Detection algorithm for magnetic dipole target based on CEEMDAN and pattern recognition

Abstract Due to the physical characteristics that the magnetic dipole target signal (MDTS) decays with the third power of distance and the fact that the measured data contain usually environmental magnetic noise such as diurnal variation noise and cultural noise, it is very difficult to detect long-distance magnetic targets. In this paper, the complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) algorithm which is an improved version of empirical mode decomposition (EMD) algorithm and pattern recognition algorithm are combined to construct a novel magnetic target detection algorithm(CEEMDAN-PR). CEEMDAN algorithm can effectively decompose the measured magnetic signal into multiple intrinsic mode functions (IMFs), and reduce the aliasing effect between modes. Then, the pattern formed by the characteristics of magnetic dipole signal is used to match the signal reconstructed by the sum of several IMFs to obtain the optimal reconstructed signal. Some simulation tests illustrate that this algorithm has good detection performance.