A Short-Term Measure of Dark Energy in Sparse Atomic Estimations

A sparse atomic estimation involves finding a representation of a signal that uses terms with finite support drawn from an overcomplete dictionary. These terms can constructively and destructively interfere with each other in the reconstruction, and some may even disappear. For this reason, we refer to the interference between atoms in the sparse estimation as "dark energy" (DE). Building upon our previous work, we present a short-term measure of DE (STDE) that can be used to examine the distribution of interference in the signal estimation. We present results for several signals, and discuss some applications of this measure.

[1]  A. Willsky,et al.  HIGH RESOLUTION PURSUIT FOR FEATURE EXTRACTION , 1998 .

[2]  Stéphane Mallat,et al.  Matching pursuits with time-frequency dictionaries , 1993, IEEE Trans. Signal Process..

[3]  Xavier Rodet,et al.  Analysis of sound signals with high resolution matching pursuit , 1996, Proceedings of Third International Symposium on Time-Frequency and Time-Scale Analysis (TFTS-96).

[4]  Rémi Gribonval.  Sparse decomposition of stereo signals with Matching Pursuit and application to blind separation of more than two sources from a stereo mixture , 2002, 2002 IEEE International Conference on Acoustics, Speech, and Signal Processing.

[5]  Martin Vetterli,et al.  Atomic signal models based on recursive filter banks , 1997, Conference Record of the Thirty-First Asilomar Conference on Signals, Systems and Computers (Cat. No.97CB36136).

[6]  Bruno Torrésani,et al.  Hybrid representations for audiophonic signal encoding , 2002, Signal Process..

[7]  Michael A. Saunders,et al.  Atomic Decomposition by Basis Pursuit , 1998, SIAM J. Sci. Comput..