Compared performances of MF-based and locally optimal-based magnetic anomaly detection

Although magnetic dipole detection was mainly investigated in the eighties, new interest for the question has been recently marked in literature. The aim of this paper is to revisit the different existing approaches based on the knowledge of the physics underlying the detection problem, and to give/recall the mathematical background on which the detection scheme is based. Thus, assessment of the detection method is possible, as well as the influence of the physical parameters on detector performances. Furthermore, the paper proposes to use a 3D information (measurement of the magnetic induction anomaly) instead of the common 1D information (use of a scalar sensor providing the total field modulus), both cases being encompassed in the same formalism. Finally, the case where noise is not Gaussian is investigated and a locally optimal receiver is studied and tested.