A Tracking Method For Target Detection And Location Of Magnetic Anomalies

With the deepening of people's understanding and widespread application of magnetism,the detection and localization of magnetic objects has attracted more and more attention.The intuitive understanding of magnetic detection and positioning is mainly the two steps of detection and positioning.Since the magnetic anomaly signal is very weak,magnetic signal extraction is a very broad research field.It is divided into many different methods according to different principles,and the positioning is similar to other kinds of targets.There are various methods to achieve.The most widely used detection method for heteromagnetic signals is blind source separation,which is an effective algorithm for conventional signal processing.However,the blind source separation algorithm has two uncertainties,amplitude uncertainty and order uncertainty,which are verified by simulation experiments.In addition,the heteromagnetic signal is generally weak,which makes the blind source separation difficulty and error larger.For signal detection,based on the general standard orthogonal basis function decomposition,the standard orthogonal basis function decomposition algorithm is improved for the separation of the alien magnetic signals.For tracking positioning,it is based on the Gaussian particle filter algorithm.An improved Cubature Kalman Gaussian particle filtering algorithm is presented.Generally,the measured complex signal can be regarded as a combination of several base signals under specific coefficients.This is the theoretical basis of the standard orthogonal basis function decomposition algorithm.The energy function of the signal is normalized,and then the characteristic time of the target signal can be seen through simulation.According to this characteristic,the base function coefficients can be calculated and the separation of the signal can be completed.On the basis of theoretical analysis,in fact,because of the phase difference of the base function signal,the coefficients obtained by the theory will be very erroneous or even wrong.The improvement of the standard orthogonal basis function decomposition algorithm is that the original three basis functions are combined according to different weights to form a new basis function,which organically combines the new basis signal function with the original basis signal function.The simulation results show that introducing a new basis function will make the energy function of the signal more SNR,easier to process,easier to obtain the coefficient of the basis function,and then complete the decomposition of the source signal.Particle filter algorithm has more advantages than traditional tracking and location methods in target tracking and location.However,the basic idea of particle filter determines that it also has some shortcomings,the degradation of particles is the most obvious phenomenon.The improved Gauss particle filter based on particle filter algorithm has more resampling links than the standard particle filter algorithm,resampling is an important measure to alleviate particle degradation.After understanding the basic theory of the Gauss particle filter algorithm,the method and effect of the algorithm in dealing with the degradation phenomenon and the exhaustion of particle diversity are analyzed.Then the Gauss particle filter algorithm is improved,and the simulation experiment is carried out.Given the non-linear state model,the simulation results are compared between the Gauss particle filter algorithm and the volume Kalman-Gauss particle filter algorithm.Experiments show that the volume Kalman-Gauss particle filter is more effective and stable than the Gauss particle filter in trajectory prediction.