F Ourier Series (French: Four-Ee-Ey)
F Ourier Series (French: Four-Ee-Ey)
I INTRODUCTION
▪ Although the idea of expanding a function in the form of such a series had been used by
Bernoulli, D’Alembert and Euler (c. 1750) to solve problems associated with the vibration of
strings, it was Joseph Fourier (1768–1830) who developed the approach to a stage where it
was generally useful.
▪ Fourier, a French physicist, was interested in heat-flow problems: given an initial
temperature at all points of a region, he was concerned with determining the change in the
temperature distribution over time.
▪ Fourier submitted a paper to the French Academy of Sciences in competition for a prize
offered for the best mathematical treatment of heat conduction which he later won.
▪ When Fourier postulated in 1807 that an arbitrary function f(x) could be represented by a
trigonometric series of the form
the result was considered so startling that it met considerable opposition from the leading
mathematicians of the time, notably Laplace, Poisson and, more significantly, Lagrange,
who is regarded as one of the greatest mathematicians of all time.
▪ They questioned his work because of its lack of rigour (ie. complexity), and it was
probably this opposition that led to the Academy’s rejection of the publication of Fourier’s
work. His classic text Théorie Analytique de la Chaleur (The Analytical Theory of Heat)
was then published in 1822.
▪ This text has since become the source for the modern methods of solving practical
problems associated with partial differential equations subject to prescribed boundary
conditions.
1
▪ Angular frequency or circular frequency or omega, ω, is defined as the factor inside the
sinusoid
where a0=2A0. 1/2 a0 is chosen since it is equal to the DC component of the signal (for
ee/ece) and it fits in the general equation for an.
▪ Orthogonality relations for sine and cosine functions shows the cases when the periodic
integral of cosine and sine and their products is not equal to zero on a specific case.
Examples are
▪ Integrating the left and right side of the Fourier series expansion, assuming each terms of
the series is integrable over a period, the orthogonality relations will be applied yielding
which is the mean value of f(t) over a period or the DC component of the signal.
▪ Multiplying the left and right side of the Fourier series expansion with sine and cosine then
integrate them respectively, the orthogonality relations will then be applied.
2
where the coefficient of the real and imaginary parts gives the values for an and bn. This is
particularly useful if the amplitude |an+jbn| of the nth harmonic is required.
II.iii Function of period 2π
▪ If the period is 2π, T=2π and ω=2π/2π=1 thus the Fourier series expansion would be
▪ Even the unit frequency (ie. ω=1) may rarely be encountered in practice, it reduces the
mathematical complexity. The new equations is still general such that at t are substituted to
▪ If the given periodic function is piecewise, that is defined for several ranges, only the
integral limits would be affected and thus must be evaluated for each piecewise range.
The summation and other terms are not affected since they are still series of sines and
cosines.
▪ Example iii.1. Obtain the Fourier series expansion of the periodic function f(t) of period 2π
defined by
f (t)=t (0<t<2 π ), f (t)=f ( t+2 π )
▪ Solution: The (0<t<2π) determines the range of the function and the rightmost term states
that the equation is periodic. Sketching the function f(t) which starts at 0,
3
▪ For bn, we use the formula
▪ Example iii.2. Sketch the graph of the given and obtain the Fourier series expansion of the
function f(t) given below.
2
f (t)=t + 1 (−π <t < π ) , f (t)=f (t +2 π )
▪ Solution: The (-π<t<π) determines the range of the function and the rightmost term states
that the equation is periodic. Sketching the function f(t) which starts at -π,
4
and by integration by parts would yield
▪ The alternative formula of using complex numbers are also possible which yields
▪ Example iii.3. A piecewise periodic function f(t) of period 2π is defined within the period 0
< t < 2π by
5
Sketch a graph of f(t) for −2π < t < 3π and find a Fourier series expansion of it.
