Non-negative orthogonal greedy algorithms - Archive ouverte HAL Access content directly
Journal Articles IEEE Transactions on Signal Processing Year : 2019

Non-negative orthogonal greedy algorithms


Orthogonal greedy algorithms are popular sparse signal reconstruction algorithms. Their principle is to select atoms one by one. A series of unconstrained least-squares subproblems of gradually increasing size is solved to compute the approximation coefficients, which is efficiently performed using a fast recursive update scheme. When dealing with non-negative sparse signal reconstruction, a series of non-negative least-squares subproblems have to be solved. Fast implementation becomes tricky since each subproblem does not have a closed-form solution anymore. Recently, non-negative extensions of the classical orthogonal matching pursuit and orthogonal least squares algorithms were proposed, using slow (i.e., non-recursive) or recursive but inexact implementations. In this paper, we revisit these algorithms in a unified way. We define a class of non-negative orthogonal algorithms and exhibit their structural properties. We propose a fast and exact implementation based on the active-set resolution of non-negative least-squares and exploiting warm start initializations. The algorithms are assessed in terms of accuracy and computational complexity for a sparse spike deconvolution problem. We also present an application to near-infrared spectra decomposition.
Fichier principal
Vignette du fichier
paper1_hal.pdf (1001.84 Ko) Télécharger le fichier
Origin : Files produced by the author(s)

Dates and versions

hal-02049424 , version 1 (26-02-2019)
hal-02049424 , version 2 (15-07-2019)
hal-02049424 , version 3 (22-08-2019)



Thi Thanh Nguyen, Jérôme Idier, Charles Soussen, El-Hadi Djermoune. Non-negative orthogonal greedy algorithms. IEEE Transactions on Signal Processing, 2019, 67 (21), pp.5643-5658. ⟨10.1109/TSP.2019.2943225⟩. ⟨hal-02049424v3⟩
361 View
1281 Download



Gmail Facebook X LinkedIn More