Continuous-wave doppler radar for human gait classification
RIS ID
130339
Abstract
Radar technology has commonly been used to estimate the speed and distance of a moving object. In recent years, it has been investigated as an alternative sensing modality for moving target classification. In contrast to optical imaging systems such as digital and infrared cameras, radars can work in all weather conditions and from a distance. Furthermore, radars have through-the-wall sensing capability and do not capture facial characteristics for identification; thus, they are less intrusive and are less likely to be considered to violate privacy. A modern Doppler radar detects not only the gross translation motion of the target but also local dynamics exhibited by moving parts attached to the target, for example, rotation of a helicopter blade, vibrations of an engine, or limb motion of a human. All these micro movements induce additional frequency modulations on the radar returns, thereby generating sidebands about the main Doppler frequency, which are referred to as the micro-Doppler (µ-D) signature [1]. Several studies have been conducted to analyze µ-D radar signatures of rigid and nonrigid moving targets [1-8]. Chen et al. formulated mathematical models and conducted experiments to investigate µ-D radar effects of targets under translation, rotation, and vibration motions [1,9]. Other researchers performed numerical simulations using a kinematic model or real radar data to analyze the µ-D signatures of nonrigid objects [3-8]. Many of these studies show that the µ-D signature reflects the kinetic motions of an object and provides a viable means for object identification. µ-D signals have been used to classify rigid targets, for example, to distinguish between a helicopter and an airplane [10], wheeled and tracked vehicles [11], different jet engines [12], and different ballistic targets [13].
Publication Details
F. Tivive, A. Bouzerdoum & B. G. Mobasseri, "Continuous-wave doppler radar for human gait classification," in Radar for Indoor Monitoring: Detection, Classification, and Assessment, A. Moeness, Ed. Boca Raton, United States: CRC Press, 2017, pp.95-120.