This article proposes a signal decomposition-based approach for inverse synthetic aperture radar (ISAR) imaging of multiple targets in a formation. The procedure of the proposed framework is divided into three steps. The first step is to preprocess the radar echoes received by the spatial multichannel radar via pulse compression. The second step is signal decomposition using the complex-valued mutual information-based least-dependent component analysis proposed in this article. In this step, the range profiles (RPs) of multiple targets are separated into individual RPs to generate an ISAR image for each target. The third step is range-Doppler imaging using the separated RPs. As compared to the conventional methods, the proposed method decomposes the superimposed radar echoes at the raw signal level. Therefore, even if multiple targets overlap in the ISAR image, or RP histories are unaligned owing to the change in the relative positions between multiple targets, a well-focused ISAR image can still be generated. Simulation results using three targets composed of point scatterer centers verify that the proposed method can effectively segment three targets closely flying in a formation.

中文翻译：

---

基于复杂互信息的编队飞行多目标 ISAR 成像最小相关分量分析

本文提出了一种基于信号分解的方法，用于编队中多个目标的逆合成孔径雷达 (ISAR) 成像。拟议框架的程序分为三个步骤。第一步是通过脉冲压缩对空间多通道雷达接收到的雷达回波进行预处理。第二步是使用本文提出的基于复值互信息的最小相关分量分析进行信号分解。在此步骤中，多个目标的距离剖面 (RP) 被分成单独的 RP，为每个目标生成 ISAR 图像。第三步是使用分离的 RP 进行距离多普勒成像。与传统方法相比，所提出的方法在原始信号级别分解叠加的雷达回波。所以，即使 ISAR 图像中有多个目标重叠，或 RP 历史由于多个目标之间的相对位置变化而未对齐，仍然可以生成聚焦良好的 ISAR 图像。以点散射体中心组成的三个目标仿真结果验证了该方法能够有效分割编队飞行的三个紧密目标。