Performance of two-dimensional parametric STAP: KASSPER airborne radar data analysis

We investigate a new class of 2D parametric models for space-time adaptive processing (STAP) of ground clutter in airborne radar using KASSPER Dataset 1. The signal-to-interference-plus-noise (SINR) degradation with respect to the optimal receiver is analyzed for various parametric models, regularizations and training sample volumes. We show that a very small training sample volume (5-15 training range bins), with a suitable parametric model and estimation technique, can give acceptable STAP performance for the KASSPER scenario.