Title :
Sparse stochastic processes: A statistical framework for modern signal processing
Author_Institution :
Biomed. Imaging Group, Swiss Fed. Inst. of Technol. of Lausanne (EPFL), Lausanne, Switzerland
Abstract :
Summary form only given. We introduce an extended family of sparse processes that are specified by a generic (non-Gaussian) innovation model or, equivalently, as solutions of linear stochastic differential equations driven by white Lévy noise. We present the mathematical tools for their characterization. The two leading threads of the exposition are - the statistical property of infinite divisibility, which induces two distinct types of behavior - Gaussian vs. sparse at the exclusion of any other; - the structural link between linear stochastic processes and splines. This allows us to prove that these processes admit a parsimonious representation in some matched wavelet-like basis. We show that these models have predictive power for image compression and that they are applicable to the derivation of statistical algorithms for solving ill-posed inverse problems, including compressed sensing.
Keywords :
Gaussian processes; linear differential equations; signal representation; splines (mathematics); stochastic processes; Gaussian behavior; compressed sensing; generic innovation model; ill-posed inverse problems; image compression; infinite divisibility; linear stochastic differential equations; matched wavelet-like basis; mathematical tools; modern signal processing; parsimonious representation; sparse behavior; sparse stochastic processes; splines; statistical algorithms; white Lévy noise; Abstracts; Biological system modeling; Biomedical imaging; Signal processing; Stochastic processes; Technological innovation;
Conference_Titel :
Systems, Signals and Image Processing (IWSSIP), 2013 20th International Conference on
Conference_Location :
Bucharest
Print_ISBN :
978-1-4799-0941-4
DOI :
10.1109/IWSSIP.2013.6623431