摘要
全极化合成孔径雷达(SAR)可对不同极化通道分别独立进行压缩感知(CS)稀疏重建来增强成像性能,但分别独立处理没有利用极化信息的冗余性与互补性,有可能破坏极化信息的完整性。依据雷达目标在全极化下的散射特性构建联合稀疏度量函数,将全极化SAR高分辨成像转化为多通道联合稀疏约束的最优化重建问题,并用改进的正交匹配追踪算法进行求解。由于有效利用全极化信息,多通道联合CS成像相比于单通道CS成像能够获得更好的成像质量,还能全面准确反映目标全极化散射特性。通过对Backhoe挖掘机电磁仿真数据的处理,验证了算法的有效性,并且在微波暗室搭建了全极化SAR半实物仿真系统,利用其获取的全极化实测数据进一步验证了该方法的工程可行性。
The imaging performance of full polarization synthetic aperture radar （SAR） can be improved by applying the sparse reconstruction technology based on compressed sensing （CS） to different polarization-channel data independently. However, with respectively independent processing the method cannot utilize the redundancy and complementarity of the polarization information, which may destroy the integrity of the polarization information. A new joint-sparsity measure function is built according to the scattering characteristics of radar target in full polarization condition. Then, the full polarization SAR high-resolution imaging can be mathematically converted to a muhichannel joint sparse constraint optimal reconstruction problem, which can be solved via the improved orthogonal matching pursuit algorithm. Because of effectively using the full polarization information, compared with the single-channel CS imaging, the muhichannel joint CS imaging not only performs better with fewer measurements and obtains better imaging quality, but also fully and accurately reflects the fully polarization scattering characteristics of the target. Finally, the processing of the Backhoe excavator simulation data verifies the effectiveness of the proposed method; a full polarization SAR hardware-in-loop system was constructed in an anechoic chamber, the full polarization test data obtained on the system further verify the engineering feasibility of the proposed method.
出处
《仪器仪表学报》
EI
CAS
CSCD
北大核心
2017年第5期1257-1266,共10页
Chinese Journal of Scientific Instrument
基金
国家自然科学基金(61472324)项目资助
关键词
合成孔径雷达
全极化
压缩感知
联合稀疏重建
高分辨成像
synthetic aperture radar （SAR）
full polarization
compressed sensing （CS）
joint sparse reconstruction
high-resolution imaging
