Download presentation
Presentation is loading. Please wait.
Published byStanley Dorsey Modified over 9 years ago
1
Numerical Analysis – Digital Signal Processing Hanyang University Jong-Il Park
2
Division of Electrical and Computer Engineering, Hanyang University Digital Signal Processing Discrete Fourier Transform Fast Fourier Transform(FFT) Multi-dimensional FFT Convolution
3
Division of Electrical and Computer Engineering, Hanyang University Sampling and aliasing
4
Division of Electrical and Computer Engineering, Hanyang University Discrete Fourier Transform Fourier Transform Discrete Fourier Transform DFT: IDFT:
5
Division of Electrical and Computer Engineering, Hanyang University Fast Fourier Transform(FFT) [Danielson&Lanczos][Cooley&Tukey]
6
Division of Electrical and Computer Engineering, Hanyang University Decimation-in-time FFT Cooley-Tukey Algorithm
7
Division of Electrical and Computer Engineering, Hanyang University Sande-Tukey Algorithm
8
Division of Electrical and Computer Engineering, Hanyang University Decimation-in-frequency FFT(I) Sande-Tukey Algorithm
9
Division of Electrical and Computer Engineering, Hanyang University Decimation-in-frequency FFT (II)
10
Division of Electrical and Computer Engineering, Hanyang University Why FFT? Further reading: http://en.wikipedia.org/wiki/Cooley-Tukey_FFT_algorithm
11
Division of Electrical and Computer Engineering, Hanyang University Computation of FFT(I) input and output of four1() in NR in C
12
Division of Electrical and Computer Engineering, Hanyang University Computation of FFT(II) Eg. FFT
13
Division of Electrical and Computer Engineering, Hanyang University 2D FFT(I)
14
Division of Electrical and Computer Engineering, Hanyang University 2D FFT(II) * Generalization to L-dimension
15
Division of Electrical and Computer Engineering, Hanyang University 2D FFT(III) Eg. 2D FFT
16
Division of Electrical and Computer Engineering, Hanyang University Convolution(I) Def.
17
Division of Electrical and Computer Engineering, Hanyang University Convolution(II) Convolution theorem o direct convolution complex computation o FFT and multiplication less computation
18
Division of Electrical and Computer Engineering, Hanyang University Convolution(III) Convolution of discrete sampled function
19
Division of Electrical and Computer Engineering, Hanyang University Convolution(IV) Trouble in using DFT of finite duration End effects Treated by zero padding End effect
20
Division of Electrical and Computer Engineering, Hanyang University Convolution(V) Zero padding
21
Division of Electrical and Computer Engineering, Hanyang University Convolution(VI) Convolving very large data sets
Similar presentations
© 2024 SlidePlayer.com. Inc.
All rights reserved.