多维参数谱重构的FDA-MIMO雷达超分辨目标定位方法

打开文本图片集

中图分类号：TN957.52 文献标志码：A DOI：10.12305/j.issn.1001-506X.2025.05.09

Abstract：Two low-complexity angle-distance super-resolution estimation algorithms are proposed to solve coupling problem of angle and distance in the target localization process of frequency division array multipleinput multiple-output （FDA-MIMO）radar，which are namely the two-dimensional rooting（2D-root） algorithm and thereduced dimensional estimation of signal parameters via rotational invariance techniques （RD-ESPRIT） algorithm.Firstly，anangle-distance two-dimensional spectral reconstruction method is proposed to perform one-dimensional equivalent stripping of the two-dimensional parameter spectrum.Differ from traditional rooting super-resolution algorithms，the 2D-root algorithm utilizes a matrix subspace rooting method to construct dual rooting polynomials for high-precision estimation of anglesand distances，and automatically pairs the parameters.Meanwhile，in order to further reduce the computational complexity of rooting computation，the

RD-ESPRIT algorithm utilizes the linear relationship inherent in the reconstructed angle spectrum to obtain rotational invariance of the signal subspace，thereby achieving eficient solution of the target angle.Computer simulation experimental results show that the two proposed algorithms can improve parameter estimation accuracy while reducing the complexity of traditional search-based target localization algorithms.

Keywords： frequency diverse array multiple-input multiple-output （FDA-MIMO） radar；super resolution; spectrum reconstruction；decoupling

0 引言

自雷达系统的概念提出以来，其主要作用之一就是对空间目标进行精准定位，即利用先进的信号处理方法从雷达回波中提取目标的角度和距离信息[1]。（剩余14065字）