空间域中压缩感知在车载雷达目标定位中的应用（英文）

Target Location via Spatial Compressed Sensing in Automotive Radar

针对毫米波汽车雷达在对目标定位时用传统Nyquist采样定理导致采样率过高、利用MUSIC算法计算目标到达角(DOA)快拍数要求高的问题,以及传统方法中使用匹配滤波器造成目标速度和距离计算分辨率不高的问题,提出空间域中基于压缩感知的方法,实现对目标距离-速度-角度的联合定位.用Matlab仿真空间域中基于压缩感知方法定位目标的过程,仿真结果表明在空间域中利用压缩感知技术可以提高雷达对目标距离和速度的分辨率并且在计算目标DOA时相比MUSIC算法可以降低对快拍数的要求同时达到更好的分辨率,并且利用压缩感知方法可以省略传统雷达上用到的匹配滤波器.

One of the task of an automotive radar is to get the direction information which includes the range,speed,and azimuth of the target.There are some challenges in automotive radars such as the high sampling rate caused by the traditional Nyquist theorem,the so many snaps in the multiple signal classification(MUSIC) algorithm for targets' azimuth estimation,and the low resolution of speed and range.In order to solve these problems,a novel method based on compressed sensing(CS)in spatial domain is proposed in this paper.This paper presents the signal processing by MATLAB. The simulation results show that using the CS technique in automotive multiple-input multiple-output(MIMO)radar can reduce the sampling rate effectively,improve the resolution of range and speed without the use of a matching filter.Moreover,the CS technique needs less snaps than the MUSIC algorithm and achieves a better resolution,which has a promising prospect of automotive radar signal processing.