CEN352 Digital Signal Processing

16
CEN352 Digital Signal Processing Lecture No. 12 to 14 Department of Computer Engineering, College of Computer and Information Sciences, King Saud University, Riyadh, Kingdom of Saudi Arabia October, 2012 BY Dr. Anwar M. Mirza Office No. 2185 Phone: 4697362 [email protected] or [email protected] / ا رز مي د ي ج م وز ن ا وز ت ك الد

description

By Dr . Anwar M. Mirza Office No. 2185 Phone: 4697362 [email protected] or [email protected]. CEN352 Digital Signal Processing. الدكتور / انور مجيد ميرزا. Lecture No. 12 to 14. Department of Computer Engineering, College of Computer and Information Sciences, - PowerPoint PPT Presentation

Transcript of CEN352 Digital Signal Processing

Page 1: CEN352 Digital Signal Processing

CEN352Digital Signal

Processing

Lecture No. 12 to 14Department of Computer Engineering,

College of Computer and Information Sciences,King Saud University, Riyadh, Kingdom of Saudi Arabia

October, 2012

BY

Dr. Anwar M. Mirza

Office No. 2185Phone: 4697362

[email protected] or [email protected]

ميرزا / مجيد انور الدكتور

Page 2: CEN352 Digital Signal Processing

Fast Fourier Transform

It is a very efficient algorithm for computing the DFT coefficients.

Page 3: CEN352 Digital Signal Processing

x

y

z=x+y x

yz=x - y

-1

xw

z=wx

Definitions of the graphical operations

(N/2)-point DFT

(N/2)-point DFT

x(4)x(5)x(6)x(7)

x(0)x(1)x(2)x(3)

b(0)b(1)b(2)b(3)

a(0)a(1)a(2)a3)

-1-1-1-1

WN0

WN1

WN2

WN3

X(1)X(3)X(5)X(7)

X(0)X(2)X(4)X(6)

The first iteration of the 8-point FFT

Method of Decimation-in-Frequency

Page 4: CEN352 Digital Signal Processing

The second iteration of the 8-point FFT

x(4)x(5)x(6)x(7)

x(0)x(1)x(2)x(3)

-1-1-1-1

X(1)X(3)X(5)X(7)

X(0)X(2)X(4)X(6)

WN0

WN1

WN3

WN2

WN0

WN2

WN2

WN0

-1-1

-1-1

(N/4)-point DFT

(N/4)-point DFT

(N/4)-point DFT

(N/4)-point DFT

Method of Decimation-in-Frequency

Page 5: CEN352 Digital Signal Processing

Block diagram for the 8-point FFT (total twelve multiplications)

X(1)X(3)X(5)X(7)

X(0)X(2)X(4)X(6)

x(4)x(5)x(6)x(7)

x(0)x(1)x(2)x(3)

-1-1-1-1

WN0

WN1

WN3

WN2

WN2

WN0

-1-1

-1-1

WN0

WN0

WN0

WN0

-1

-1

-1

-1

Method of Decimation-in-Frequency

Page 6: CEN352 Digital Signal Processing

Example 4.12

Solution

First Iteration Second Iteration

x(0)=1

x(1)=2

x(2)=3

x(3)=4

4

6

-2

-2 W40 = 1

W41 = -j

-2

2j

10

-2

-2+2j

-2-2j

W40 = 1

W40 = 1

10

-2

-2+2j

-2-2j

00

01

10

11

Bit Index00

10

01

11

Bit Reversal

= X(0)

= X(2)

= X(1)

= X(3)

Page 7: CEN352 Digital Signal Processing

First Iteration Second Iteration

X(0)=10

X(1)=-2+j2

X(2)=-2

X(3)=-2-j2

8

-4

12

j4

12

-4

10

-2

-2+2j

-2-2j

4

12

8

16

00

01

10

11

Bit Index00

10

01

11

Bit Reversal

= x(0)

= x(2)

= x(1)

= x(3)

Solution

Example 4.13

Page 8: CEN352 Digital Signal Processing

The first iteration of the 8-point FFT using Decimation-in-Time

Method of Decimation-in-Time

Page 9: CEN352 Digital Signal Processing

The second iteration of the 8-point FFT using Decimation-in-Time

Method of Decimation-in-Time

Page 10: CEN352 Digital Signal Processing

The 8-point FFT using Decimation-in-Time (12 complex multiplications)

Method of Decimation-in-Time

Page 11: CEN352 Digital Signal Processing

The 8-point IFFT using Decimation-in-Time

Method of Decimation-in-Time

Page 12: CEN352 Digital Signal Processing

The 4-point FFT using Decimation-in-Time

Page 13: CEN352 Digital Signal Processing

The 4-point FFT using Decimation-in-Time

Page 14: CEN352 Digital Signal Processing

That’s it for today !!

Page 15: CEN352 Digital Signal Processing

Example 4.12

Solution

First Iteration Second Iteration

x(0)=6

x(1)=4

x(2)=2

x(3)=0

8

4

4

4 W40 = 1

W41 = -j

4

-4j

12

4

4-4j

4+4j

W40 = 1

W40 = 1

12

4

4-4j

4+4j

00

01

10

11

Bit Index00

10

01

11

Bit Reversal

= X(0)

= X(2)

= X(1)

= X(3)

Page 16: CEN352 Digital Signal Processing

Example 4.12

Solution

First Iteration Second Iteration

x(0)=1

x(1)=2

x(2)=3

x(3)=4

4

6

-2

-2 W40 = 1

W41 = -j

-2

2j

10

-2

-2+2j

-2-2j

W40 = 1

W40 = 1

10

-2

-2+2j

-2-2j

00

01

10

11

Bit Index00

10

01

11

Bit Reversal

= X(0)

= X(2)

= X(1)

= X(3)