Math 1102-Ch-3-Lecture Note Fourier Series
Math 1102-Ch-3-Lecture Note Fourier Series
Fourier Series
functions helps us to find the unknown coefficients a0 , a1 , a2 in a simple fashion. These are
called Fourier coefficients of f(x) with respect to y 0 , y1 , y 2 , . If we multiply both side of the above
expansion by p(x)yn(x) for a fixed n, and then integrate over a x b, we obtain, assuming term by
term integration is permissible,
b b
b b
a m p( x) y m ( x) y n ( x) a n p( x) y n2 ( x)dx a n y n (x) , all other
2
m0 a a
integrals being zero in the right hand side, because of the orthogonality of the set.
1
b
p ( x) f ( x) y n ( x)dx
Thus a n a
2
.
y n ( x)
Definition: A function f is said to be a periodic function with period p if p is the least positive number
such that f(x +p) = f(x) for all x in the domain of f. It follows that f(x+np) = f(x) for all x in the
domain of f and all integers n.
Example : f(x) = sinx and g(x) = cos x are the familiar periodic functions with period p = 2.
The constant function h(x) = c is a periodic function, since h(x +p) =c = h(x) for all p (0, ).
Proposition: If f and g are periodic functions with period p, then H(x) = af(x) + bg(x), for some
constants a & b, is a periodic function with period p.
Proof: H(x+p) = a f(x+p) +bg(x +b) = af(x) + bg(x) = H(x) since f(x+p) = f(x) &g(x+p) = g(x)
Therefore, H(x) is a periodic function with period p.
Definition : A functional series of the form
a0
a1 cos x b1 sin x a2 cos 2 x b2 sin 2 x an cos nx bn sin nx ...
2
a0
= a n cos nx bn sin nx is called a trigonometric series; where a0 , an , bn (n = 1, 2, 3,
2 n 1
are real constants,) are called the coefficients of the trigonometric series.
a
If the series converges, say to a function f(x), i.e, 0 (a n cos nx bn sin nx) f ( x)
2 n 1
Then f is a period function with period p = 2 , by the above proposition.
a0
Thus f(x+2) = f(x) for all x in the domain of f, where f ( x) an cos nx bn sin nx .
2 n1
3.2 Fourier Series
3.2.1 Fourier Series of function with period
Definition : The Fourier series for the periodic function f(x) in an interval α < x < α + 2 is given by.
a0
f ( x) an cos nx bn sin nx And the coefficients
2 n1
a0 , a n and bn , n = 1, 2, 3, are called the Fourier coefficients.
To evaluate the Fourier Coefficients, the following integrals, involving sine and cosine functions are
useful.
2 2
i)
cos nxdx
sin nx 0 , n = 1, 2, 3,
2
2 2
1
ii)
cos mx cos nxdx
2
[cos(m n) x cos(m n) x]dx
1 1 1 2
sin(m n) x sin(m n) x = 0, m n.
2 m n mn
2
iii)
cos mx sin nxdx 0
2 2
Corollary : 1. If = 0, the interval becomes 0 < x < 2 , and Euler’s formulae are given by:
2 2
1 2
a0 1
f ( x) sin nxdx .
1
f ( x)dx , an f ( x) cos nxdx , bn
0 0
0
2. If = - , then the interval becomes -< x < , and the Euler’s Formulae and given by:
1 1
f ( x) cos nxdx , f ( x) sin nxdx .
1 a0 bn
a0
f ( x)dx ,
x 2 cos nx
2
3
2 2 2 2
1 1 x 1
an f ( x) cos nxdx cos nxdx ( x)
2
cos nxdx )
0
0 2 4 0
2
1 ( x) 2 cos(nx) ( sin(nx) 1 2 2 1
sin(nx) 2( x) 2 = 2
4 n n 2
n3 0
4 n 2 n 2 n
2 2 2
1 1 x
bn
0
f ( x) sin(nx)dx
0 2
sin(nx)dx
2
1 2 cos(nx) sin(nx) 2
4 ( x) 2( x) 3 cos(nx) 0
n
2 2
n n 0
x 2 cos(nx)
2
Therefore, f ( x)
2 12 n1 n 2
DIRICHLET’S CONDITIONS:
Suppose that:
a) f(x) is defined and single – valued except possibly at a finite number of points in(α, α + 2).
