The Novel Method of Magnetic Anomaly Recognition Based on the Fourth Order Aperiodic Stochastic Resonance

In the background of strong interference noise, it is one of the crucial technologies of magnetic anomaly detection system to effectively detect and identify weak magnetic anomaly signals (MAS) gene- rated by targets. The detection and recognition of target signal under very low signal-to-noise ratio(SNR)below -10dB have not been achieved in the existing magnetic anomaly detection technology. To tackle these challenges, this paper proposes a magnetic anomaly recognition method based on fourth-order aperiodic stochastic resonance with the aid of sto- chastic resonance theory. The main nature of new algorithm is a four-layer fusion of single potential well stochastic resonance and has real-time recognition ability for MAS without prior information. When simulated magnetic anomaly signal of single object and mutil-targets were assumed in different condition, effect of parameters of algorithm was analyzed and range of optimal parameters was obtained. Through simulation and experimental verification of targets data, the new algorithm could realize target signal recognition in complex interference environment with SNR lower than -15dB,which demonstrates the efficacy of our proposed method and offer some useful design insights to practical MAD system.

Automated summary

This paper proposes a magnetic anomaly recognition method based on fourth-order aperiodic stochastic resonance, which can effectively detect and identify weak magnetic anomaly signals generated by targets in the background of strong interference noise.