论文部分内容阅读
将压缩感知(CS)理论用于逆合成孔径雷达(ISAR)成像,可以有效利用缺损的雷达回波数据,解决了因数据缺损造成成像质量下降的问题。目前压缩感知中常用的高斯或伯努利等随机测量矩阵独立随机元数目过多,存储空间过大,从而导致硬件实现成本过高。所构造的稀疏带状测量矩阵,通过将测量矩阵进行带状循环移位置零稀疏化,可大幅减少测量矩阵中非零元素数目,降低系统采样要求,节约硬件实现成本,使得压缩感知ISAR成像工程化更容易实现。最后通过仿真和微波暗室实验数据验证了点目标模型下稀疏带状测量矩阵进行ISAR成像的可行性和有效性。
Compress sensing (CS) theory is applied to inverse synthetic aperture radar (ISAR) imaging, which can effectively use the missing radar echo data and solve the problem of the image quality degradation caused by the data defect. At present, Gaussian or Bernoulli, which are commonly used in compressed sensing, have too many independent random elements and excessive storage space, resulting in high hardware implementation costs. By constructing the sparse strip matrix, the number of non-zero elements in the measurement matrix can be greatly reduced by reducing the striping matrix to zero and zeroing the measurement matrix, reducing the sampling requirements of the system and saving the hardware implementation cost, so that the compression-aware ISAR imaging project Easier to achieve. Finally, the feasibility and effectiveness of ISAR imaging with sparse banded measurement matrixes under the point target model are verified by simulation and experimental data in darkroom.