b) f(x) is periodic outside (α, α + 2) with period 2.
c) f(x) and f (x) are sectionally continuous in (α, α + 2).
a0
Then the series (an cos(nx) bn sin(nx) with coefficients
2 n 1
a 2 a 2 a 2
1 1 1
a0
a
f ( x)dx, an
a
f ( x) cos(nx)dx & bn
f ( x) sin(nx)dx converges to
a
( x) dx ( x)dx
a0
a c
a 2
1 1
a 2
c c
a0 ( x) cos(nx)dx ( x) cos(nx)dx and bn
a ( x) sin(nx)dx
( x) sin(nx)dx
a c c
4
, if x 0
1 1 1 1 1 2
f ( x) and deduce that .
x, if 0 x n 0 ( 2n 1)
2
12 32 5 2 7 2 8
a0
Solution: Let f ( x) an cos(nx) bn sin(nx)dx
2 n1
1
0 0
1 1 2 1 1
Then a0 dx 0 0
xdx dx xdx x
2 0 2 2
1
0
an cos(nx)dx x cos(nx)dx cos(nx)dx x cos(nx)dx
1 sin(nx)
0 0 0 n 0
1 x sin(nx) sin(nx) cos(nx)
2 1 1
1
dx
n
n 0 0
n n 0 n
2
1
0 0
1
and bn sin(nx)dx x sin(nx)dx sin(nx)dx
0
x sin(nx)dx
0
cos(nx)
0
1 x cos(nx)
cos(nx) 1 (1)
n
1 (1) n Sin (nx)
dx
n n 0 0
n n n n 2 0
5
3.2.2 Fourier Series of Functions with arbitrary period P = 2L
In many engineering problems, the period of the function required to be expanded is not 2 but some
other interval say 2L. In order to apply the foregoing discussion to functions of period 2L, this interval
must be converted to the length 2 . This involves only a proportional change in the scale.
Consider the periodic function f(x) defined on (α, α + 2L). To change the problem to period 2 put
L
t x which implies that x t . This gives when x (a, a 2L) . Thus the function f(x) of period
L
L L
2L in (α, α + 2L) is transformed to function f t of period 2 in , 2 . Hence f t
L L
can be expressed in Fourier series as:
L a
g (t ) f t 0 an cos(nt ) bn sin(nt ) , where
2 n1
2
2
L
2
L 1 L
1 1
a0 f t dt , an f t cos(nt )dt , and bn f t sin(nt )dt , where
a L
and x t .
L
Marking the inverse substitution t x and noting that dt dx in the above formula, the Fourier
L L
series Expansion of f(x) in the interval (α, α +2L) is given by
a0
f ( x) a n cos n x bn sin n x where
2 n1 L L
2L
2 L 2 L
1
a0
1
L f ( x)dx, an
1
L f ( x) cos n x dx and bn
L L
f ( x)sin n x dx
L
Corollary :
i. Putting α = 0 in these formulae, we get the corresponding Fourier Coefficients for the
interval (0, 2L)
2L 2L
1 1
a0
L
0
f ( x)dx, an f ( x) cos n x dx
L0 L
2L
1
bn
L f ( x) sin n L x dx
0
6
ii. Putting a = - L in the above formulae, we get the results:
L
1
L L
1 1
a0 f ( x)dx, an f ( x) cos n x dx and bn f ( x)sin n x dx are the Fourier coefficients
L L L L L L L L
in the interval (-L, L).
