Approximate Methods for Obtaining the Complex Natural Electromagnetic Oscillations of an Object

摘要

The complex natural resonances (poles) of an object are excitation (aspect and polarization) independent. The primary goal of this dissertation is to develop a rational function approximation (RFA) for use in pole extraction from the measured scattering spectrum data of broadband radar systems. Applications of the RFA are demonstrated for a variety of simple as well as complex objects. The results obtained are compared with those poles extracted using other methods. To overcome noise problems, two preprocessing techniques--a sum operator and a zero-phase-shift digital filter are developed. The minimum signal-to-noise ratio (additive) is explored such that reliable pole extraction is still feasible. Finally, RFA are applied to the measured data of complex geometric targets (scale models of aircraft).