Ma207 Lect10 Beamer Akp PDF

MA 207 - Differential Equations-II
Amiya Kumar Pani
Department of Mathematics,
Indian Institute of Technology Bombay,
Powai, Mumbai 76
[email protected]
November 9, 2020
Amiya Kumar Pani Lecture 10

Start with .......
“ It is just as foolish to complain that people are
selfish and treacherous as it is to complain that the
magnetic field does not increase unless the electric
field has a curl. Both are laws of nature”
by John von Neumann

Start with .......
“ It is just as foolish to complain that people are
selfish and treacherous as it is to complain that the
magnetic field does not increase unless the electric
field has a curl. Both are laws of nature”
by John von Neumann
“It is India that gave us the ingenious method of

expressing all numbers by means of ten symbols,
each symbol receiving a value of position as well as
an absolute value; a profound and important idea
which appears so simple to us now that we ignore
its true merit. But its very simplicity and the great
ease which it has lent to computations put our
arithmetic in the first rank of useful inventions; and
we shall appreciate the grandeur of the
achievement the more when we remember that it
escaped the genius of Archimedes and Apollonius,
two of the greatest men produced by antiquity.”
by Pierre Simon Laplace
Outline of the lecture
Fourier Transforms ( introduced by Joseph Fourier in 1822)

Basic Properties
Inversion and Plancherel Theorem
Applications

Fourier Transforms on R
Let f (x) be a piecewise

Zsmooth on any finite
interval and
∞
absolutely integrable |f (x)| dx < ∞ real-valued
−∞
function.


interval and
∞
−∞
function.
Z ∞
1
F(f ) = fˆ(ω) = √ f (x)e −iωx dx is called the
2π −∞
FOURIER TRANSFORM of f (x).


interval and
∞
−∞
function.
Z ∞
1
2π −∞

Note that,
Z ∞
1
|fˆ(ω)| = | √ f (x)e −iωx dx|
2π −∞


interval and
∞
−∞
function.
Z ∞
1
2π −∞

Note that,
Z ∞
1
ˆ
|f (ω)| = | √ f (x)e −iωx dx|
2π −∞
Z ∞
1
≤ √ |f (x)| |e −iωx | dx
2π −∞


interval and
∞
−∞
function.
Z ∞
1
2π −∞

Note that,
Z ∞
1
ˆ
|f (ω)| = | √ f (x)e −iωx dx|
2π −∞
Z ∞
1
≤ √ |f (x)| |e −iωx | dx
2π −∞
Z ∞
= |f (x)| dx < ∞.
−∞

Quotations by Jean Baptiste Joseph Fourier —-
“The integrals which we have obtained are not only general expressions which
satisfy the differential equation, they represent in the most distinct manner the
natural effect which is the object of the phenomenon... when this condition is
fulfilled, the integral is, properly speaking, the equation of the phenomenon; it
expresses clearly the character and progress of it, in the same manner as the
finite equation of a line or curved surface makes known all the properties of
those forms.”

Quotations by Jean Baptiste Joseph Fourier —-
“The integrals which we have obtained are not only general expressions which
satisfy the differential equation, they represent in the most distinct manner the
natural effect which is the object of the phenomenon... when this condition is
fulfilled, the integral is, properly speaking, the equation of the phenomenon; it
expresses clearly the character and progress of it, in the same manner as the
finite equation of a line or curved surface makes known all the properties of
those forms.”
“Primary causes are unknown to us; but are subject to
simple and constant laws, which may be discovered by
observation, the study of them being the object of natural
philosophy. Heat, like gravity, penetrates every substance
of the universe, its rays occupy all parts of space. The
object of our work is to set forth the mathematical laws
which this element obeys. The theory of heat will
hereafter form one of the most important branches of
general physics.”

Examples
Compute the Fourier transforms of :

1. f (x) = k, 0 < x < a, 0 otherwise.

