RADLAB
menu MENU
Home > People > Faculty > Kamal Sarabandi > Research Areas & Projects > Modeling and Analysis of Millimeter-Wave Doppler Signature of Walking Human

Modeling and Analysis of Millimeter-Wave Doppler Signature of Walking Human

Millimeter-wave radars are envisioned for a number of safety and security applications such as collision warning for vehicles, autonomous vehicle control, and detection of concealed weapons and explosives carried by individuals. Detection and identification of pedestrian in highly cluttered radar scattering scenes such as urban and highway environments is also of great importance. The scattered field from a human subject undergoes Doppler modulation of a particular signature if the human subject begins to move. Measuring the radar backscatter of a moving human subject over an extended period of time, a complete Doppler spectrum can be extracted. It is shown that this spectrum is unique enough to distinguish a human subject from other objects in the radar scene. A signal processing method is also presented to decompose that radar backscatter of a human subject into its components (torso, hands, and legs).

Figure 1: Human body is illuminated by a uniform plane wave in the same direction of walking.
Figure 1: Human body is illuminated by a uniform plane wave in the same direction of walking.
Figure 2: Distribution of the magnitude of electric current over a human body surface at W-band frequencies illuminated with a uniform wave as shown in Figure 1. Also show is the estimated Doppler spectrum of the radar backscatter.
Figure 2: Distribution of the magnitude of electric current over a human body surface at W-band frequencies illuminated with a uniform wave as shown in Figure 1. Also show is the estimated Doppler spectrum of the radar backscatter.