Example : Find the Fourier series of the periodic function f(x) of period 2, where
1, for 1 x 0 1 1 1 2 1 1 1
f ( x) and show that 1 2 2 2 and 1
2 x, for 0 x 1 3 5 7 8 3 5 7 4
a0
Solution: 2L = 2 L = 1, and f ( x) an cos nx bn sin nx
2 n1
1 1
L 1 0 1
1 1
where a0 f ( x)dx f ( x)dx dx 2 xdx x x 2 1 1 0
L L 1 1 1 0 0 0
1 0 1
1
an f ( x) cos(nx)dx cos(nx)dx 2 x cos(nx)dx
1 1 1 0
1 1
sin(nx)
2
x sin(nx) 10 22 2 cos(nx)
n 0 n n
0
2 2
2
(1) n
1
(1 (1) n ) 2 2 1 (1) n 1
n
2 2
n
2 2
n
1 0 1
1
bn f ( x) sin(nx)dx sin(nx)dx 2 sin(nx)dx
1 1 1 0
cos(nx)
0
2 x cos(nx) 1 1 cos(nx)dx
n 1 n 0 0
a0
Thus f ( x) an cos(nx) bn sin(nx)
2 n1
0
2 1 (1) n1
cos( nx )
1 3(1) n1
sin(nx)
n 1 n2 2
n
4 4 2 4 4
cos x sin x sin 2x cos 3x sin 3x
2 32 2 3
2 4 4
sin 4x 2 2 cos(5x) sin(5x)
4 5 5
7
4 cos x cos 3x cos 5x 2 sin 2x 2 sin 4x
2
2 sin x sin 3x
1 2
3 2
5 2
2 3 4
lim f ( x) lim f ( x)
When x = 0, the series converges to x0 x0 1
2 2
1 4 1 1 1 2
Therefore, 2 2 2 2 (0 0 0 )
2 1 3 5
4 1 1 1 1 1 1
1 2 .
=
2 12 3 2 5 2
12 3 2 5 2
n 0 ( 2n 1)
2
8
1 1
When x , f 1 , giving 1 24 (0 0 0 ) 2 2 0 2 0 2
2 2 2 3 3 3
1 1 1 1
(1) n
2 2 2 2
3 5 7
4 1 1 1
2 1
3 5 7 3 5 7
2n 1
n 0 4
3.2.3 Fourier Series of Odd and Even Functions
Definition: A function f(x) is said to be odd iff f (-x) = -f(x)
A function f(x) is said to be even iff(-x) = f(x)
Example : The functions sin(nx) and tan (nx) are odd functions. Graph of odd function is symmetric
about the origin.
Example : The functions cos(nx), x2, sec(nx) are even functions. Graphs of even functions are
symmetric about y-axis.
Proposition : If f(x) is a periodic function with period p = 2L, then
L
L f ( x)dx 0
2 f ( x)dx, if f is even
L
0,
if f is odd
Recall that a periodic function f(x) defined in (- L, L) can be represented by the Fourier series:
a
nx nx
f ( x) 0 an cos bn sin ,
2 n1 L L
where a0 f ( x)dx, an f ( x) cos n x dx&
L L L
1 1 1
L L L L L
bn
L L
f ( x) sin n x dx
L
L L
1 2
When f(x) is an even function a 0
L L
f ( x)dx f ( x)d x ,
L0
1 L 2 L
an
L L
f ( x) cos n x dx f ( x) cos n x dx, sin ce f ( x) cos n x
L L 0
L L
L
1
is even, and bn
L L
f ( x) sin n
L
x dx 0, sin ce f ( x) sin n
L
x is odd.
8
Therefore, the Fourier series expansion of a periodic even function f(x) contains only the cosine terms
whose coefficients are a0 f ( x) dx and a n f ( x) cos n x dx
L L
2 2
L L0 L
2
L L
1
f ( x) f ( x) dx
n 1 L
f ( x) cos n x dx
L cos
n x .
L 0 0 L
L L L L L L
L
and bn 1 2 L
f ( x) sin n x dx
L
f ( x) sin n x dx , since f ( x) sin n x is even.
L L L 0
L L
Thus, if a period function f(x) is odd, its Fourier series expansion contains only the sine terms, whose
coefficients are bn 2 f ( x)sin n x dx , so that
L
L0 L
L
2
f ( x) f ( x) sin n x dx sin n x .
