2

-A

A

DC?

A

0 1 2-1-2

DC?

Page 15: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Symmetry

A

A

0 π/2 π 3π/2

f(-t) = -f(t)

• Even function

• Odd function

0

-A

A

π/2 π 3π/2

f(-t) = f(t)

Page 16: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Symmetry• Note that the integral over a period of an odd function is?

,...3,2,1....)sin()(2 01

1

00

ndttntfT

bTt

tn

If f(t) is even:

EvenOddX = Odd

,...3,2,1....)cos()(2 01

1

00

ndttntfT

aTt

tn

EvenEvenX = Even

Page 17: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Symmetry• Note that the integral over a period of an odd function is

zero.

,...3,2,1....)cos()(2 01

1

00

ndttntfT

aTt

tn

If f(t) is odd:

OddEvenX = Odd

,...3,2,1....)sin()(2 01

1

00

ndttntfT

bTt

tn

OddOddX = Even

Page 18: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Symmetry•If the function has:

▫even symmetry: only the cosine and associated coefficients exist

▫odd symmetry: only the sine and associated coefficients exist

▫even and odd: both terms exist

Page 19: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Symmetry

• If the function is half-wave symmetric, then only odd harmonics exist

Half wave symmetry: f(t-T0/2) = -f(t)

-1 1 2

-A

A

Page 20: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Properties – Discontinuities•If the function has

▫ Discontinuities: the coefficients will be proportional to 1/n

▫ No discontinuities: the coefficients will be proportional to 1/n2

• Rationale:

-1 1 2

-A

A

A

0 1 2-1-2

Which is closer to a sinusoid?

Which function has discontinuities?

Page 21: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Example

• Without any calculations, predict the general form of the Fourier series of:

-1 1 2

-A

A

DC? No, a0 = 0;

Symmetry? Even, bn = 0;

Half wave symmetry?

Yes, only odd harmonics

Discontinuities?No, falls of as

1/n2Prediction an 1/n2 for n = 1, 3, 5, …;

Page 22: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Example• Now perform the calculation

2/

00

00

0

001

1

)cos()(4

)cos()(2

TTt

tn dttntf

Tdttntf

Ta

;20 T 2

20

...5,3,1...8

22 n

n

Aan

)cos(14

)cos(2222

1

0

nn

AdttnAtan

zero for n even

Page 23: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Example• Now compare your calculated answer with your

predicted form

DC? No, a0 = 0;

Symmetry?Even, bn = 0;

Half wave symmetry?

Yes, only odd harmonics

Discontinuities?No, falls of as

1/n2

Page 24: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Class exercise

• Discuss the general form of the solution of the function below and write it down

• Compute the Fourier series representation of the function

• With your partners, compare your calculations with your predictions and comment on your solution

A

0 1 2-1-2

Page 25: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines

,...3,2,1....)sin()(2

,...3,2,1....)cos()(2

2

)sin()cos()(

01

1

01

1

00

00

00

01

01

ndttntfT

b

ndttntfT

a

T

tnbtnatf

Tt

tn

Tt

tn

nn

nn

n

nn

nnn

nn

n

a

b

bac

ac

tncctf

1

22

00

01

0

tan

)cos()(

Page 26: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines

•Gives the frequency composition of the function▫Amplitude, phase of sinusoidal components

•Could provide information not found in time signal▫E.g. Pitch, noise components

•May help distinguish between signals ▫E.g speech/speaker recognition

Page 27: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines Example

-3 -2 -1 0 1 2 30.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

t/pi

y(t

)

exp(-t/2)

QUESTIONS --DC Yes____ ao = ? No_____ ao = 0Symmetry

Even____ an = ? bn = 0Odd____ an = 0 bn = ?Nether even nor odd ____ an = ? bn = ?

Halfwave symmetryYes_____ only odd harmonicsNo______ all harmonics

DiscontinuitiesYes_____ proportional to1/nNo______ proportional to1/n2

Note ? means find that variable.Comment on the general form of the Fourier Series coefficients [an and/or bn.]

X

X

X

X

Page 28: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines Example

,3,2,1....0

)2sin(2/

,...3,2,1....01

1

)0sin()(

0

2

,3,2,1....0

)2cos(2/

,3,2,1....01

1

)0cos()(

0

2

22

0

20

5042.02/01

1

)(

00

0

11

ndttnt

e

n

Tt

tdttntf

Tnb

ndtntt

e

n

Tt

tdttntf

Tna

T

dtt

e

Tt

tdttf

Ta

-3 -2 -1 0 1 2 30.2

0.3

0.4

0.5

0.6

0.7

0.8

0.9

1

t/pi

y(t

)

exp(-t/2)

Page 29: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines Example

,....3,2,1.....1

216

4.0342n

12

16

)2/

1(2/

82

12

16

)2/

44(2/

22

,...3,2,1.....1

216

2/4)2sin()2cos(4(

2/22

,...3,2,1.....1

216

008.1

12

16

)2/

1(2/

22

,...3,2,1.....1

216

2/)2cos()2sin(4(

2/22

nnn

ene

n

nene

nn

nennnenb

nnn

ee

nn

ennnena

Page 30: Lecture 11: Introduction to Fourier Series Sections 2.2.3, 2.3.

Spectral Lines Example

-3 -2 -1 0 1 2 30

0.5

1

t/pi

y(t)

exp(-t/2)

0 1 2 3 4 5 6 7 8 9 100

0.5

1

a n

n

0 1 2 3 4 5 6 7 8 9 100

0.2

0.4

b n

n

0 1 2 3 4 5 6 7 8 9 100

0.5

1C

n

n

0 1 2 3 4 5 6 7 8 9 10-2

-1

0

n [ra

d]

n

n

nn

nnn

nn

n

a

b

bac

ac

tncctf

1

22

00

01

0

tan

)cos()(