Stap method based on iterative subspace techniques

We address the problem of low-complexity space-time adaptive processing (STAP) with small sample support requirement. Fast iterative subspace techniques as projection approximation subspace tracking deflation (PASTd), modified PASTd (MPASTd), fast approximated power iteration (FAPI), and modified FAPI (m-FAPI) are computationally efficient algorithms for estimating the clutter subspace. We give deep investigation of the performance of these techniques combined with the eigencanceler for clutter suppression in STAP. Simulation results suggest that better performance can be achieved even with a lower clutter subspace dimension when limited training samples are available.