Examples

1. f (x) = k, 0 < x < a, 0 otherwise.
By definition,
Z ∞
1
fˆ(ω) = √ f (x)e −iωx dx
2π −∞

Examples

1. f (x) = k, 0 < x < a, 0 otherwise.
By definition,
Z ∞
1
2π −∞
Z a
1
= √ k e −iωx dx
2π 0

Examples

1. f (x) = k, 0 < x < a, 0 otherwise.
By definition,
Z ∞
1
2π −∞
Z a
1
2π 0
k e −iωx a

= √
2π −iω x=0

Examples

1. f (x) = k, 0 < x < a, 0 otherwise.
By definition,
Z ∞
1
2π −∞
Z a
1
2π 0
k e −iωx a

= √
2π −iω x=0
k (e −iωa − 1) k (1 − e −iωa )
= √ = √
2π −iω 2π iω

Examples

1. f (x) = k, 0 < x < a, 0 otherwise.
By definition,
Z ∞
1
2π −∞
Z a
1
2π 0
k e −iωx a

= √
2π −iω x=0
k (e −iωa − 1) k (1 − e −iωa )
= √ = √
2π −iω 2π iω

2. f (x) = e −|x| x ∈ R.

2. f (x) = e −|x| x ∈ R.
By definition,
Z ∞
1
ˆ
f (ω) = √ f (x)e −iωx dx
2π −∞

2. f (x) = e −|x| x ∈ R.
By definition,
Z ∞
1
ˆ
2π −∞
Z ∞
1
= √ e −|x| e −iωx dx
2π −∞

2. f (x) = e −|x| x ∈ R.
By definition,
Z ∞
1
ˆ
2π −∞
Z ∞
1
2π −∞
Z 0 Z ∞
1 1
= √ e (1−iω)x dx + √ e −(1+iω)x dx
2π −∞ 2π 0

2. f (x) = e −|x| x ∈ R.
By definition,
Z ∞
1
ˆ
2π −∞
Z ∞
1
2π −∞
Z 0 Z ∞
1 1
= √ e (1−iω)x dx + √ e −(1+iω)x dx
2π −∞ 2π 0
r
2 1
= (CHECK !)
π 1 + ω2

Properties of Fourier Transforms
1. LINEARITY
The Fourier transform is a linear operation; that is, for any
functions f (x) and g (x) whose Fourier transforms exist and
any constants a and b,

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
Proof : Exercise (use definition)

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
2. SHIFTING
If f (x) has a Fourier transform, so does f (x − a), that is,

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
2. SHIFTING
F(f (x − a)) = e −iωa F(f (x))

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
2. SHIFTING
F(f (x − a)) = e −iωa F(f (x))

Z ∞
1
Proof : F(f (x − a)) = √ f (x − a)e −iωx dx
2π −∞

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
2. SHIFTING
F(f (x − a)) = e −iωa F(f (x))

Z ∞
1
2π −∞
Set x − a = y , then y + a = x, dx = dy .

1. LINEARITY
F(af + bg ) = aF(f ) + bF(g )
2. SHIFTING
F(f (x − a)) = e −iωa F(f (x))

Z ∞
1
2π −∞
Set x − a = y , then y + a = x, dx = dy . Using a change of
variables, we obtain

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π−∞
Z ∞
−iωa 1
= e √ f (y )e −iωy dy
2π −∞

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).
3. SHIFTING OF ω AXIS
If fˆ(ω) is the Fourier transform of f (x), then fˆ(ω − a) is the
Fourier transform of e iax f (x).

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).
Fourier transform of e iax f (x). That is,
F(e iax f (x)) = fˆ(ω − a).

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).
Proof :
Z ∞
1
F(e iax f (x)) = √ e iax f (x)e −iωx dx
2π −∞

Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).
Proof :
Z ∞
1
2π −∞
Z ∞
1
= √ e −i(ω−a)x f (x) dx = fˆ(ω − a)
2π −∞
Z ∞
1
F(f (x − a)) = √ f (y )e −iω(y +a) dy
2π −∞
Z ∞
−iωa 1
2π −∞
−iωa
= e F(f (x)).
Proof :
Z ∞
1
2π −∞
Z ∞
1
= √ e −i(ω−a)x f (x) dx = fˆ(ω − a)
2π −∞
Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
Use the Fourier transform of f (x) to compute the FT of e αx f (x).

