Experimental study of through-wall human detection using ultra wideband radar sensors

Measurement of weak signals is very challenging since the signals may be buried by noise. In this paper, we study measurement of sense-through-wall signals using UWB radar sensors, and measurements show that the collected signals due to respiratory movement of human target are very weak. Based on the measurements, we perform through-wall human detection. The detection of human targets hidden by walls, or trapped in buildings is of interest for rescue, surveillance and security operations. In this paper, experiments on through-wall human detection using the ultra wideband (UWB) radar PulsOn 220 in monostatic mode are carried out in two scenarios: through gypsum wall, and through wooden door. And three analytic methods are employed for the detection: normalized difference square matrix method, reference moving average method with Discrete Fourier Transform (DFT), and Empirical Mode Decomposition (EMD) from Hilbert Huang Transform, of which the breathing information of the human target is contained in the third intrinsic mode function (IMF3). The experimental results for the human target detection behind the wall are demonstrated and thus compared using these three methods. It shows that the distance of the human target to the UWB radar could be approximately estimated by the normalized difference square matrix method and the reference moving average method. As to Hilbert Huang Transform, even though the existence of the through-wall human target can be clearly detected, the range information is ambiguous.