Magnetic anomaly signal processing method based on periodic potential stochastic resonance system

The magnetic anomaly signals of dynamic targets are transient and accidental, making the signal processing problem of magnetic anomaly detection very challenging. The stochasticresonance method has attracted wide attention due to its unique advantages in weak signal detection. The effectiveness of the stochastic resonance method depends on the selection of structural parameters and the degree of matching the initial value and the signal morphology. The proposed stochastic resonance method based on periodic potential enables robust processing of diverse magnetic anomaly signals without dependence on initial value configurations. Compared to the parallel stochastic resonance method, it achieved a 50 % improvement in computational efficiency and a 56 % increase in successful multi-target detection rates. Additionally, a signal scaling factor calculation method based on three-dimensional magnetic signals was employed to enhance processing reliability. Experimental results demonstrate that the proposed method achieves a 10.13 dB improvement in SNR compared to the original input signal. Furthermore, the processed signal waveform exhibits the highest similarity to the true signal, outperforming other classical methods and showing its superiority in magnetic anomaly signal processing.