exp_fs1 15/03/2018, 16+19

In [ ]: cd matlab
pwd

Exponential Fourier Series

Scope and Background Reading

This session builds on our Revision of the to Trigonometrical Fourier Series
(https://github.com/cpjobling/EG-247-Resources/blob/master/fourier_series/trig_fseries.ipynb).

Trigonometric Fourier series uses integration of a periodic signal multiplied by sines and cosines at the
fundamental and harmonic frequencies. If performed by hand, this can a painstaking process. Even with
the simplifications made possible by exploiting waveform symmetries, there is still a need to integrate
cosine and sine terms, be aware of and able to exploit the tigonometrc identities, and the properties of
orthogonal functions before we can arrive at the simplified solutions. This is why I concentrated on the
properties and left the computation to a computer.

However, by exploiting the exponential function eat , we can derive a method for calculating the
coefficients of the harmonics that is much easier to calculate by hand and convert into an algorithm that
can be executed by computer.

The result is called the Exponential Fourier Series and we will develop it in this session.

The material in this presentation and notes is based on Chapter 7 (Starting at Section 7.8) of Steven T.
Karris, Signals and Systems: with Matlab Computation and Simulink Modelling, 5th Edition.
(http://site.ebrary.com/lib/swansea/docDetail.action?docID=10547416) from the Required Reading List.
Some clarificattion was needed and I used Chapter 4 of Benoit Boulet, Fundamentals of Signals and
Systems (https://ebookcentral.proquest.com/lib/swansea-ebooks/reader.action?
ppg=150&docID=3135971&tm=1518703383001) from the Recommended Reading List for this.

Agenda
Background

Trigonometric Fourier Series (https://github.com/cpjobling/EG-247-

Resources/blob/master/fourier_series/trig_fseries.ipynb)

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 1 of 16
exp_fs1 15/03/2018, 16+19

First Hour

Exponents and Euler's Equation

The Exponential Fourier series
Symmetry in Exponential Fourier Series
Example

Second Hour

Line spectra
Power in periodic signals
Steady-State Response of an LTI System to a Periodic Signals

The Exponential Function eat

You should already be familiar with eat because it appears in the solution of differential
equations.
It is also a function that appears in the definition of the Laplace and Inverse Laplace Transform.
It pops up again and again in tables and properies of the Laplace Transform.

Case when a is real.

When a is real the function eat will take one of the two forms illustrated below:

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 2 of 16
exp_fs1 15/03/2018, 16+19

When a < 0 the response is a decaying exponential (red line in plot)

When a = 0 eat = 1 -- essentially a model of DC
When a > 0 the response is an unbounded increasing exponential (blue line in plot)

(Image generated with this Matlab script: expon.m (matlab/expon.m)

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 3 of 16
exp_fs1 15/03/2018, 16+19

Case when a is imaginary

This is the case that helps us simplify the computation of sinusoidal Fourier series.

It was Leonhard Euler (http://en.wikipedia.org/wiki/Leonhard_Euler) who discovered the formula

(http://en.wikipedia.org/wiki/Euler%27s_formula) visualized above.

Some important values of omega*t

These are useful when simplifying expressions that result from integrating functions that involve the
imaginary exponential

Give the following:

ejωt when ωt = 0
ejωt when ωt = π/2
ejωt when ωt = π
ejωt when ωt = 3π/2
ejωt when ωt = 2π

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 4 of 16
exp_fs1 15/03/2018, 16+19

Case where a is complex

We shall not say much about this case except to note that the Laplace transform equation includes such
a number. The variable s in the Laplace Transform
∞

∫0
f (t)e−st dt

is a complex exponential.

The consequences of a complex s have particular significance in the development of system stability
theories and in control systems analysis and design. Look out for them in EG-243.

Two Other Important Properties

By use of trig. identities, it is relatively straight forward to show that:

ejωt + e−jωt
cos ωt =
2
and

ejωt −e−jωt
sin ωt =
j2

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 5 of 16
exp_fs1 15/03/2018, 16+19

We can use this result to convert the Trigonometric Fourier Series into an Exponential Fourier Series
which has only one integral term to solve per harmonic.

The Exponential Fourier Series

As as stated in the notes on the Trigonometric Fourier Series (https://github.com/cpjobling/EG-247-
Resources/blob/master/fourier_series/trig_fseries.ipynb) any periodic waveform f (t) can be represented
as

$$\begin{gathered} f(t) = \frac{1}{2}{a_0} + {a_1}\cos \Omega_0 t + {a_2}\cos 2\Omega_0 t + \cdots \

{b_1}\sin \Omega_0 t + {b_2}\sin 2\Omega_0 t + \cdots \ \end{gathered} $$

If we replace the cos and sin terms with their imaginary expontial equivalents:

$$\begin{gathered} f(t) = \frac{1}{2}{a_0} + {a_1}\left(\frac{e^{j\Omega_0 t} + e^{-j\Omega_0 t}}{2}\right) +

{a_2}\left(\frac{e^{j2\Omega_0 t} + e^{-j2\Omega_0 t}}{2}\right) + \cdots \

{b_1}\left(\frac{e^{j\Omega_0 t} - e^{-j\Omega_0 t}}{j2}\right) + {b_2}\left(\frac{e^{j2\Omega_0 t}

