WIT Press


Readability Of Reassigned Scalograms And Extraction Of Spectra Features For Signal Analysis

Price

Free (open access)

Paper DOI

10.2495/BIO090271

Volume

13

Pages

14

Page Range

287 - 300

Published

2009

Size

1,232 kb

Author(s)

S. Mekaoui, A. Houacine & T. Gharbi

Abstract

In this paper, a wavelet scalogram is used. The wavelet scalogram presents some disadvantages. This is particularly true for time-frequency analysis and representation, which present inconvenient cross-terms. Mechano-myogram (MMG) signals are acquired via a home probe highly sensitive optical sensor. The data obtained from two healthy subjects and two patients tested under drastic conditions are analyzed to characterize the dynamic properties of the MMG and also to determine their frequency contents. We developed the reassignment form of the scalogram, which improves its resolution and readability. A plot of the scalogram contours is also presented to test the direct readability of the scalogram representations. Spectra features are extracted and relevant parameters are assessed, such as the power spectral density, the mean frequency, the average frequency and the well known ratio HF% that characterizes the dynamic characteristics of the tested muscles. For that purpose, the number of subjects had been increased to 24 healthy subjects and up to 18 patients affected by different specific muscular diseases. Keywords: mechano-myogram (MMG) signals, reassigned wavelet scalogram, power spectral density, mean and average frequencies, MMG rms value, MMG average value, HF% ratio value.

Keywords

mechano-myogram (MMG) signals, reassigned wavelet scalogram, power spectral density, mean and average frequencies, MMG rms value, MMG average value, HF% ratio value.