Sparse reconstruction based reweighted non-convex optimization using homotopy-DCD algorithm

This paper proposed a simple, efficient and easy to hardware implementation iteratively reweighted algorithm to improve the recovery performance for sparse reconstruction from compressive sensing (CS). The algorithm focusses on solving the weighted non-convex penalty minimization problem using homotopy and dichotomous coordinate descent (DCD) algorithm compared with the traditional l₁ norm and weighted l₁ norm optimization. The numerical experiential results demonstrate that the new proposed method outperforms in signal reconstruction quality.