Sparse Signal Recovery Using Markov Random Fields

Compressive Sensing (CS) combines sampling and compression into a single subNyquist linear measurement process for sparse and compressible signals. In this paper, we extend the theory of CS to include signals that are concisely represented in terms of a graphical model. In particular, we use Markov Random Fields (MRFs) to represent sparse signals whose nonzero coefficients are clustered. Our new model-based recovery algorithm, dubbed Lattice Matching Pursuit (LaMP), stably recovers MRF-modeled signals using many fewer measurements and computations than the current state-of-the-art algorithms.

• Type: Conference
• Year: 2008
• Where: NIPS
• Authors: Volkan Cevher, Marco F. Duarte, Chinmay Hegde, Richard G. Baraniuk
• Keywords: Compressible Signals, Information Technology, Linear Measurement Process, NIPS 2008, Sparse Signals