Uxo Discrimination Using Blind Source Separation

Statistical signal processing techniques have shown progress in discriminating UXO from clutter<br>when the objects occur in isolation. Under this condition, only a single object contributes to the sensor<br>measurement. For multiple closely-spaced subsurface objects, however, the unprocessed sensor measurement<br>is a mixture of the responses from several objects. Consequently, the unprocessed measurements<br>cannot be used directly to discriminate UXO from clutter. In this paper, we implement blind source separation<br>(BSS) techniques, specifically independent component analysis (ICA), to recover the unobserved<br>object signatures from the mixed measurement data obtained by electromagnetic induction (EMI) sensors,<br>and then use the recovered signatures for UXO/clutter discrimination. Discrimination performance<br>depends on multiple factors, including the number of clutter objects in proximity to the UXO and the<br>separation distance between the UXO and clutter. Simulation results are presented illustrating the impact<br>of these factors on discrimination performance.