Adaptive Compressed Sensing Using Sparse Measurement Matrices

Compressed sensing methods using sparse measurement matrices and iterative message-passing recovery procedures are recently investigated due to their low computational complexity and excellent performance. The design and analysis of this class of methods is inspired by a large volume of work on sparsegraph codes such as Low-Density Parity-Check (LDPC) codes and the iterative Belief-Propagation (BP) decoding algorithms. In particular, we focus on a class of compressed sensing methods emerging from the Sudocodes scheme that follow similar ideas used in a class of sparse-graph codes called rateless codes. We are interested in the design and analysis of adaptive Sudocodes methods and this paper provides initial steps in this direction.