FAST FOURIER TRANSFORM

FAST FOURIER TRANSFORM

Carnegie Mellon

Slide 2

ECE Department

Carnegie Mellon

Slide 3

ECE Department

Decimation in time (DIT) FFT

Carnegie Mellon

Slide 4

ECE Department

Decimation in time (DIT) FFT

Carnegie Mellon

Slide 5

ECE Department

Decimation in time (DIT) FFT

Carnegie Mellon

Slide 6

ECE Department

Decimation in time (DIT) FFT

Carnegie Mellon

Slide 7

ECE Department

Decimation in time (DIT) FFT

Carnegie Mellon

Slide 8

ECE Department

The decimation-in-frequency (DIF) FFT algorithm

 Introduction: Decimation in frequency is an alternate way of

developing the FFT algorithm

 It is different from decimation in time in its development,

although it leads to a very similar structure

Carnegie Mellon

Slide 9

ECE Department

The decimation in frequency FFT (continued)

 Consider the original DFT equation .
N 1

X[k] !

n !0

nk x[n]WN

 Separate the first half and the second half of time samples:
( N / 2) 1

[k] !

n !0

nk x[n]WN 

N 1

n! N / 2

nk x[n]WN ( N / 2) 1

( N / 2)1

n !0

( nk x[n]WN  WNN / 2)k

n !0

nk x[n  ( N / 2)]WN

n !0 ( N / 2)1

nk ?x[n]  ( 1)k x[n  ( N / 2)]AWN

 Note that these are not N/2-point DFTs

Carnegie Mellon
Slide 10 ECE Department

Continuing with decimation in frequency ...

(N / 2) 1

X[k] !

n !0 (N / 2)1

nk x[n]  (  1) k x[n  ( N / 2)] WN

 For k even, let k ! 2r

X[k] !

n !0

x[n]  ( 1)

2r

n2r x[n  ( N / 2)] WN !

( N / 2) 1 n !0

nr ?x[n]  x[n  ( N / 2)]AWN / 2

 For k odd, let k ! 2r  1

( N / ) 1

X[k] !

n !0 ( N / )1

n( x[n]  (1) r (1)x[n  (N / )] WN r 1)

 These expressions are the N/2-point DFTs of

Carnegie Mellon
Slide 11

n!0

n nr ?x[n]  x[n  (N / )]AWN WN /

n x[n]  x[n  ( N / 2)] and [x[n]  x[n  ( N / 2)]]WN

ECE Department

Continuing with decimation in frequency ...

Carnegie Mellon

Slide 12

ECE Department

Continuing with decimation in frequency ...

Carnegie Mellon

Slide 13

ECE Department

Continuing with decimation in frequency ...

Carnegie Mellon

Slide 14

ECE Department

Continuing with decimation in frequency ...

Carnegie Mellon

Slide 15

ECE Department

 and replacing the N/4-point DFTs by butterflys we obtain

Carnegie Mellon

Slide 16

ECE Department

The DIF FFT is the transpose of the DIT FFT

 To obtain flowgraph transposes:
Reverse direction of flowgraph arrows Interchange input(s) and output(s)

DIT butterfly:

DIF butterfly:

 Comment:
We will revisit transposed forms again in our discussion of filter implementation
Carnegie Mellon
Slide 17 ECE Department

The DIF FFT is the transpose of the DIT FFT

 Comparing DIT and DIF structures:
DIT FFT structure: DIF FFT structure:

 Alternate forms for DIF FFTs are similar to those of DIT FFTs
Carnegie Mellon
Slide 18 ECE Department

Alternate DIF FFT structures

 DIF structure with input natural, output bit-reversed

Carnegie Mellon

Slide 19

ECE Department

Alternate DIF FFT structures (continued)

 DIF structure with input bit-reversed, output natural

Carnegie Mellon

Slide 20

ECE Department

Summary

 This morning we considered a number of alternative ways of

computing the FFT:
The decimation-in-time structure The decimation-in-frequency structure Using standard FFT structures for inverse FFTs

Carnegie Mellon

Slide 21

ECE Department