基于反射信号时域重构的探地雷达超薄层状介质参数反演方法

探地雷达实际应用场景中介质表面往往存在着其他覆盖物,将会严重影响实际探测效果,如冬季路面表面的冰层、沥青铺路时防止沥青混合物粘在滚筒表面而喷出的水汽层等,且这些覆盖层的厚度较薄,通常小于探地雷达可分辨的最小厚度。因此,本文提出了一种基于反射信号时域重构的层状介质参数反演方法,利用广义反射系数建立雷达发射回波的时域模型,再通过遗传算法对模型的代价函数进行优化进而反演出覆盖物薄层的参数。最后,通过时域差分工具gprMax仿真软件模拟沥青混凝土实施铺路场景,对所提算法的有效性和准确性进行验证分析。 

An inversion method of ultra-thin layered media of GPR based on time domain reconstruction of reflected signal

In application scenarios, there often exists thin coverings on the surface of the ground penetrating radar medium, which will affect detection effect, we propose a layered medium parameter inversion method based on the reflection signal reconstruction. The novelty of our method is that it solves the problem that the thickness of the cover layer is usually less than the minimum resolvable thickness of the ground penetrating radar. In this process, we first use the generalized reflection coefficient to establish a time-domain model of radar emission echo; Then optimize the cost function of the model through genetic algorithm to inverse the parameters of the thin layer of the cover; Finally, we use gprMax simulation software simulates the paving scene of asphalt concrete, which verifies and analyzes the effectiveness and accuracy of the proposed algorithm.