FAST Fourier Transform(FFT)

FAST FOURIER TRANSFORM(FFT)
FOURIER TRANSFORM(FT)
• Fourier transform comes under the fourier analysis.
• Fourier analysis is used to convert time domain function into
frequency domain function.
• The function representation is either energy function or stable
function then fourier transform is used.
• In time domain simplification is tough compare to frequency
domain simplification.
• The device used to convert time domain to frequency domain is
called spectrum analysis.
BASIC FORMULA USED IN FT
Time domain Frequency domain
x(t) X[f] or X[w]
Fourier Transform of x(t)=[X(f)] or X(w)
FT of [x(t)] = X(w)=
IFT of [X(w)] = x(t)=
FREQUENCY RESPONSE
• The frequency response of a system is the quantitative measure of

the magnitude and phase of the output as a function of input
frequency.
• Magnitude is changing w.r.t to frequency.
• Phase is changing w.r.t to frequency.
• To find the exact magnitude and phase response of any signal or
system the F.T is used.
• The frequency response is widely used in the design and analysis of
systems, such as audio and control system.
WHAT IS DFT?
• The Discrete Fourier Transform is a very important
mathematical tool which is used for discrete-time signal
processing.
• It is used to convert time domain signal into frequency
domain.
• If the given sequence is finite then the DFT is used as the
transform.
• DFT has many applications in digital signal processing such
as linear filtering, spectrum analysis etc.
EQUATION OF DFT
• Equation of DFT,
X(k) =
N = length of input sequence
x(n) = input sequence
K , n = 0,1,2……N-1
DECIMATION-IN-TIME FFT ALGORITHM
• The principle of decimation-in-time is most conveniently illustrated

by considering the special case of N an integer power of 2.
• Since N is an even integer, we can consider computing X(k) by

separating x(n) into two N/2-point sequences consisting of the
even-numbered points in x(n) and the odd-numbered points in x(n).
With X(k) given by
• Separating x(n) into even-numbered and odd-
numbered points, we obtain
or with the substitution of variables n=2r for an

even and n=2r+1 for odd ,
DECIMATION IN TIME (RADIX 2)
Let’s explore an approach that subdivides time interval into
successively smaller intervals. Assume N, the size of the DFT,
is an even integer so that N/2 is also an integer. Define two
auxiliary signals:
a[n]  x[2n], n 0, 1, 2, ..., N / 2  1
• bLet Ak and Bk denote
[n]  x[2n  1],
two (N/2)-point DFTs of a[n] and b[n]:
n 0, 1, 2, ..., N / 2  1
( N / 2) 1
Ak   a[n]W
n 0
kn
N /2 , k 0, 1, ..., N / 2  1
( N / 2) 1
Bk   b[
k 0
n ]W  kn
N /2, k 0, 1, ..., N / 2  1
EXAMPLE ON DFT
• If x(n) = {1,1,0,0}, find DFT of given input sequence.
Solution:-
N=4
n = N-1=> 4-1 =3
n = 0,1,2,3
X(k) =
X(k) =
K=0,1,2,……N-1
K=0,1,2,3
X(K=0)=
=
=
= X(0)+X(1)+X(2)+X(3)
= 1+1+0+0
=2
SIMILARLY,
X(K=1)=1-J, X(K=2)=0, X(K=3) =1+J

X(K)= {2,1-J,0,1+J}
WHAT IS FFT?
• A fast fourier transform(FFT) is an algorithm that computes the

discrete fourier transform(DFT) of a sequence or its
inverse(IDFT).
• The FFT employed in the field of Digital signal processing(DSP)
like filtering, spectral analysis and so on.
• The FFT plays vital role in digital communication such as digital
video broadcasting, Orthogonal Frequency Division Multiplexing (OFDM) .
• Discrete implementation of DFT has O(N2) complexity, which reduces to
O(Nlog2N) through FFT.
RADIX-2 FFT ALGORITHMS
• Let N=2v; then we choose M=2 and L=N/2 and divide x(n)
into two N/2-point sequence.
• This procedure can be repeated again and again. At each
stage the sequences are decimated and the smaller DFTs
combined. This decimation ends after v stages when we have
N one-point sequences, which are also one-point DFTs.
• The resulting procedure is called the decimation-in-time FFT
(DIT-FFT) algorithm, for which the total number of complex
multiplications is: CN=Nv= N*log2N;
• using additional symmetries: CN=Nv= N/2*log2N

• The FFT algorithm is classified into two: Decimation In
Time (DIT) and Decimation In Frequency (DIF)
algorithms.
• In decimation in time, the input sequence is divided

into smaller sequences and the outputs are combined
in a specific pattern to obtain the intended DFT output.
• In the decimation in frequency approach, the frequency

samples of the DFT are divided into smaller sub-
sequences in a certain manner.
• DIT and DIF algorithms are further classified as: Radix-