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
Use the Fourier transformZof∞f (x) to compute the FT of e αx f (x).
1
2π −∞

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
Z a
1
= √ e −iωx dx
2π 0

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π 0 2π −iω x=0

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π 0 2π −iω x=0
1 (e −iωa − 1) 1 (1 − e −iωa )
= √ = √
2π −iω 2π iω

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π 0 2π −iω x=0
1 (e −iωa − 1) 1 (1 − e −iωa )
= √ = √
2π −iω 2π iω
Now, F(e iax f (x)) = fˆ(ω − a).

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π 0 2π −iω x=0
1 (e −iωa − 1) 1 (1 − e −iωa )
= √ = √
2π −iω 2π iω
Now, F(e iax f (x)) = fˆ(ω − a). That is,

F(e αx f (x)) = F(e i(−iα)x f (x)) = fˆ(ω + iα).

Example 1 :
• Let (
1 if 0 < x < a
f (x) =
0 elsewhere.
1
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π 0 2π −iω x=0
1 (e −iωa − 1) 1 (1 − e −iωa )
= √ = √
2π −iω 2π iω
Now, F(e iax f (x)) = fˆ(ω − a). That is,

F(e αx f (x)) = F(e i(−iα)x f (x)) = fˆ(ω + iα).
1 (1 − e −i(ω+iα)a )
=⇒ F(e αx f (x)) = √
2π i(ω + iα)
Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.
Use the Fourier transform of f (x) to compute the FT of e iαx f (x).

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
Z a
1
= √ e −iωx dx
2π −a

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
2π −a 2π −iω x=−a

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
1 (e −iωa − e iωa )
= √
2π −iω

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
1 (e −iωa − e iωa ) 1 sin aω
= √ = √ 2
2π −iω 2π ω

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
1 (e −iωa − e iωa ) 1 sin aω p sin aω
= √ = √ 2 = 2/π
2π −iω 2π ω ω

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
= √ = √ 2 = 2/π
F(e iax f (x)) = fˆ(ω − a)

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
= √ = √ 2 = 2/π
F(e iax f (x)) = fˆ(ω − a) =⇒ F(e iαx f (x)) =

Example 2 :
• Let (
1 if −a < x < a
f (x) =
0 elsewhere.

Z ∞
1
ˆ
2π −∞
1 e −iωx a
Z a
1 −iωx
= √ e dx = √
= √ = √ 2 = 2/π
sin a(ω − α)
F(e iax f (x)) = fˆ(ω − a) =⇒ F(e iαx f (x)) =
p
2/π .
(ω − α)

Properties (contd..)
4. Fourier transform of the derivative of f (x)

Theorem
If
• f (x) is continuous on R,


Theorem
If
• f (x) → 0 as |x| → ∞,


Theorem
If
• f (x) → 0 as |x| → ∞,
• f 0 (x) is absolutely integrable in R; then


Theorem
If
• f (x) → 0 as |x| → ∞,
F(f 0 (x)) = iωF(f (x))


Theorem
If
• f (x) → 0 as |x| → ∞,
F(f 0 (x)) = iωF(f (x))
Proof : From the definition of FT we have


Theorem
If
• f (x) → 0 as |x| → ∞,
F(f 0 (x)) = iωF(f (x))
Proof : From the definition of FT we have

Z ∞
0 1
F(f (x)) = √ f 0 (x)e −iωx dx
2π −∞

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
Integrating by parts, we obtain

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
Since f (x) → 0 as |x| → ∞, we obtain

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).
Remark : If
• f (x) and its n − 1 derivatives are continuous on R,

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).
Remark : If
djf
• → 0 as |x| → ∞, j = 0, 1, n − 1.
dx j

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).
Remark : If
djf
• j
→ 0 as |x| → ∞, j = 0, 1, n − 1.
dxn
• f (x) is absolutely integrable in R; then

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).
Remark : If
djf
• j
→ 0 as |x| → ∞, j = 0, 1, n − 1.
dxn
F(f n (x)) = (iω)n F(f (x)) .

