Radar observations of human activity have a variety of applications in security, defense, and rescue operations. In radar signatures, Doppler modulations caused by human motions are of particular interest, because they can be used to identify particular human activities. Such modulations have been referred to as ``micro-Doppler" signatures in the relevant literature, and can be examined using a time frequency representation. Range-Doppler signatures are a useful tool for remotely retrieving information regarding human activity. However, interpretation of the signatures requires an understanding of the scattering processes.
This dissertation investigates the development of an approximate model for human scattering in order to explain the impact of multi-frequency, polarization, and a multi-path scattering on human motion radar signatures. The model utilizes a finite dielectric cylinder and multi-path approximation. The human model is described as a collection of dielectric cylinders. The use of cylinders is motivated by the quasi-cylindrical nature of the human limbs, torso, and head. Because human motions occur above the ground, multi-path approximation is used to consider ground effects. To validate the model, the simulation results are compared using a commercial electromagnetic simulation tool that is based on the method of moments. Additionally, a joint radar and motion capture measurement is conducted to validate the model experimentally.
This dissertation also examines the advantages of using polarimetric radar by investigating the polarimetric effect on human motion. Although polarization of the incidence and scattered waves is known to influence the electromagnetic scattering process, only a small number of previous studies of human Doppler signatures have considered polarization effects. In this study, the multi-polarized backscattering cross section of human walking is simulated, and analyzed using a statistical approach. In addition, this dissertation examines the effect of multi-path scattering on human radar signatures for multiple polarizations and investigates the advantages of using UWB radar systems in observations of range resolved human Doppler signatures.