2, Radix- 4, Radix-8, Split-Radix [3].
• The decimation-in-time (DIT) radix-2 FFT recursively

partitions a DFT into two DFTs that have length half of
the total length of sequence such as even-indexed and
odd- indexed time samples.
• The N-point data sequence is splitted into two point data sequences
F1(k) and F2(k), corresponding to the even- numbered and odd-
numbered samples of x(n) respectively.
Basic Butterfly in DIT FFT Algorithm
• The basic computational method is to take two complex numbers, i.e.,

the pair (a, b). Then do the multiplication of b and then add and
subtract the product value from a to form two new complex numbers (A,
B). This basic calculation is called a butterfly.
• The name Radix-2 is given because of its base is equals to 2 and the
representation is , here M represents the stage.
• A 32-points FFT can be computed in 5 stages where each stage consists

of 16 butterfly operations. Butterfly operation in a Radix-2 algorithm is
that part of FFT computation.
• In this case 32-points DFT can be divided into sixteen 2-point DFTs.
Then these can be combined to get eight 4-point DFTs. Further they are
combined to get four 8-point DFTs. These are combined to get two 16-
point DFTs. Finally, we get 32- point DFT.
DETAIL OF “BUTTERFLY”
.. THE COMPUTATIONAL FLOW OR THE
SIGNAL FLOW IN COMPUTING X(K) FOR AN
8-POINT SEQUENCE, I.E. N=8 IS SHOWN IN
FIGURE BELOW.
BIT REVERSAL
The bit reversal algorithm used to perform the re-ordering of

signals.
The decimal index, n, is converted to its binary equivalent.
The binary bits are then placed in reverse order, and converted
back to a decimal number.
Bit reversing is often performed in DSP hardware in the data
address generator (DAG).
DIT ALGORITHM
• Result is the sum of two N/2 length DFTs

X [k ]  G
[k ]  WNk  H
[k ]
N/2 DFT N/2 DFT
of even samples of odd samples
Then repeat decomposition of N/2 to N/4 DFTs, etc.

N/2
• x[0,2,4,6] DFT
• N/2
• x[1,3,5,7] DFT X[0…7]
WN0...7
DECIMATION-IN-TIME ALGORITHM
• Consider expressing DFT with even and odd input samples:

N1
X [k ]  x[n]WNnk
n 0
  x[n]WNnk   x[n]WNnk
n even n odd
N 1 N 1
2 2
 x[2r ](WN2 ) rk  WNk  x[2r  1](WN2 ) rk

r 0 r 0
N 1 N 1
2 2
 x[2r ]WNrk/ 2  WNk  x[2r  1]WNrk/ 2

r 0 r 0
DECIMATION-IN-TIME FFT ALGORITHM
DECIMATION-IN-FREQUENCY FFT
ALGORITHM
FFT VS. DFT
• The FFT is simply an algorithm for efficiently calculating the

DFT.
• Computational efficiency of an N-Point FFT:
• DFT: N2 Complex Multiplications
• FFT: N (N/2) log2DFT
(N) Complex
FFT Multiplications
FFT efficiency
Multiplication Multiplication
256 65,536 1,024 64:1
512 262,144 2,304 114:1
1024 1,048,576 5,120 205:1

2048 4,194,304 11,264 372:1
4096 16,777,216 24,576 683:1
APPLICATION OF FFT
• FFT is most widely used in digital signal processing(DSP) algorithm.
• It is the basic building for a large percentage of algorithm in
current usage.
• Spectrum analysis is used to detecting and analysing the signals.
• Audio and speech signals are often coded in the frequency domain
using FFT.
• Another recent application is in a modulation scheme called OFDM
which is used for digital TV broadcasting and digital radio (audio)
broadcasting(DAB).
• Background noise reduction for mobile telephony, speech and
audio signals is often implemented in the frequency domain using
FFTs.
ADVANTAGES OF FFT
• Fast capture waveform: The waveform is analysed digitally
and can be captured in short time. This short capture time
can have many advantages, it can allow for the capture of
transient or short lived waveforms.
• Able to analyse the signal phase: As part of signal capture
process, data is gained which can be processed to reveal the
phase of signals.
• Waveforms can be stored using FFT technology. It is possible
to capture the waveform and analyse it later should this be
required.
DISADVANTAGES OF FFT
• Frequency limitation: The main limit of frequency and bandwidth of FFT
spectrum analyzers is the analog to digital converter, ADC that is used to
convert the analog signal into digital format. While technology is improving
this component still places a major limitation on the upper frequency limits
or the bandwidth if a down-conversion stage is used.
• Cost: The high level of performance required by the ADC means that this
item is a very high cost item. In addition to all the other processing and
display circuitry required, this results in the costs rising for these items
THANK YOU !

FAST Fourier Transform(FFT)

Uploaded by

Copyright:

Available Formats

FAST Fourier Transform(FFT)

Uploaded by

Document Information

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

FAST Fourier Transform(FFT)

Uploaded by

Copyright:

Available Formats

FAST FOURIER TRANSFORM(FFT)

• The frequency response of a system is the quantitative measure of

• The principle of decimation-in-time is most conveniently illustrated

• Since N is an even integer, we can consider computing X(k) by

or with the substitution of variables n=2r for an

X(K=1)=1-J, X(K=2)=0, X(K=3) =1+J

• A fast fourier transform(FFT) is an algorithm that computes the

• using additional symmetries: CN=Nv= N/2*log2N

• In decimation in time, the input sequence is divided

• In the decimation in frequency approach, the frequency

• DIT and DIF algorithms are further classified as: Radix-

• The decimation-in-time (DIT) radix-2 FFT recursively

Basic Butterfly in DIT FFT Algorithm

• The basic computational method is to take two complex numbers, i.e.,

• A 32-points FFT can be computed in 5 stages where each stage consists

The bit reversal algorithm used to perform the re-ordering of

• Result is the sum of two N/2 length DFTs

Then repeat decomposition of N/2 to N/4 DFTs, etc.

• Consider expressing DFT with even and odd input samples:

 x[2r ](WN2 ) rk  WNk  x[2r  1](WN2 ) rk

 x[2r ]WNrk/ 2  WNk  x[2r  1]WNrk/ 2

• The FFT is simply an algorithm for efficiently calculating the

512 262,144 2,304 114:1

1024 1,048,576 5,120 205:1

You might also like

Pfad - The Proxy pFad of © 2024 Garber Painting. All rights reserved.