Z ∞
1
F(f 0 (x)) = √ f 0 (x)e −iωx dx
2π −∞
r ∞ Z ∞
0 1 −iωx
−iωx
F(f (x)) = f (x)e − (−iω) f (x)e dx
2π
−∞ −∞
F(f 0 (x)) = iωF(f (x)).
Remark : If
djf
• j
→ 0 as |x| → ∞, j = 0, 1, n − 1.
dxn
F(f n (x)) = (iω)n F(f (x)) .

Properties (Contd..)
5. If f (x) and x m f (x) are absolutely integrable, then

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
d ˆ fˆ(ω + h) − fˆ(ω)
(f (ω)) = lim
dω h→0 h

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
(f (ω)) = lim
dω h→0 h
Z ∞ −i(ω+h)x
1 e − e −iωx
= lim √ f (x) dx
h→0 2π −∞ h

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
(f (ω)) = lim
dω h→0 h
Z ∞ −i(ω+h)x
1 e − e −iωx
= lim √ f (x) dx
h→0 2π −∞ h
Z ∞ −ihx
1 −iωx e −1
= √ e lim f (x) dx
2π −∞ h→0 h

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
(f (ω)) = lim
dω h→0 h
Z ∞ −i(ω+h)x
1 e − e −iωx
= lim √ f (x) dx
h→0 2π −∞ h
Z ∞ −ihx
1 −iωx e −1
2π −∞ h→0 h
Z ∞
1
= √ (−ix) e −iωx f (x) dx
2π −∞

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
(f (ω)) = lim
dω h→0 h
Z ∞ −i(ω+h)x
1 e − e −iωx
= lim √ f (x) dx
h→0 2π −∞ h
Z ∞ −ihx
1 −iωx e −1
2π −∞ h→0 h
Z ∞
1
= √ (−ix) e −iωx f (x) dx = (−i) F(xf (x)).
2π −∞

d ˆ
(f (ω)) = −iF (xf (x)) (A)
dω
dm ˆ
(f (ω)) = (−i)m F (x m f (x)) (B)
dω m
Proof of (A):
(f (ω)) = lim
dω h→0 h
Z ∞ −i(ω+h)x
1 e − e −iωx
= lim √ f (x) dx
h→0 2π −∞ h
Z ∞ −ihx
1 −iωx e −1
2π −∞ h→0 h
Z ∞
1
= √ (−ix) e −iωx f (x) dx = (−i) F(xf (x)).
2π −∞
Use induction to prove (B).
Example : Tut. Sheet 6 (iii)
2
Find the FT of f (x) = e −ax , x ∈ R, a > 0.

2
Z ∞
1
ˆ
We know, F(f (x)) = f (ω) = √ f (x)e −iωx dx,
2π −∞

2
Z ∞
1
ˆ
2π −∞
d ˆ
(f (ω)) = −iF (xf (x)).
dω

2
Z ∞
1
ˆ
2π −∞
d ˆ
(f (ω)) = −iF (xf (x)). Idea : Form a DE in fˆ.
dω

2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
As f (x) = e −ax , = −2axf (x) =⇒ − f 0 (x) = xf (x) .
dx 2a

2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
dx 2a
d ˆ
Now, (f (ω)) = −iF (xf (x))
dω

2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
dx 2a
d ˆ
Now, (f (ω)) = −iF (xf (x))
dω Z ∞
i
= −√ xf (x)e −iωx dx
2π −∞

2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
dx 2a
d ˆ
Now, (f (ω)) = −iF (xf (x))
dω Z ∞
i
2π −∞
Z ∞
i 1
= −√ − f 0 (x)e −iωx dx
2π −∞ 2a

2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
dx 2a
d ˆ
Now, (f (ω)) = −iF (xf (x))
dω Z ∞
i
2π −∞
Z ∞
i 1
= −√ − f 0 (x)e −iωx dx
2π −∞ 2a
i
= F(f 0 (x))
2a
2
Z ∞
1
ˆ
2π −∞
d ˆ
dω
2 df 1
dx 2a
d ˆ
Now, (f (ω)) = −iF (xf (x))
dω Z ∞
i
2π −∞
Z ∞
i 1
= −√ − f 0 (x)e −iωx dx
2π −∞ 2a
i i ω
= F(f 0 (x)) = (iω)F(f (x)) = − fˆ(ω)
2a 2a 2a
d fˆ ω
=⇒ + fˆ(ω) = 0, with
dω 2a

