Highly Nonlinear Approximations for Sparse Signal Representation


SPMPTrgFFT Greedy Pursuit Algorithm

SPMPTrgFFT is a dedicated Self Projected Matching Pursuit (SPMP) algorithm, for sparse spectral modeling of music sound.

It enables the sparse representation of a piece of music signal, as a linear superposition of spectral components.

The algorithm is tailored to be applied with trigonometric dictionaries. Its distinctive feature being that it avoids the need for the actual construction of the whole dictionary, by implementing the required operations via the Fast Fourier Transform.

The achieved sparsity is theoretically equivalent to that rendered by the Orthogonal Matching Pursuit method. The contribution of the proposed dedicated implementation is to extend the applicability of the standard Orthogonal Matching Pursuit algorithm, by reducing its storage and computational demands.

The algorithm details are given in the paper:

"A Dedicated Greedy Pursuit Algorithm for Sparse Spectral Modeling of Music Sound"
by Laura Rebollo-Neira and Gagan Aggarwal

The MATLAB routines for implementing the SPMPTRrgFFT method and reproduce the numerical examples in the paper are available in the archive SPMPTrgFFT_Web.zip.