In this paper, we propose a method, named the cyclic matching pursuit, that outperforms the standard matching pursuit and the perceptual matching pursuit while preserving the computationally efficiency of those methods. We exemplify the application of the method to audio modeling and coding using a perceptual distortion measure and demonstrate using audio signals that the method leads to improved modeling capabilities.
Published in:
Signals, Systems and Computers, 2007. ACSSC 2007. Conference Record of the Forty-First Asilomar Conference on
Date of Conference: 4-7 Nov. 2007