Title :
Improved Frequency Resolution DFT Eases Teaching FFT Analysis and Provides Better Amplitude Accuracy
Author :
Keradec, Jean Pierre ; Margueron, Xavier
Author_Institution :
Lab. d´´Electrotechnique de Grenoble, Saint-Martin-d´´Heres
Abstract :
For a signal extension T, DFT and FFT calculate the transform at frequencies 1/T apart. Due to this, amplitude of the transform can show more than 36% of discrepancy relatively to exact analytical Fourier transform and this disagreement looks random. Calculating the transform with a frequency resolution 7 times thinner lowers this difference to within 1%. As a result, obtained spectrums are easier to explain. This high resolution spectrum can be obtained either directly from time file or by interpolating frequency file that results from standard calculation. Both algorithms are supplied and spectrums of simple signals, obtained with standard and improved resolution, are compared
Keywords :
design for testability; fast Fourier transforms; interpolation; spectral analysis; FFT analysis; design for testability; fast Fourier transforms; signal extension; Discrete Fourier transforms; Education; Embedded software; Fourier transforms; Frequency; Oscilloscopes; Power engineering and energy; Signal resolution; Software algorithms; Spectral analysis; DFT calculation; Fourier Transform; high resolution spectrum; student experimentation;
Conference_Titel :
Instrumentation and Measurement Technology Conference, 2005. IMTC 2005. Proceedings of the IEEE
Conference_Location :
Ottawa, Ont.
Print_ISBN :
0-7803-8879-8
DOI :
10.1109/IMTC.2005.1604324