RELAX参数估计算法在雷达目标特征提取中的应用

从理论上推导分析了聚束式SAR(Spotlight SAR)的chirp信号回波模型。在回波模型中,回波被分解为两部分--强散射点的回波信号以及噪声和杂波,强散射点的回波(正弦信号)为离散谱,而噪声和杂波为连续谱。基于该回波模型,将RELAX算法应用于对可分解为若干强散射点的特定目标的特征提取,从噪声和杂波的连续谱中估计出强散射点回波的离散谱参数,频率对应于强散射点的位置、幅值对应于强散射点的RCS,从而实现对目标的特征提取。最后,通过计算机仿真,验证RELAX算法的有效性和优越性。

Application of RELAX Algorithm in Radar Target Feature Extraction

In this paper, data models of returned chirp signal for Spotlight SAR are inducted and described. In this model, returned signal consists of two parts--the returned sinusoids signal of dominant scatters and the additive noise and clutter. The sinusoids have a discrete spectrum, while the noise and clutter has a continuous spectrum. Based on such model, the RELAX algorithm is introduced in the feature extraction of targets which can be divided into several dominant scatters: estimating these dominant scatters' discrete spectrum parameters from the noise and clutter's continuous spectrum. Among the estimated parameters, the frequencies correspond to the location of dominant scatters, and the amplitudes are proportional to their RCSs. By this means, we can realize target feature extraction for this specific kind of targets. Finally, computer simulation results are presented to show the validities and advantages of RELAX algorithm.