▪ Solution: The function has several values from 0 to 2π. Sketching each function f(t) within
its interval and start at -2π,
▪ As stated above, all piecewise functions are solved by determining the integral of each
range. Thus, solving for a0 would be
6
▪ Then the Fourier series expansion is
▪ The integral for one period based on the figure for even function would then be
a a
∫ f (t)dt=2∫ f (t)dt
−a 0
▪ If f(t) is an odd function then f(t) = −f(−t) for all t, and the graph of the function is
symmetrical about the origin
7
▪ The integral for one period based on the figure for odd function would then be
a
∫ f (t )dt=0
−a
▪ Using the integral simplifications above for odd and even function and the properties of
products of even and odd functions: product of two even signals or of two odd signals is an
even signal and that the product of an even signal and an odd signal is an odd signal.
Example proof for even*odd=odd
where
where
▪ One can generalize that if f(t) is even, the Fourier series expansion is only even terms that
is DC and cosine. if f(t) is odd, the Fourier series expansion is only odd terms that is the
sines.
▪ Example iv.1. A periodic function f(t) with period 2π is defined within the period −π < t < π
by
▪ Clearly f(t) is an odd function thus the formula for odd function would be used.
8
▪ Thus the Fourier series expansion is
▪ Example iv.2. A periodic function f(t) with period 2π is defined within the period −π < t < π
by
2
f (t)=t (−π <t < π ), f (t)=f (t +2 π )
Obtain the Fourier series expansion for it.
▪ Solution: Sketching the graph over the interval -4 π to 4π,
▪ which shows an odd function and the formula for odd Fourier series expansion is used.
▪ since sin nπ = 0 and cos nπ = (−1)n. Thus the Fourier series expansion of f(t) = t 2 is
9
II.v Linearity Property
▪ The sum of two functions f(t)=lg(t)+m(t) is equal to the sum of their Fourier series
expansion. This can be proven by directly substituting f(t) to the Fourier series expansion
formulae.
▪ Example v.1. Suppose that g(t) and h(t) are periodic functions of period 2π and are defined
within the period −π < t < π by
▪ Determine the Fourier series expansions of both g(t) and h(t) and use the linearity property
to confirm the expansion obtained in Example iii.2 for the periodic function f(t) defined within
the period −π < t < π by f(t) = t 2 + t.
▪ Solution: Example iv.2 already solved the Fourier series expansion of g(t)
▪ For h(t)=t,
recognizing again that cos nπ = (−1)n and sin nπ = 0. Thus the Fourier series expansion of
h(t) = t is
▪ Using linearity, combining the Fourier series expansion of g(t) and h(t) yields the Fourier
series for f(t).
▪
II.vi Exercises
▪ In each of the following a periodic function f(t) of period 2π is specified over one period. In
each case sketch a graph of the function for −4π < t < 4π and obtain a Fourier series
representation of the function.
10
II.vii Functions of period T
▪ Topic II.iii to topic II.vii all involved periods of 2π. As stated in topic II.iii, periods of 2π are
rarely encountered in practice and is used for ease of manipulation in determining the
Fourier coefficients while becoming acquainted with Fourier series
▪ Example vii.1. A periodic function f(t) of period 4 (that is, f(t + 4) = f(t)) is defined in the
range −2 < t < 2 by
Sketch a graph of f(t) for −6≤t≤6 and obtain a Fourier series expansion for the function.
▪ Solution: A graph of f(t) for −6≤t≤6 is shown below.
▪ Sketch a graph of f(t) for − 4≤t≤4 and determine a Fourier series expansion for the
function.
▪ Solution: A graph of f(t) for − 4≤t≤4 is shown below.
11
▪ Taking T = 2 in the Fourier series expansion formulae, we have
▪ Example vii.3. Obtain the Fourier series expansion of the rectified sine wave
12
▪
II.viii Exercises
▪ Determine the Fourier series expansion of the half-wave rectified sinusoidal wave shown
below.
▪ The Fourier series of such function can be obtained using Fourier series expansion
formulae in topic II.ii.
II.xi Half-range cosine and sine series
▪ As shown below, a non-periodic function (a) can be made periodic by interpolating the
succeeding graphs (b) to be symmetrical as even or odd function.
13
▪ The Fourier series of such function can be obtained using Fourier series expansion
formulae for odd and even in topic II.iv.
II.xii Exercises
▪ Exercise xii.1 Determine the Fourier series expansion of the graph below consisting only of
sine terms.
▪ Solution: Make the graph odd symmetric. See topic II.10 citing topic II.iv.
14