- e^{-j2\Omega_0 t}}{j2}\right) + \cdots \ \end{gathered} $$

Gouping terms with same exponents

( 2 j2 ) ( 2 j2 ) ( 2 j2 ) ( 2 j2 )
a2 b a1 b 1 a1 b a2 b
f (t) = ⋯ + − 2 e−j2Ω0 t + − 1 e−jΩ0 t + a0 + + 1 ejΩ0 t + + 2
2

New coefficents
The terms in parentheses are usually denoted as

2( j ) 2
1 b 1
C−k = ak − k = (ak + jbk )

2( j ) 2
1 b 1
Ck = ak + k = (ak −jbk )

1
C0 = a0
2

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 6 of 16
exp_fs1 15/03/2018, 16+19

The Exponential Fourier Series

f (t) = ⋯ + C−2 e−j2Ω0 t + C−1 e−jΩ0 t + C0 + C1 ejΩ0 t + C2 ej2Ω0 t + ⋯

or more compactly
n
Ck e−jkΩ0 t
∑
f (t) =
k= −n

Important

The Ck coefficents, except for C0 are complex and appear in conjugate pairs so

C−k = Ck∗

Evaluation of the complex coefficients

The coefficients are obtained from the following expressions*:
2π
1
2π ∫0
Ck = f (Ω0 t)e−jk(Ω0 t) d(Ω0 t)

or
T
1
T ∫0
Ck = f (t)e−jkΩ0 t dt

These are much easier to derive and compute than the equivalent Trigonemetric Fourier Series
coefficients.

* The analysis that leads to this result is provided between pages 7-31 and 7-32 of the text book
(https://ebookcentral.proquest.com/lib/swansea-ebooks/reader.action?
ppg=243&docID=3384197&tm=1518704101461). It is not a difficult proof, but we are more interested in
the result.

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 7 of 16
exp_fs1 15/03/2018, 16+19

Trigonometric Fourier Series from Exponential Fourier Series

By substituting C−k and Ck back into the original expansion

1
Ck + C−k = (ak −jbk + ak + jbk )
2
so

ak = Ck + C−k

Similarly

1
Ck −C−k = (ak −jbk −ak −jbk )
2
so

bk = j (Ck −C−k )

Thus we can easily go back to the Trigonetric Fourier series if we want to.

Symmetry in Exponential Fourier Series

Since the coefficients of the Exponential Fourier Series are complex numbers, we can use symmetry to
determine the form of the coefficients and thereby simplify the computation of series for wave forms that
have symmetry.

Even Functions
For even functions, all coefficients Ck are real.

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 8 of 16
exp_fs1 15/03/2018, 16+19

Proof

Recall

2( j ) 2
1 b 1
C−k = ak − k = (ak + jbk )

and

2( j ) 2
1 b 1
Ck = ak + k = (ak −jbk )

From knowledge of the trig. fourier series, even functions have no sine terms so the bk coefficients are 0.
Therefore both C−k and Ck are real.

Odd Functions
For odd functions, all coefficients Ck are imaginary.

By a similar argument, all odd functions have no cosine terms so the ak coefficients are 0. Therefore
both C−k and Ck are imaginary.

Half-wave symmetry
If there is half-wave symmetry, Ck = 0 for k even.

Proof

From Trigonometric Fourier Series, if there is half-wave symmetry, all even harnonics are zero, thus both
ak and bk are zero for k even. Hence C−k and Ck are also zero when k is even.

No symmetry
If there is no symmetry the Exponential Fourier Series of f (t) is complex.

Relation of C-k to Ck
C−k = Ck∗ always

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 9 of 16
exp_fs1 15/03/2018, 16+19

End of Pre-Class Presentation

Example 1
Compute the Exponential Fourier Series for the square wave shown below assuming that ω = 1

Some questions for you

Square wave is an [odd/even/neither] function?
DC component is [zero/non-zero]?
Square wave [has/does not have] half-wave symmetry?

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 10 of 16
exp_fs1 15/03/2018, 16+19

Hence

C0 = [?]
Coefficients Ck are [real/imaginary/complex]?
Subscripts k are [odd only/even only/both odd and even]?
What is the integral that needs to be solved for Ck ?

Solution

2π [∫0 ] 2π [ −jk ∣π ]
π 2π
1 1 A −jk(Ω0 t) ∣π −A −jk(Ω0 t) ∣2π
∫π
−jk(Ω0 t) −jk(Ω0 t)
Ae d(Ω0 t) + (−A)e d(Ω0 t) = e ∣ + e ∣
∣0 −jk

2π [ −jk ] 2jπk
1 A −jkπ A −j2kπ A
= ( e −1) + ( e −e−jkπ ) = ( 1 −e−jkπ + e−j2kπ −e−jkπ )
jk
A A
(e −2e−jkπ −1) = (e −1)
−j2kπ −jkπ 2
2jπk 2jπk
For n odd*, e−jkπ = −1. Therefore
Cn A A A 2A
(e −1) =
−jkπ 2
= (−1 −1)2 = (−2)2 =
n = odd 2jπk 2jπk 2jπk jπk

