一种快速的基于压缩感知的多普勒高分辨方法

利用雷达目标在多普勒域的稀疏性,基于压缩感知的目标多普勒估计方法,能够在有限的相干积累时间内实现多普勒的高分辨.然而,即使采用压缩感知中的一种高效算法——正交匹配追踪算法,其运算复杂度也相对较高.为了进一步降低运算复杂度,对接收脉冲进行分组,将一维的多普勒估计问题转化为一个二维的稀疏信号重构问题,进而利用一种针对二维稀疏信号优化的低复杂度正交匹配追踪算法对其进行估计.仿真表明,该方法具有较高的运算效率,并能够获得接近直接应用传统的正交匹配追踪算法的多普勒分辨率.

Low complexity compressed sensing based Doppler high resolution algorithm

Exploiting the sparsity of radar targets in the Doppler frequency domain, compressed sensing based Doppler estimation methods can lead to high resolution estimates of targets' Doppler frequencies in the very limited coherent integration time. However, this involves a large amount of computation even via an efficient algorithm-Orthogonal Matching Pursuit (OMP). For further reduction of computational complexity, the 1D Doppler estimation is translated into a 2D sparse signal recovery problem through a pulse grouping method. Then a low complexity OMP algorithm optimized for 2D sparse signals is utilized. Simulation results indicate that high resolution Doppler estimates approximating those of the OMP can be obtained with an improved efficiency.