基于两层均匀媒质的GPEN SAR地下目标成像方法及其性能分析

地表穿透合成孔径雷达(Ground PENetration SAR,GPEN SAR)为了探测掩埋在地下的目标,通常工作在多层媒质的环境中。传统成像模型是建立在同一均匀媒质的假设上,不再适合于GPEN SAR的实际情况。本文首先建立了两层分区均匀媒质中的成像模型,然后利用后向投影(BP)算法定量分析了成像几何参数、土壤参数等对成像的影响,进而提出了一种修正的后向投影(MBP)算法。MBP算法不仅能够校正两层分区均匀媒质对成像定位的影响,还能估计目标的掩埋深度,提供目标三维位置坐标。仿真结果验证了MBP算法在不同信噪比环境下,对多目标的三维定位精度能满足实际的需要。

GPEN SAR underground targets imaging formation in two-layer homogeneous media environment and its performance analysis

Ground Penetration SAR (GPEN SAR) usually works in multiple media condition in order to detect buried targets. The traditional imaging model is based on homogeneous media, which is not suitable for GPEN SAR. A new imaging model considering two-layer homogenous media is studied in this paper. Rased on the model, the effects of imaging geometry and soil characteristic on the SAR image are analyzed quantitatively using the Hack-Projection (BP) algorithm. Furthermore, a Modified RP (MBP) is proposed, which can not only correct the effects of two - layer homogeneous media on SAR image but also estimate the depth of targets under the ground. The simulate results verify the efficiency of MBP algorithm under different signal-to-noise ratio (SNR) environment.