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 11 of 16
exp_fs1 15/03/2018, 16+19

* You may want to verify that C0 = 0 and

Cn
= 0.
n = even

Exponential Fourier series for the square wave with odd symmetry
From the definition of the exponential Fourier series

f (t) = ⋯ + C−2 e−j2Ω0 t + C−1 e−jΩ0 t + C0 + C1 ejΩ0 t + C2 ej2Ω0 t + ⋯

the exponential Fourier series for the square wave with odd symmetry is

jπ ( ) jπ ∑ n e
2A 1 −j3Ω0 t −jΩ0 t jΩ 0 t 1 j3Ω0 t 2A 1 jkΩ0 t
f (t) = ⋯ − e −e + e + e + ⋯ =
3 3 n= odd

Note sign change in first two terms. This is due to the fact that C−k = Ck∗ .

E.g. since C3 = 2A/j3π , C−3 = C3∗ = −2A/j3π

Trig. Fourier Series from Exponential Fourier Series

Since

jπ (
+ ⋯)
2A 1 −j3Ω0 t −jΩ0 t jΩ 0 t 1 j3Ω0 t
f (t) = ⋯ − e −e + e + e
3 3
gathering terms at each harmonic frequency gives

π ( ( ) 3( ) ) π (
4A ejΩ0 t −e−jΩ0 t 1 ej3Ω0 t −e−j3Ω0 t 4A 1
f (t) = ⋯+ + + ⋯ = sin Ω0 t + sin 3Ω0 t +
2j 2j 3

Computing coefficients of Exponential Fourier Series in

Matlab

Example 2
Verify the result of Example 1 using Matlab.

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 12 of 16
exp_fs1 15/03/2018, 16+19

Solution
Solution: See efs_sqw.m (matlab/efs_sqw.m).

EFS_SQW
Calculates the Exponential Fourier for a Square Wave with Odd Symmetry.

In [9]: clear all

Set up parameters

In [10]: syms t A;

tau = 1;
T0 = 2*pi; % w = 2*pi*f -> t = 2*pi/omega
k_vec = [-5:5];

Define f(t)

IMPORTANT: the signal definition must cover [0 to T0]

In [11]: xt = A*(heaviside(t)-heaviside(t-T0/2)) - A*(heaviside(t-T0/2)-heav

iside(t-T0));

Compute EFS

In [12]: [Ck, w] = FourierSeries(xt, T0, k_vec)

Ck =

[ (A*2i)/(5*pi), 0, (A*2i)/(3*pi), 0, (A*2i)/pi, 0, -(A*2i)/pi, 0,

-(A*2i)/(3*pi), 0, -(A*2i)/(5*pi)]

w =

-5 -4 -3 -2 -1 0 1 2 3 4 5

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 13 of 16
exp_fs1 15/03/2018, 16+19

Plot the numerical results from Matlab calculation.

Convert symbolic to numeric result

In [14]: Xw = subs(Ck,A,1);

Plot

In [15]: subplot(211)
stem(w,abs(Xw), 'o-');
title('Exponential Fourier Series for Square Waveform with Odd Symm
etry')
xlabel('\Omega_0 (rad/sec)');
ylabel('|c_k|');
subplot(212)
stem(w,angle(Xw), 'o-');
xlabel('\Omega_0 (rad/sec)');
ylabel('\angle c_k [radians]');

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 14 of 16
exp_fs1 15/03/2018, 16+19

End of First Hour

Summary

Exponents and Euler's Equation

The exponential Fourier series
Symmetry in Exponential Fourier Series
Example

Next Time

Line spectra
Power in periodic signals
Steady-State Response of an LTI System to a Periodic Signals

Read notes (http://nbviewer.jupyter.org/github/cpjobling/EG-247-

Resources/blob/master/week5/exp_fs2.ipynb)

Answers to in-class problems

Some important values of omega*t - Solution

When ωt = 0 : ejωt = ej0 = 1
When ωt = π/2 : ejωt = ejπ/2 = j
When ωt = π : ejωt = ejπ = −1
When ωt = 3π/2 : ejωt = ej3π/2 = −j
When ωt = 2π : ejωt = ej2π ej0 = 1
It is also worth being aware that nωt, when n is an integer, produces rotations that often map back to the
simpler cases given above. For example see ej2π above.

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 15 of 16
exp_fs1 15/03/2018, 16+19

Some answers for you

Square wave is an odd function!
DC component is zero!
Square wave has half-wave symmetry!

Hence

C0 = 0
Coefficients Ck are imaginary!
Subscripts k are odd only!
What is the integral that needs to be solved for Ck ?

2π [∫0 ]
2π π 2π
1 1
2π ∫0 ∫π
−jk(Ω0 t) −jk(Ω0 t)
Ck = f (Ω0 t)e d(Ω0 t) = Ae d(Ω0 t) + (−A)e−jk(Ω0 t) d(Ω0 t)

http://localhost:8890/nbconvert/html/week5/exp_fs1.ipynb?download=false Page 16 of 16