Resonant Mode Feature Extraction via a Compressive Multi-Frequency Linear Sampling Method

Recent research concerning qualitative inverse scattering approaches, and in particular the Linear Sampling Method (LSM), has revealed an interesting connection to interior Eigenmodes of scatterer(s) embedded in a known background. These modes have some desirable properties that are amenable for automatic target recognition (ATR) applications; chiefly scale, translation, and rotation invariance. In this work we apply a space-frequency regularized LSM to extract and exploit this connection for generating resonant based features. We establish initial proof-of-concept of the proposed approach using numerically generated multi-static data.