Detection of Human Target Location Under Simulated Randomized Rubble Using Global Fresnel’s Reflection Coefficient

There has been a lot of cases when humans get buried under a pile of rubble created by the collapsing of man-made structures and sometimes due to natural causes. In such cases, it’s imperative that they are extracted as soon as possible. Ultra-wideband impulse radar is being used for such purposes, for detecting micro as well as macro motions created by the target. This paper presents a simulated realistic rubble-target model that will incorporate real-life electrical and spatial properties of the rubble and human thorax. The global reflection coefficient is derived for the entire multi-layered rubble and target environment using Fresnel’s coefficients and validated using Green’s function with a mean absolute error of 0.6013 % and correlation of 97.37 %. Based on the thorax movement, radar is able to detect the location of the target using the variance approach. Considering the model simulation parameters, rubble properties, and attenuation suffered by the signal when travelling through the rubble, the maximum detection range of the target was set to 3 m. When the target was 3 m away, the maximum mean absolute error in the target range was obtained at 28 cm which is within the dimensions of an average human thorax.