Research on micro-motion feature extraction and auxiliary tracking method of broadband radar space target using atomic norm

Broadband radar can accurately perceive space situations and targets, which is of great significance to ensuring my country's aerospace and space security. Broadband tracking can increase the target data rate and provide richer target information for identification, but it must overcome the problems of dense multi-extended target association caused by increased false alarm density and reduced signal-to-noise ratio. This project is based on the phase-derived ranging method to obtain the fine motion parameters of broadband radar space targets, and uses the atomic norm minimization constraint method to extract target micro-motion features, which is used to assist multi-target tracking to improve tracking performance. On the basis of initially estimating the state of the moving target and obtaining the observed values of the target's micro-motion characteristics, the atomic norm constraint is used to model the sparse characteristics of the target's micro-motion in the frequency domain, and the L1 norm is used to model the sparseness of the error correlation. Study the L1 atomic norm minimization method, convert the problem of accurate estimation of target characteristic state in incomplete and distorted characteristic observation data into a semi-definite programming problem, and study the method of quickly solving the target micro-motion characteristics. The obtained target micro-motion feature information is used to correct the correlation weight for secondary filtering to obtain a more accurate estimate of the target's motion state, thereby improving target tracking performance. The research results of this project can be used to improve the measurement accuracy and target recognition capabilities of my country's broadband radar for space targets.