电子学报 ›› 2022, Vol. 50 ›› Issue (5): 1150-1158.DOI: 10.12263/DZXB.20210479

• 学术论文 • 上一篇    下一篇

基于矩阵化原子范数的高频雷达距离-多普勒二维估计方法

吕明久1, 马建朝1, 韦旭2, 陈文峰3, 杨军3, 马晓岩3   

  1. 1.空军预警学院雷达士官学校, 湖北 武汉 430019
    2.空军预警学院研究生队, 湖北 武汉 430019
    3.空军预警学院预警技术系, 湖北 武汉 430019
  • 收稿日期:2021-04-14 修回日期:2021-05-21 出版日期:2022-05-25 发布日期:2022-06-18
  • 作者简介:吕明久 男,1985年出生于安徽庐江,现为空军预警学院雷达士官学校讲师,主要从事压缩感知在雷达成像中的应用研究.
  • 基金资助:
    国家自然科学基金(61671469)

Two-Dimensional Range-Doppler Estimation Method for High Frequency Radar Based on Matrix Atomic Norm

LÜ Ming-jiu1, MA Jian-chao1, WEI Xu2, CHEN Wen-feng3, YANG Jun3, MA Xiao-yan3   

  1. 1.Radar NCO School,Air Force Early Warning Academy,Wuhan,Hubei 430019,China
    2.Graduate team,Air Force Early Warning Academy,Wuhan,Hubei 430019,China
    3.Early Warning Technology,Air Force Early Warning Academy,Wuhan,Hubei 430019,China
  • Received:2021-04-14 Revised:2021-05-21 Online:2022-05-25 Published:2022-06-18

摘要:

针对传统合成带宽高频雷达存在相参时间长,传统处理算法二维分辨能力弱等问题,提出一种基于矩阵化原子范数的多周期脉冲内随机稀疏步进频率信号距离-多普勒二维估计方法. 通过在每个脉冲重复时间内连续发射一串由多个载频步进子脉冲组成的脉冲序列来减少相参处理周期. 构造矩阵形式的连续原子集合,将距离-多普勒二维估计转化为矩阵形式的原子范数最小化模型,并基于半正定规划问题进行快速求解. 所提方法有效降低了运算量且实现了网格失配目标的二维高分辨. 理论分析与仿真实验验证了所提方法的有效性.

关键词: 高频雷达, 连续压缩感知, 矩阵化原子范数, 网格失配, Vandermonde分解, 稀疏重构

Abstract:

In order to solve the problems of long coherent time and weak two-dimensional resolution in traditional synthetic bandwidth high frequency radar, a two-dimensional range-Doppler estimation method based on matrix atomic norm for multi-cycle intra-pulse random sparse stepped frequency signal is proposed. Firstly, a series of pulse, which is composed of sequences multiple carriers stepped sub-pulses, has been continuously transmitted in each pulse repetition time to reduce the coherent processing period. Secondly, a continuous set of atoms in matrix form is constructed, and the two-dimensional range-Doppler estimation is transformed into a matrix form of atomic norm minimization problem. The above optimization problem is transformed into a positive semi-definite programming problem, and can be solved. The proposed method can not only effectively reduce the amount of computation, but also achieve the high resolution under the condition of off grid. Theoretical analysis and simulation experiments verify the effectiveness of the proposed method.

Key words: high frequency radar, compressed sensing, matrix atomic norm minimization, off grid, Vandermonde decomposition, sparse reconstruction

中图分类号: