Super Resolution Multi-target Localization Method for Overlapped Magnetic Anomaly Field

Abstract

Significant progress has been made in the isolated ferrous target localization based on magnetic anomaly field. However, localizing multiple nearby targets remains an ongoing challenge, especially in cases of overlapped magnetic anomaly fields. Here, we propose a super resolution multi-target localization method based on the normalized source strength (NSS) data. Multiple uniformly distributed magnetic dipoles are preset in the solution space, so the measured NSS is the is the sum of the NSS generated by all magnetic dipoles, and an underdetermined linear system is established for the unknown vector of magnitude of magnetic moment of all diploes. Since the targets are sparse in the solution domain, $L1$ -minimization technique is used to estimate the nonzero dipoles, thereby determining the number and position of the targets. Experiments and results have been demonstrated, indicating the target resolution is the interval of the magnetic dipoles in subsurface space, which is smaller than the spatial resolution of the magnetic anomaly field, and when the magnetic anomaly field is aliased, the proposed approach is able to estimate the number, position, and magnetic moment of the targets.