d fˆ ∞
Z
ω 1 2
=⇒ + fˆ(ω) = 0, with fˆ(0) = √ e −ax dx.
dω 2a 2π −∞

d fˆ ∞
Z
ω 1 2
=⇒ + fˆ(ω) = 0, with fˆ(0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
On solving, f (ω) = f (0)e

d fˆ ∞
Z
ω 1 2
=⇒ + fˆ(ω) = 0, with fˆ(0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
On solving, f (ω) =Zf (0)e
∞
1 2
Now, fˆ(0) = √ e −ax dx.
2π −∞

d fˆ ∞
Z
ω 1 2
=⇒ + fˆ(ω) = 0, with fˆ(0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
∞
1 2
√ 2π √−∞
Choose y = ax, a dx = dy .

d fˆ ∞
Z
ω 1 2
=⇒ + fˆ(ω) = 0, with fˆ(0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
∞
1 2
√ 2π √−∞
Choose y = ax, a dx = dy . Thus,

d fˆ
Z ∞
ω ˆ 1 2
=⇒ ˆ
+ f (ω) = 0, with f (0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
∞
1 2
√ 2π √−∞
Z ∞
1 1 2
ˆ
f (0) = √ √ e −y dy
a 2π −∞

d fˆ
Z ∞
ω ˆ 1 2
=⇒ ˆ
+ f (ω) = 0, with f (0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
∞
1 2
√ 2π √−∞
Z ∞
1 1 2
ˆ
f (0) = √ √ e −y dy
a 2π −∞
Z ∞
1 1 2
= √ √ e −y dy =
2a π −∞

d fˆ
Z ∞
ω ˆ 1 2
=⇒ ˆ
+ f (ω) = 0, with f (0) = √ e −ax dx.
dω 2a 2π −∞
ˆ ˆ −ω 2 /4a
∞
1 2
√ 2π √−∞
Z ∞
1 1 2
ˆ
f (0) = √ √ e −y dy
a 2π −∞
Z ∞
1 1 2 1
= √ √ e −y dy = √ .
2a π −∞ 2a
1 2 2 1 2
Therefore, fˆ(ω) = √ e −ω /4a , that is, F(e −ax ) = √ e −ω /4a .
2a 2a

Convolution
6. The convolution f ∗ g of functions f and g is defined by

Convolution
h(x) = (f ∗ g )(x)

Convolution

Z ∞
h(x) = (f ∗ g )(x) = f (y )g (x − y ) dy
−∞

Convolution

Z ∞ Z ∞
h(x) = (f ∗ g )(x) = f (y )g (x − y ) dy = f (x − y )g (y ) dy
−∞ −∞

Convolution

Z ∞ Z ∞
−∞ −∞
Theorem
Suppose that
f (x) and g (x) are piecewise continuous, bounded and absolutely
integrable on R.

Convolution

Z ∞ Z ∞
−∞ −∞
Theorem
Suppose that
integrable on R. Then
√
F(f ∗ g ) = 2πF(f )F(g )

Convolution

Z ∞ Z ∞
−∞ −∞
Theorem
Suppose that
integrable on R. Then
√
F(f ∗ g ) = 2πF(f )F(g )

Proof :

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞
An interchange of order of integration gives

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

∞ Z ∞
r Z
1 −iωx
F(f ∗ g ) = f (y ) g (x − y )e dx dy
2π −∞ −∞

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

∞ Z ∞
r Z
1 −iωx
F(f ∗ g ) = f (y ) g (x − y )e dx dy
2π −∞ −∞
∞ Z ∞
r Z
1 −iωs
= f (y ) g (s)e ds e −iωy dy (x − y = s)
2π −∞ −∞

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

∞ Z ∞
r Z
1 −iωx
F(f ∗ g ) = f (y ) g (x − y )e dx dy
2π −∞ −∞
∞ Z ∞
r Z
1 −iωs
2π −∞ −∞
r Z ∞ Z ∞
1 −iωy −iωs
= f (y )e dy × g (s)e ds
2π −∞ −∞