n 1 L 0 L L
Coefficients
Example: Find a series to represent f(x) = x2 in the interval (, ) . Deduce the values of i)
(1) n 1 1 2
1 1 1 1 2
n 1 n 2
1 2
1
3
1
2 2 2 =
12
and ii)
n 1 n
2
2 2 2
1 2 3 6
4 n 4
n
x dx 2 2 (1) n sin n x 4 2(12 ) and bn 0
2 n
x cos x cos
n
2 2
0 0 n n 0 n
2 3
(1) n n 2 4 2 cos x cos
2 2
4 x cos x
Therefore, = f ( x) x 2
3
2
n2
cos
x
2 ,
n 1 3 1 2
2 2
32
2
4 1 2
1 1
which is the required Fourier series. i) Putting x = 0, we get f (0) 0 2 2 2 2
3 1 2 3
(1) n 1 2
i.e,
n 1 n2
12
9
lim f ( x) lim f ( x)
x 2
2 2
4 2 1 1 1 1
ii) Putting x , we get x
2
2
2 2 2 2
2 2 3 1 2 3 4
1 2
1 4 1 1 1 1 1 1 1 1 2
1 2 2 2 2 2 2 2 2 2
2
n 2
6
3 1 2 3 4 1 2 3 4 3 4
n 1
2
3
B -2
Hence the Fourier series for f(x) over the full period (-2, 2) will contain only sine series terms given by
nx nx
2 2
nx Where b 2
f ( x) bn sin n f ( x) sin dx x sin dx
n 1 2 2 0 2 0 2
10
2
4(1) n 4(1) n 1
2 x cos n x
0
2
2 4 Sin n
n n 2 2
x n n
2 0
0
3
4
(1) n1 nx 4 sin x sin x
f ( x)
n1 n
sin
2
sin x
2
2
2 3
b) The graph of f(x) = x in (0, 2) is the line OA. Let us extend the function f(x) in the interval (-2, 0)
(shown by the B O ) so that the new function is symmetric about the
y – axis and, therefore, represents an even function in (-2, 2).
Y
B A
X
-2 0 2
Hence the Fourier series for f(x) = x over the full period (-2, 2) will contain only cosine terms given by
a
nx
f ( x) 0 an cos a0 2 f ( x)dx xdx 2 and an 2 f ( x) cos nx dx
2 2 2
2 n1 2 20 0 20 2
2
nx 4 cos nx 4 (1) 2 1 4 (1) 1
2
nx 2 x sin nx
2 2 n
2
x cos
n 2 0 n 0 2
dx sin dx
0 2 n 2 2 2 0 n
2 2
n 2 2
2 4
(1) n 1
Therefore, the desired result is: f ( x) 2
2
n 1 n 2
cos n x
2
3 5
4 2 2 3 2 5 cos cos cos x
1 2 2 Cos x 2 Cos x 2 Cos 1 8 2 2 2
1 2 3 2 5 2 2 12 32 52
If
f (t ) dx converges, i.e.; f(x) is absolutely integrable in (- ,), then the 1st – term on the right
L
1 1
2 L L 2 L
side of (*) approaches 0 as L, since f (t ) dt f (t ) dt .
1 n (t x)
The 2nd – term on the right side of (*) tends to lim
L
L n1
f (t ) cos
L
dt
1
lim
0
f (t ) cos n (t x)dt on writing
n 1 L
.Thus as L , (*) becomes
1
f ( x)
f (t ) cos (t x)dtd
0
called the Fourier Integral of f(x).
1
Remark: 1. If function f is continuous at x, then f ( x)
f (t ) cos (t x)dtd .
0
f ( x 0) f ( x 0) 1
If f is not continuous at x, then f (t ) cos (t x) dtd
2 0
where f ( x 0) lim f (t ) and f(x - 0) lim f (t ) .
t x t x -
2. Fourier sine and cosine integrals. Expanding cos(t-x) = cos (t - x)
= cost cosx + sin t sin t, the Fourier integral of f(x) may be written as
1 1
f ( x)
cos x f (t ) cos t dt d
0
sin x f (t ) sin t dt d .
0
If f(x) is an odd function, f(t) cos (t) is also an odd function while f(t) sin (t) is even. Then the 1st
2
term on the right side of the above equation vanishes, and f ( x)
sinx f(t) sin(t)dt d , which is
0 0
1, for 0 x
Example: Express f ( x) as a Fourier sine integral and hence
0, for x
12
1 cos( )
evaluate sin( x )d .
0
2
Solution: The Fourier sine integral of f(x) is f ( x)
sin(x) d f(t)sin(t)dt
0 0
2
sin (x) d sin(t)dt 2 sin(x)d cos(t )
2 1 cos( )
0
sin( x )d
0 0 0 0
2 1 cos( )
Therefore, f ( x) sin (x) d 1 cos(x) sin(x) d f ( x) 2 , for 0 x
2 0, for x
0 0
At x =, which is a point of discontinuity of f(x), then the value of the above integral is
f ( x) lim f ( x)
lim
1 1 cos( )
sin x d , at x
x x
0 , i.e,
2 2 2 2 4 0
4
13