Magnetic anomaly detection using entropy filter

We address the detection of a ferromagnetic target that generates an anomaly in the ambient Earth magnetic field. Detection of an anomaly buried in magnetic noise requires the use of a magnetic anomaly detector (MAD), such as the orthonormal basis functions (OBFs) detector. In contrast to the OBFs detector that relies on target signal waveform ensemble, we propose the adaptive minimum entropy detector (MED) to detect any changes in the magnetic noise pattern. Hence, we have constructed the MED based on the magnetic noise probability density function. The MED was tested on real-world magnetic noise and compared to the OBFs detector. Higher detection rate was exemplified for the MED over the OBFs detector in detecting a ferromagnetic target with low signal-to-noise ratio (SNR). Appreciable advantage of the MED over the OBFs detector is shown, when the target does not move according to the assumed pattern. The low-computational complexity makes the MED appropriate for real time applications.