FPGA implementation of a feature extraction technique based on Fourier transform
Authors:
- 1 December 2012
- conference paper
- Published by Institute of Electrical and Electronics Engineers (IEEE)
- p. 1-4
Abstract
In this paper, a feature extraction method based on short-time Fourier transform (STFT) was implemented on FPGA for real-time pattern recognition. The proposed technique was implemented using Xilinx System Generator (XSG) in MATLAB/SIMULINK environment. The response signals obtained during feature extraction process of blue whale call are represented. Also, the classification performance based on the fixed-point XSG implementation is compared to that based on the floating-point MATLAB one using isolated blue whale calls.Keywords
This publication has 5 references indexed in Scilit:
- Serial combination of multiple classifiers for automatic blue whale calls recognitionExpert Systems with Applications, 2012
- Real-Time Speaker Verification System Implemented on Reconfigurable HardwareJournal of Signal Processing Systems, 2012
- Blue whale calls classification using short-time Fourier and wavelet packet transforms and artificial neural networkDigital Signal Processing, 2010
- Pattern recognition methods applied to respiratory sounds classification into normal and wheeze classesComputers in Biology and Medicine, 2009
- Chip design of MFCC extraction for speech recognitionIntegration, 2002