Abstract—An optimum frequency resolution is useful for extracting features of any signals for further analysis. The purpose of this research was to determine the best frequency resolution of Surface Electromyography (sEMG) signals by using a sliding window with Discrete Fourier Transform to produce spectrogram plots. The process was carried out in two stages of investigations. The first investigation was to hold the sampling frequency constant at 8192 Hz and to use a new algorithm with sliding window sizes ranging from 16 to 512 samples through the signal. The results showed that the spectrogram that produced the best visual frequency resolution was with a window size of 64 samples. The calculated time period of sampling frequency of 8192 Hz with window size of 64 samples is hence 1/128 seconds. The second investigation was to use the time period of 1/128 seconds found in the investigation one. This time period of 1/128 second is held constant in order to determine window sizes that passed through different sampling frequencies which were set at 1024 Hz, 2048 Hz and 4096 Hz. Hence the calculated window sizes sample values are 8, 16 and 32 respectively. The spectrogram was plotted for each window sizes and it was found that a window size of 32 samples with the sampling frequency at 2048 Hz gave the best visual frequency resolution for the analysis of sEMG signals.
Index Terms—Fourier, spectrogram, electromyography, signal Processing.
J. Kilby and K. Prasad are with the School of Engineering at AUT University, Auckland, New Zealand (e-mail: jkilby@ aut.ac.nz, kprasad@laut.ac.nz).
[PDF]
Cite: J. Kilby and K. Prasad, "Analysis of Surface Electromyography Signals Using Discrete Fourier Transform Sliding Window Technique,"
International Journal of Computer Theory and Engineering vol. 5, no. 2, pp. 321-325, 2013.
