13 the fast fourier transform fft r x y a yr

Info iconThis preview shows page 1. Sign up to view the full content.

View Full Document Right Arrow Icon
This is the end of the preview. Sign up to access the rest of the document.

Unformatted text preview: re (3.1) we have the original time series and the associated DFT (the real and imaginary part). In Figure (3.2) the original time series after zero padding ( zeros) is used to compute the DFT. ' 1 In the following example I show how to pad with zeros a time series. This codes was utilized to generate Figures (3.1) and (3.2). d GF  £¥©£¥b a@"c@¦¨A4 )¨©T(B¨¦9AQ¦¨&$ )0  § S§¥£R E§ ¥£P I GF £CB E§ CB£8 @4¨)4AD))A@9 5 cu¢&¨¦¢)i†4¦¨4 v8 E S§¥£R CB… b 6 e CB£ 1 Wr ‡#))@A¦' 20 6„ur ƒ8€ U W&¨¦@‚# y1 # w1 cu eS§¥£R € x 6U v8 E 6 %tsf# 1 # d 6W¦ r r dU 5 g ¦U ¨qW)pihgd H f&S 1 #  e g © e  6 V VU a`0YXUWQ1  65H 7¤91 8 653 1 7&420 £'©$#! © §¥£¡ )(&%" ¤¦¨¨¦¤¢ 41 The FFT is not a new transform; the FFT is just a fast algorithm to compute DFTs. The FFT is based on the halving trick, that is a trick to compute the DFT of length N time series using the DFT of two sub-series of length N/2. Let’s start assuming that we a have :...
View Full Document

This note was uploaded on 11/29/2012 for the course GEOPHYSICS 426 taught by Professor Sacchi during the Winter '12 term at University of Alberta.

Ask a homework question - tutors are online