基于多维特征的全息雷达“低慢小”目标识别

    LSS Target Recognition in Holographic Radar Based on Multi-dimensional Features

    • 摘要: 由于我国机场“黑飞”和“扰航”现象频发,严重威胁公共安全,以至于对可靠的无人机和飞鸟等“低慢小”目标的探测与识别技术的需求显著增加。文中深入分析多旋翼无人机、固定翼无人机、飞鸟、直升机等“低慢小”目标的旋翼或发动机叶片旋转以及飞鸟翅膀拍动等的微观运动特性、电磁散射特性、宏观运动特性等,形成综合利用目标微多普勒特性、雷达散射截面积特性以及运动特性等多维特征的全息雷达“低慢小”目标识别技术。采用全息雷达系统在外场采集的多种“低慢小”目标实测数据对所提方法的有效性和可行性进行实验验证与分析,提取的特征对各目标均能达到较高的识别率,实验结果证实了基于多维特征的“低慢小”目标分类识别技术的有效性。

       

      Abstract: The frequent occurrences of “black flight” and “interference” at airports have seriously threatened public safety, so that the demand for reliable detection and recognition technology of “low altitude, slow speed, small” (LSS) targets such as unmanned aerial vehicles (UAV) and birds has significantly increased. This paper deeply analyzes the microscopic motion characteristics of rotor or engine blade rotation and bird wing flapping, electromagnetic scattering characteristics, and macroscopic motion characteristics of LSS targets such as multi-rotor UAV, fixed-wing UAV, bird, and helicopter. A LSS target recognition technology for holographic radar is formed by comprehensively using multi-type feature such as micro-Doppler (m-D) feature, radar cross section (RCS) feature and motion feature of the target. In this paper, the validity and feasibility of the proposed method are verified and analyzed by the measured data of LSS targets collected by the holographic radar system in the field. The experimental results show that the m-D features, RCS and motion features are effective for the classification of LSS targets.

       

    /

    返回文章
    返回