稀疏带状测量矩阵在压缩感知ISAR成像中的应用-红外与激光工程.PDFVIP

第42 卷第11 期 红外与激光工程 2013 年11 月 Vol.42 No.11 Infrared and Laser Engineering Nov.2013 稀疏带状测量矩阵在压缩感知 ISAR 成像中的应用 谭 歆，冯晓毅，王保平 (西北工业大学 电子信息学院，陕西 西安 710129) 摘 要 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 存储空间过大，从而 ， 立随机元数目过多， 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。 关键词 压缩感知； ISAR 成像； 稀疏带状测量矩阵； 微波暗室实验； 点目标模型 中图分类号 TN959.1 文献标志码 A 文章编号 1007-2276(2013) 11-3137-07 Application in compressed sensing ISAR imaging based on sparse banded measurement matrices Tan Xin, Feng Xiaoyi, Wang Baoping (School of Electronics Information, Northwestern Polytechnical University , Xi 忆an 710129, China) Abstract: The application of compressed sensing (CS) theory to inversed synthetic aperture radar (ISAR) can make effective use of the data of defective radar echo, and solve the lower quality of imaging, which is caused by the defective data . By analyzing the present situation, it was found that the number of independent random elements in the commonly used Gauss and Bernoulli random measurement matrices was too huge, and the storage space was too large, which led to the high hardware implementation cost. Sparse banded measurement matrices were constructed in this paper, which significantly reduced the number of nonzero elements in measur