Wave-oriented data processing for frequency- and time-domain scattering by nonuniform truncated arrays

Understanding observable-based observables (OBP) phenomenology should make it possible to link observables, extracted from measured or computed data, to the wave mechanisms from which they originated, aiding in data inversion: we have referred to such a scenario as wave-oriented data processing. Thus, in wave-oriented data processing, one first parameterizes the forward problem in terms of a concise set of observables (forward problem), and then signal-processing algorithms are developed to extract these observables from data (inverse problem). To close the loop, we must show that the data processing algorithms extract wave phenomenology which is consistent with the forward OBP. The authors describe the forward and inverse phases of this strategy in greater detail. A general strategy is then implemented for a very specific scattering problem: frequency-domain (FD) and time domain (TD) scattering from periodic and weakly aperiodic arrays. That scattering problem is described and the results of their previous forward modeling are summarized. The remainder of the article emphasizes inverse, wave-oriented data processing. Processing options for phase-space processing are summarized,and the results for FD and TD scattering are presented.<>