Two-stage Fast Matching Pursuit Algorithm for Multi-target Localization

For large-scale high-dimensional positioning scenes, the massive number of grid points brings challenges to multi-target algorithms based on compressed sensing. To cope with challenges, a fast localization method direction arrival is proposed. A sensing model constructed for DOA sequence measured by nodes. Then, two-stage matching pursuit algorithm presented sparse reconstruction, which consists preliminary estimation and supports rectification. process similar orthogonal adopted get estimate result, but no nonlinear operations employed complexity reduction. Then another iterative carried out rectify chosen in result sequentially. Simulation results verify effectiveness accuracy proposed localization.