Abstract
Analysis of nonstationary signals, such as music signals, is a challenging task. The purpose of this study is to explore an efficient and powerful technique to analyze and classify music signals in higher frequency range (44.1 kHz). The pursuit methods are good tools for this purpose, but they aimed at representing the signals rather than classifying them as in Y. Paragakin et al., 2009. Among the pursuit methods, matching pursuit (MP), an adaptive true nonstationary time-frequency signal analysis tool, is applied for music classification. First, MP decomposes the sample signals into time-frequency functions or atoms. Atom parameters are then analyzed and manipulated, and discriminant features are extracted from atom parameters. Besides the parameters obtained using MP, an additional feature, central energy, is also derived. Linear discriminant analysis and the leave-one-out method are used to evaluate the classification accuracy rate for different feature sets. The study is one of the very few works that analyze atoms statistically and extract discriminant features directly from the parameters. From our experiments, it is evident that the MP algorithm with the Gabor dictionary decomposes nonstationary signals, such as music signals, into atoms in which the parameters contain strong discriminant information sufficient for accurate and efficient signal classifications.
Highlights
Since most of the real-world signals are non-stationary, the study and analysis of non-stationary signals is receiving more and more attention in the scientific community
We propose a parametric analysis method to study the atoms obtained from the decomposition and extract the discriminant features from the atom parameters
After a long try and comparing process, the optimum feature set, which brings up the best classification accuracy, is found to be the standard deviation of octave, the median of octave, the standard deviation of innerProdI, the standard deviation of realGG, and the central energy
Summary
Since most of the real-world signals are non-stationary, the study and analysis of non-stationary signals is receiving more and more attention in the scientific community. Time series and frequency spectrum contain all the information about the underlying processes of signals. The best representations of non-stationary processes may not be well presented. Due to the time-varying behavior, techniques which give joint time frequency (TF) information are needed to analyze non-stationary signals. Called basis functions, are signals localized in both time and frequency domains. This signal analysis method devises a joint function of time and frequency, that is, a distribution that will describe the energy density or intensity of a signal simultaneously in time and frequency [2]. Features extracted from TF analysis contain the combined time-frequency dynamics of the given signal, as opposed to features along either the time or the frequency axis alone, as provided by conventional techniques [3]
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call
