Code covered by the BSD License  

Highlights from
Sprint Race for Fast Butterflies

Be the first to rate this file! 17 Downloads (last 30 days) File Size: 18.9 KB File ID: #40097
image thumbnail

Sprint Race for Fast Butterflies

by

 

30 Jan 2013 (Updated )

Benchmarking of Various FFT Algorithm Implementations Based on Execution Time.

| Watch this File

File Information
Description

The following FFT implementations are provided: 1) Radix-2 DIT Recurcive FFT, 2) In Place Radix-2 DIT Iterative FFT, 3) Radix-2 DIT FFT, 4) Radix-4 DIT FFT, 5) Radix-2 DIT Iterative mex-coded FFT, 6) Split-Radix DIT FFT, 7) Radix-2 DIF FFT. DIT = Decimation In Time, DIF = Decimation In Frequency.

Acknowledgements

Mat Lab Solutions: "Introduction To Digital Signal Processing: A Computer Laboratory Textbook". inspired this file.

Required Products MATLAB
MATLAB release MATLAB 8.0 (R2012b)
Tags for This File   Please login to tag files.
Please login to add a comment or rating.
Comments and Ratings (2)
01 Feb 2013 Ilias Konsoulas

To Mikhail: Yes. The script fft_it2.m is used to generate fft_it2_mex via MatLab's codegen. The procedure is described at the bottom of file fft_it2.m. As can be seen the coding in files fft_it.m and fft_it2.m are identical. The purpose of this "dual" effort is to demonstrate the acceleration obtained via mex coded functions.

01 Feb 2013 Mikhail

Hi! Are you using fft_it2_mex instead of fft_it2 by intention in your bench_fft.m?

Updates
31 Jan 2013

I have only modified the title of this submission.

Contact us