小波分析在毫米波综合孔径近场成像中的应用  

The Application of Wavelet Analysis for the Near-Field Millimeter Wave Synthetic Apeture Imaging

作　　者：张育浩 王海辉[1] 李超[1] 

机构地区：[1]北京航空航天大学数学与系统科学学院,北京

出　　处：《应用数学进展》2017年第1期62-68,共7页Advances in Applied Mathematics

摘　　要：毫米波综合孔径的成像技术在近场成像中有着广阔的应用前景,无需机械扫描,可稀疏阵列,具有较高的空间分辨率。毫米波综合孔径近场可视度模型可转化为V=AR+n,这是一个欠定方程组,解不唯一,且含有噪声。综合小波阈值去噪算法及均值滤波在去噪方面的优点,提出了一种改进的小波阈值去噪-滤波算法,并且改进了传统小波阈值去噪中的软、硬阈值法。通过将基于偏微分方程的正则化方法及改进的小波阈值去噪-滤波算法应用在一个实际的近场仿真实验中,比较发现改进的小波阈值去噪-滤波算法要优于基于偏微分方程的正则化方法。Millimeter wave imaging technology using synthetic aperture has broad application prospects in the near field. It has high spatial resolution without mechanical scanning. Near-field millimeter wave synthetic aperture visibility model can be transformed into V=AR+n , this is an underde-termined equation, the solution is not unique, and it contains noise. An improved wavelet thre-sholding-filtering algorithm combining wavelet thresholding algorithm’s advantages and mean filtering’s advantages in terms of de-noising is presented. The method improves traditional soft and hard threshold method. At the last, we make the regularization method based on partial differential equations and the improved wavelet thresholding-filtering algorithm used in an actual near-field simulation experiment. By comparison, we find that the imaging results of the improved algorithm are better than the regularization method based on PDE.

关 键 词：毫米波成像 近场 小波阈值去噪–滤波算法 

分 类 号：TP39[自动化与计算机技术—计算机应用技术]