Acoustic discrimination between aircraft and land vehicles using nonlinear dynamical signal models

Abstract

One of the most important applications of nonlinear dynamics is the estimation of empirical dynamical models from data, in order to explain time series derived from physical processes. Such derived models can then be used for a variety of data processing applications, in particular for detection and classification problems. Previously, we presented a theory and numerical approach for the estimation of such nonlinear dynamical models in the detection and classification of acoustic data. Here, we apply these ideas to perform discrimination between aircraft and land vehicles based on acoustic signature recordings, derived from a field test of the USA Army.