稀疏线性调频步进信号ISAR成像观测矩阵自适应优化方法

摘要

The ISAR imaging technology with sparse Stepped-Frequency Chirp Signals (SFCS) based on Compressive Sensing (CS) theory can construct the target image from a few of measurements with high probability,where the measurement matrix optimization is an effective way of improving the imaging quality and reducing the measurements.However,most of the existing measurement matrix optimization methods do not utilize the target characteristic,which leads to low adaptive ability of target.Therefore,an adaptive measurement matrix optimization method for Inverse Synthetic Aperture Radar (ISAR) Imaging with sparse SFCS is proposed in this paper,where the actual physical observation process is considered and the target characteristics are utilized to optimize the measurement matrix.In the method,a parametric sparse representation model of ISAR imaging is established to solve the Doppler sensitivity firstly.On the basis,the measurement matrix is optimized with the goal of obtaining the best target image with the minimum measurements under a given image quality requirement.As a result,the expected imaging results can be obtained with minimum measurements by using the optimized measurement matrix.The effectiveness of the proposed method is demonstrated by experiments.%基于压缩感知(CS)理论的稀疏线性调频步进信号(SFCS)逆合成孔径雷达(ISAR)成像技术能够从少量观测数据中高概率重构出目标像,其中,观测矩阵的优化设计是提高成像质量和减少观测数据量的有效途径.然而,现有的观测矩阵优化设计研究通常没有考虑目标特征信息的有效利用,对目标的自适应能力不足.因此,该文在充分利用目标特征信息的基础上,结合稀疏SFCS信号的实际物理观测过程,提出一种ISAR成像观测矩阵自适应优化方法.该方法首先建立参数化稀疏表征成像模型以解决稀疏SFCS信号多普勒敏感问题,在此基础上,以在达到成像质量要求条件下使用最少观测数据量获得最优成像结果为目标对观测矩阵进行自适应优化设计,最终能够利用最少的数据量获得满意的目标成像结果.仿真实验验证了该算法的有效性.