Title: Classification of sums of complex exponentials

Abstract: Numerous signals in relevant signal processing applications can be modeled as a sum of complex exponentials. Each exponential term entails a particular property of the modeled physical system, and it is possible to define families of signals that are associated with the complex exponentials. In this paper, we formulate a classification problem for this guiding principle and we propose a data processing strategy. In particular, we exploit the information obtained from the analytical model by combining it with data-driven learning techniques. As a result, we obtain a classification strategy that is robust under modeling uncertainties and experimental perturbations. To assess the performance of the new scheme, we test it with experimental data obtained from the scattering response of targets illuminated with an impulse radio ultra-wideband radar.