Basis pursuit for spectrum cartography

A nonparametric version of the basis pursuit method is developed for field estimation. The underlying model entails known bases, weighted by generic functions to be estimated from the field’s noisy samples. A novel field estimator is developed based on a regularized variational least-squares (LS) criterion that yields estimates spanned by thin-plate splines. Robustness considerations motivate well the adoption of an overcomplete set of basis functions, together with a sparsity-promoting regularization term, which endows the estimator with the ability to select a few of these bases that “better” explain the data. This parsimonious field representation becomes possible because the sparsity-aware spline-based method of this paper induces a group-Lasso estimator of the thin-plate spline basis expansion coefficients. The novel spline-based approach to basis pursuit is motivated by a spectrum cartography application, in which a set of sensing cognitive radios collaborate to estim...

• Type: Journal
• Year: 2011
• Where: ICASSP
• Authors: Juan Andrés Bazerque, Gonzalo Mateos, Georgios B. Giannakis
• Keywords: Basis Pursuit, ICASSP 2011, Noisy Samples, Rf Power, Signal Processing