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

∞ Z ∞
r Z
1 −iωx
F(f ∗ g ) = f (y ) g (x − y )e dx dy
2π −∞ −∞
∞ Z ∞
r Z
1 −iωs
2π −∞ −∞
r Z ∞ Z ∞
1 −iωy −iωs
2π −∞ −∞
√
= 2π fˆ(ω)ĝ (ω)

Proof :
∞ ∞
r Z Z
1 −iωx
F(f ∗ g ) = e f (y )g (x − y ) dy dx
2π −∞ −∞

∞ Z ∞
r Z
1 −iωx
F(f ∗ g ) = f (y ) g (x − y )e dx dy
2π −∞ −∞
∞ Z ∞
r Z
1 −iωs
2π −∞ −∞
r Z ∞ Z ∞
1 −iωy −iωs
2π −∞ −∞
√
= 2π fˆ(ω)ĝ (ω)
√
= 2π F(f )F(g ).

Inverse Fourier Transform
The INVERSE FOURIER TRANSFORM of fˆ is defined by


Z ∞
1
f (x) = √ fˆ(ω)e iωx dω
2π −∞


Z ∞
1
2π −∞
When can we say that the inverse FT exists?


Z ∞
1
2π −∞
PLANCHEREL’S IDENTITY


Z ∞
1
2π −∞
PLANCHEREL’S IDENTITY Z ∞
If f (x) is square integrable, that is, |f (x)|2 dx < ∞, then fˆ
−∞
exists and fˆ is square integrable.


Z ∞
1
2π −∞
−∞
Moreover,


Z ∞
1
2π −∞
−∞
Moreover,
Z ∞ Z ∞
2
|f (x)| dx = |fˆ(ω)|2 dω .
−∞ −∞


Z ∞
1
2π −∞
−∞
Moreover,
Z ∞ Z ∞
2
|f (x)| dx = |fˆ(ω)|2 dω .
−∞ −∞
Exercises: • Tutorial Sheet 6, Problem 6
• Kreyszig, Page 575, Qns 1-10

Application - Heat equation in unbounded domain
ut − uxx = 0 − ∞ < x < ∞, t > 0

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
Idea : Take Fourier transform to obtain an IVP in ω, that is
simple to solve.

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
simple to solve. Take inverse transforms to retrieve the solution
u(x, t).

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
u(x, t). Take Fourier transforms to obtain

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
Note that ûxx = (iω) û(ω, t)

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
On substituting in the equation, we obtain

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
ût (ω, t) + ω 2 û(ω, t) = 0

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
ût (ω, t) + ω 2 û(ω, t) = 0
û(ω, 0) = uˆ0 (ω)

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
ût (ω, t) + ω 2 û(ω, t) = 0
û(ω, 0) = uˆ0 (ω)
Solving the above ODE (IVP) in time,

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
ût (ω, t) + ω 2 û(ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
û(ω, t) = uˆ0 (ω)e −ω t .

ut − uxx = 0 − ∞ < x < ∞, t > 0
u(x, 0) = u0 (x) −∞<x <∞
ût (ω, t) − ûxx (ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
ût (ω, t) + ω 2 û(ω, t) = 0
û(ω, 0) = uˆ0 (ω)
2
û(ω, t) = uˆ0 (ω)e −ω t .
Now we need to find the inverse Fourier transform.
2
Set ĝ (ω) = e −ω t .

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
Using convolution theorem, we have û(ω, t) = √ F(f ∗ g )
2π

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
Taking inverse FT, we obtain
1
u(x, t) = √ (f ∗ g )
2π

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
Now we need to find g such that ĝ (ω, t) = e −ω

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
2 1 2
We know that F(e −ax ) = √ e −ω /4a
2a

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
2 1 2
2a
2t 2 /(1/t))
Now, e −ω = e −ω .

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
2 1 2
2a
2t 2 /(1/t)) 1
Now, e −ω = e −ω . Comparing, 4a = 1/t =⇒ a =
4t

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
2 1 2
2a
2t 2 /(1/t)) 1
2 /4a √ 2
4t
Thus, ĝ (ω, t) = e −ω = 2aF(e −ax ) with a = 1/(4t).

