Fourier Array Processing for Buried Victims Detection Using Ultra Wide Band Radar with Uncalibrated Sensors

The purpose of this paper is to propose a new way to detect victims buried in or under layers of rubble or debris in case of disasters such as earthquakes, fires or terrorist attacks. The method is based on Fourier processing and principal component analysis (PCA). It utilizes a moving array of sensors arranged in the line of motion. It combines the speed and robustness of Fourier-based processing with a capability of space-time adaptive processing (STAP) to detect objects moving at extremely low velocities. The concept is demonstrated on several experiments carried out using an ultra wide band (UWB) ground penetrating radar (GPR). A fast, yet robust and extremely sensitive technique is being tested to achieve detection of the slightest motion (such as breathing) of a subject using uncalibrated sensors.