Abstract: Traditional airborne magnetic anomaly detection methods rely on prior knowledge of target signal characteristics or noise statistical characteristics，resulting in poor detection probability and short detection distance.In order to improve the performance of airborne magnetic anomaly detectors，an adaptive magnetic anomaly detection method based on the high-order zero-crossing method is proposed, in which the characteristics of the target signal and the statistical characteristics of noise are considered.The characteristics of target signal and noise are analyzed based on the target signal model of ferromagnetic submarine，the measured high-altitude signal of target，and the measured magnetic noises collected over a long period in stable non-magnetic and complex magnetic environments.The high-order zero-crossing statistic characteristics of target signal and noise are researched.Then the test statistic is proposed based on the first five orders of the high-order zero-crossing sequence of signals，and an adaptive high-order zero-crossing detector suitable for detection in complex environment was designed. Theoretical performance analysis and measured data verification show that the magnetic noises collected in stable non-magnetic and complex magnetic environments are quite different， and the noise cannot be simply regarded as a Gaussian distribution. The high-order zero-crossing characteristics of the ferromagnetic and magnetic noise signals are obviously different， and the first five orders of high-order zero-crossing rate are clearly distinguishable. Under the condition of the same input amplitude signal-to-noise ratio， the amplitude of output signal-to-noise ratio of adaptive high-order zero-crossing detector is greater than 5. The anti-noise capability of the adaptive high-order zero-crossing detector is stronger and has good weak signal detection capacity. The detection performance of adaptive high-order zero-crossing detector is better than that of conventional high-order zero-crossing detector and orthonormal basis functions detector. The adaptive high-order zero-crossing detector is suitable for low signal-to-noise ratio detection.

Key words: submarine, airbornemagneticanomalydetection, signaldetection, high-orderzero-crossingmethod, detectionprobability, performanceanalysis