基于频率分集逆合成孔径雷达的多重信号分类目标成像算法

Multiple Signal Classification Target Imaging Algorithm Based on Frequency Diversity Inverse Synthetic Aperture Radar

将频率分集的思想应用在逆合成孔径雷达(inverse synthetic aperture radar,ISAR)成像中,通过单频信号合成宽带信号,可解决系统发射接收宽带信号复杂的问题。但窄带的合成可视为宽带信号的稀疏采样,由此带来了旁瓣提高等难点。提出一种基于频率分集ISAR体制的多重信号分类(multiple signal classification,MUSIC)目标成像算法,该算法将合成阵列接收的回波信号协方差矩阵进行特征值分解,得到信号子空间与噪声子空间,然后根据二者的正交性构建谱函数对目标位置进行估计,得到目标的超分辨二维像。将MUSIC算法与后向投影(back projection,BP)算法做了对比,仿真结果表明:在有较强噪声环境下,前者仍能有较好的成像效果,证明本文方法的应用可有效解决频率稀疏带来的高旁瓣问题。

The idea of frequency diversity is applied to inverse synthetic aperture radar(ISAR)imaging.The complex problem of transmitting and receiving wideband signals can be solved by synthesizing wideband signals through single frequency signals.However,the synthesis of narrow band can be regarded as the sparse sampling of wideband signal,which brings difficulties such as high sidelobes.A multiple signal classification(MUSIC)target imaging algorithm based on frequency diversity ISAR system was proposed.The signal subspace and noise subspace were obtained by eigenvalue decomposition of the covariance matrix of the echo signal received by the synthetic array,and then the target position was estimated by constructing spectral function according to the orthogonality of the two subspaces to obtain the super-resolution two-dimensional image of the target.The MUSIC algorithm was compared with the back projection(BP)algorithm.The simulation results show that the former algorithm can still achieve better imaging effect in the strong noise environment,which indicates that the application of the proposed method can effectively address the problem of high sidelobes caused by frequency sparsity.