Recipes on Hard Thresholding Methods

• Lausanne: EPFL, 2011

We present and analyze a new set of sparse recovery algorithms within the class of hard thresholding methods. We provide optimal strategies on how to set up these algorithms via basic ``ingredients'' for different configurations to achieve complexity vs. accuracy tradeoffs. Simulation results demonstrate notable performance improvements compared to state-of-the-art algorithms both in terms of data reconstruction and computational complexity.

Keywords: Sparse signal recovery, model-based compressive sensing

Reference

• EPFL-WORKING-169253

Record created on 2011-10-10, modified on 2012-03-21