A Generalized Framework for Learning and Recovery of Structured Sparse Signals

Loading...
Thumbnail Image

Date

2012-02

Journal Title

Journal ISSN

Volume Title

Publisher

Submitted to 2012 IEEE Statistical Signal Processing Workshop

Research Projects

Organizational Units

Journal Issue

Abstract

We report on a framework for recovering single- or multi-timestep sparse signals that can learn and exploit a variety of probabilistic forms of structure. Message passing-based inference and empirical Bayesian parameter learning form the backbone of the recovery procedure. We further describe an object-oriented software paradigm for implementing our framework, which consists of assembling modular software components that collectively define a desired statistical signal model. Lastly, numerical results for an example structured sparse signal model are provided.

Description

Engineering: 1st Place (The Ohio State University Edward F. Hayes Graduate Research Forum)

Keywords

compressed sensing, structured sparse signal recovery, time-series, sparse linear regression, message passing, approximate message passing, multiple measurement vector, loopy belief propagation, graphical models, probabilistic signal processing

Citation