基于频域稀疏非均匀采样的雷达目标一维高分辨成像

Radar Target One Dimensional High Resolution Imaging Based on Sparse and Non-Uniform Samplings in Frequency Domain

雷达目标的高分辨距离像具有稀疏和可压缩的特点,可以在频域进行稀疏非均匀采样获得目标的宽带散射数据,再通过恰当的信号处理手段得到一维高分辨像.本文描述了基于频域稀疏非均匀采样的雷达一维成像数学模型,从参数估计的角度比较了稀疏非均匀采样与均匀采样的成像性能,提出了非均匀采样点选取方法;分别应用参数估计方法和稀疏像重构方法实现了基于频域稀疏非均匀采样的雷达目标一维高分辨成像.采用暗室测量数据比较了两种方法的性能,验证了频域稀疏非均匀采样在降低数据量、提高分辨力方面的优势.

The high range resolution profiles （HRRP） of radar targets are always sparse and compressible, so it is possible to obtain the wideband scattering data by sparse and non-uniform samplings in the frequency domain and reconstnact the HRRPs by proper signal processing methods. This paper describes the mathematical model of one dimensional （1D） radar imaging, compares the imaging performance between sparse non-uniform samplings and uniform samplings from the aspect of parameter estimation, and proposes a scheme to select the non-uniform samplings with better imaging performance. Moreover, we apply both the parameter es- timation method and the sparse reconstruction method to realize 1D imaging based on sparse and non-uniform frequency samplings. The anechoic chamber data are used to compare the performances of the two methods and to verify the advantages of the sparse and non-uniform frequency samolinz scheme in reducin~ data amount and increasino range resolution.