2
Set ĝ (ω) = e −ω t .
Then, û(ω, t) = uˆ0 (ω)ĝ (ω).
1
2π
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy .
2π −∞
2t
2 1 2
2a
2t 2 /(1/t)) 1
2 √ 2
4t
Thus, ĝ (ω, t) = e −ω /4a = 2aF(e −ax ) with a = 1/(4t).
1 2 1 2
That is, ĝ (ω, t) = √ F(e −x /4t ) =⇒ g (x, t) = √ (e −x /4t ).
2t 2t
Recall,
1
u(x, t) = √ (f ∗ g )
2π

Recall,
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy
2π −∞

Recall,
1
u(x, t) = √ (f ∗ g )
2π
Z ∞
1
= √ u0 (y )g (x − y , t) dy
2π −∞
Z ∞
1 1 2
= √ √ u0 (y )e −(x−y ) /4t dy
2π 2t −∞
Z ∞
1 2 /4t
Hence, u(x, t) = √ u0 (y )e −(x−y ) dy
2 πt −∞

Z ∞ Z ∞
Moreover, 2
|f (x)| dx = |fˆ(ω)|2 dω (Plancherel’s identity)
−∞ −∞

Z ∞ Z ∞
Moreover, 2
−∞ −∞
Now,
Z ∞ Z ∞
2
2
|u(x, t)| dx = |uˆ0 (ω)|2 e −2ω t dω
−∞ −∞

Z ∞ Z ∞
Moreover, 2
−∞ −∞
Now,
Z ∞ Z ∞
2
2
|u(x, t)| dx = |uˆ0 (ω)|2 e −2ω t dω
−∞
Z−∞
∞
≤ |uˆ0 (ω)|2 dω
−∞

Z ∞ Z ∞
Moreover, 2
−∞ −∞
Now,
Z ∞ Z ∞
2
|u(x, t)| dx2
= |uˆ0 (ω)|2 e −2ω t dω
−∞
Z−∞
∞
≤ |uˆ0 (ω)|2 dω
Z−∞
∞
= |u0 (x)|2 dx (Plancherel0 s identity)
−∞
Z ∞ Z ∞
2
=⇒ |u(x, t)| dx ≤ |u0 (x)|2 dx.
−∞ −∞

Z ∞ Z ∞
Moreover, 2
−∞ −∞
Now,
Z ∞ Z ∞
2
|u(x, t)| dx2
= |uˆ0 (ω)|2 e −2ω t dω
−∞
Z−∞
∞
≤ |uˆ0 (ω)|2 dω
Z−∞
∞
= |u0 (x)|2 dx (Plancherel0 s identity)
−∞
Z ∞ Z ∞
2
=⇒ |u(x, t)| dx ≤ |u0 (x)|2 dx.
−∞ −∞
Stability property

Ma207 Lect10 Beamer Akp PDF

Uploaded by

Copyright:

Available Formats

Ma207 Lect10 Beamer Akp PDF

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Ma207 Lect10 Beamer Akp PDF

Uploaded by

Copyright:

Available Formats

MA 207 - Differential Equations-II

Amiya Kumar Pani

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

“It is India that gave us the ingenious method of

Fourier Transforms ( introduced by Joseph Fourier in 1822)

Amiya Kumar Pani Lecture 10

Let f (x) be a piecewise

Amiya Kumar Pani Lecture 10

Let f (x) be a piecewise

FOURIER TRANSFORM of f (x).

Amiya Kumar Pani Lecture 10

Let f (x) be a piecewise

FOURIER TRANSFORM of f (x).

Amiya Kumar Pani Lecture 10

Let f (x) be a piecewise

FOURIER TRANSFORM of f (x).

Amiya Kumar Pani Lecture 10

Let f (x) be a piecewise

FOURIER TRANSFORM of f (x).

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Compute the Fourier transforms of :

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

F(f (x − a)) = e −iωa F(f (x))

Amiya Kumar Pani Lecture 10

F(f (x − a)) = e −iωa F(f (x))

Amiya Kumar Pani Lecture 10

F(f (x − a)) = e −iωa F(f (x))

Amiya Kumar Pani Lecture 10

F(f (x − a)) = e −iωa F(f (x))

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10

Amiya Kumar Pani Lecture 10