系统工程与电子技术

• 传感器与信号处理 • 上一篇    下一篇

基于压缩感知的高频地波雷达二维DOA估计

赵春雷1, 王亚梁1, 毛兴鹏1,2, 于长军2,3   

  1. 1. 哈尔滨工业大学电子与信息工程学院, 黑龙江 哈尔滨 150001; 2. 哈尔滨工业大学信息感知技术协同创新中心, 黑龙江 哈尔滨 150001; 3. 哈尔滨工业大学(威海)
    信息与电气工程学院, 山东 威海 264209
  • 出版日期:2017-03-23 发布日期:2010-01-03

Compressive sensing based two-dimensional DOA estimation for high frequency surface wave radar

ZHAO Chunlei1, WANG Yaliang1, MAO Xingpeng1,2, YU Changjun1,2   

  1. 1. School of Electronics and Information Engineering, Harbin Institute of Technology, Harbin 150001, China; 2. Collaborative Innovation Center of Information Sensing and Understanding, Harbin Institute of Technology, Harbin 150001, China; 3. School of Information and Electronical Engineering, Harbin Institute of Technology (Weihai), Weihai 264209, China
  • Online:2017-03-23 Published:2010-01-03

摘要:

针对高频地波雷达(high frequency surface wave radar, HFSWR)目标角度估计问题,提出了一种基于压缩感知(compressive sensing, CS)的距离多普勒(range Doppler, RD)域单快拍二维波达方向(direction of arrival,DOA)估计方法。分析了在RD域进行DOA估计的理论可行性及单快拍处理的优势;基于子空间追踪思想提出了一种针对L型阵列的二维DOA估计降维后的角度配对和修正方法,在低信噪比和目标回波幅度接近的情况下仍能保证完成正确的匹配并消除无效的角度组合,从而实现精确的二维DOA估计和目标空域分辨。仿真结果和高频地波雷达实测数据的处理结果分别验证了其性能优势和有效性。

Abstract:

To achieve accurate angle estimation of targets in high frequency surface wave radar (HFSWR), a two-dimensional direction-of-arrival (2-D DOA) estimation method based on compressive sensing with a single snapshot in rangeDoppler (RD) domain is proposed. The theoretical feasibility of estimating DOA in RD domain and the benefits of processing with a single snapshot are analyzed. Based on subspace pursuit, an approach of angle pair matching and correction after dimensionality reduction is proposed for L shaped array. It eliminates invalid angle pairs and ensures correct matching even under conditions of low signal-to-noise ratio (SNR) or nearly equal echo amplitudes. Therefore, accurate 2-D DOA estimation accompanied with resolving targets in spatial domain are realized. Simulation results and experiment results of practical HFSWR data verify the performance advantages and effectiveness of the proposed method respectively.