Enhanced discrimination among UXO-like targets using extremely low-frequency magnetic fields

We show that the ability to discriminate among unexploded ordnance (UXO) targets is enhanced by extending the measurement range of our continuous-wave (CW) electromagnetic induction (EMI) system down to extremely low frequencies (ELFs). In particular, we demonstrate that for a typical UXO target set, when data in the range 30 Hz to 24 kHz are extended downward toward 1-Hz discrimination performance improves by a factor of 1.5. Measurements and numerical simulations show that very low frequency magnetic fields penetrate deep into the target, enabling one to discriminate between thick- and thin-walled ferrous cylinders that are otherwise visually identical. Data acquisition at extremely low frequencies is a time-consuming process especially if data averaging is required to achieve acceptable signal-to-noise ratio. Therefore, in practice, discrimination improvement using ELF is probably best carried out using a "cued" approach. In a "cued" approach, target locations would be marked using another detection technique (e.g., a simple total-field magnetometer), and the low-frequency CW EMI system would be deployed to these known target locations for classification.