Scribd
Upload
47 views1 page

The Discrete Fourier Transform

The pdf file shows sample problems regarding discrete Fourier transform/series, with proving and graphing problems. Refer to Arfken for study notes.

Uploaded by

Jhames Niño Trinidad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
47 views1 page

The Discrete Fourier Transform

The pdf file shows sample problems regarding discrete Fourier transform/series, with proving and graphing problems. Refer to Arfken for study notes.

Uploaded by

Jhames Niño Trinidad
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 1

Mathematical Physics 113

Problem Set 3
National Institute of Physics
(Dated: September 19, 2015)

Deadline: 2 October 2015


AR refers to Arfken 5th .

10 pts

Prove the orthogonality relation


2N
1
X
p=0

I. DISCRETE FOURIER SERIES [50 pts]

Consider the periodic function f (t) with repeating unit


(
t,
0 t < T /2,
f (t) =
(1)
T t, T /2 t < T.
That is, the period of the function is T .
10 pts

10 pts

Construct a Fourier basis {sin m t, cos n t} with


elements that have period T . In other words, what
are the possible frequencies i ?

10 pts

Construct the orthogonality integrals over the interval t [0, T ].

10 pts

Obtain the complete infinite Fourier series representation of f (t).

10 pts

Provide a plot that shows that f (t) can be approximated by a few terms of this infinite series (try the
terms with the largest amplitudes).

II. DISCRETE FOURIER TRANSFORM [40 pts]

Consider the periodic function f (t) given in the previous problem. Lets say that instead of the time variable
t being continuous, we only have a discrete set of data
fk f (tk ) known at time intervals:
kT
,
2N

k = 0, 1, 2, . . . , 2N 1.

(2)

That is, the interval [0, T ] is divided into 2N equally


spaced points and f is known only on those points. Further, define the following discrete frequencies:
p =
10 pts

2p
,
T

p = 0, 1, 2, . . . , 2N 1.

(3)

Sketch the points fk for tk [0, 2T ] for your choice


of N .

(4)

Calculate the discrete Fourier transform


F (p ) =

10 pts

Sketch a few periods of this function. Label your


axes.

tk =

10 pts

(eip tm ) eip tk

= mk .
2N
2N

2N
1
X
1
f (tk )eip tk F[f (tk )].
2N k=0

(5)

Prove that the inverse transformation f (tk ) =


F 1 [F (p )] is given by

f (tk ) =

2N
1
X
1
F (p )eip tk F 1 [F (p )].
2N p=0

(6)

III. CONTINUOUS FOURIER TRANSFORM [20 pts]

Consider an arbitrary Fourier transform pair g(t) and


G() Let the forward Fourier transform F () = F[g(t)]
be
Z
1
eit g(t) dt.
(7)
G() =
2
Prove that the inverse transformation g(t) = F 1 [G()]
is given by
Z
1
g(t) =
eit G() dt.
(8)
2
That is g(t) = F 1 {F[g(t)]}.

You might also like