Buried unexploded ordnance identification via complex natural resonances

A classification technique focused on the identification of buried unexploded ordnance (UXO) using complex natural resonance (CNR) signature is considered. The total least square (TLS) Prony's (1795) method is used to extract CNRs from time-domain data, Full-scale UXO computational models and the body of revolution moment method (BORMM) code is used to obtain the backscattered fields, which are then used to give the theoretical free-space CNRs. Baum's (1993) transformation is used to relate a CNR in a lossy simple medium to the corresponding CNR in free space. A practical UXO classification yields an estimate of the UXO length from its CNR information. Successful UXO classification examples from actual measured data are presented.