Estimation of Doppler Velocities from Sub-Nyquist Ultra-Wideband Radar Measurements

Abstract

Many conventional Doppler analysis techniques for radar sensors suffer from velocity ambiguity when confronted with targets that are moving at speeds higher than the Nyquist velocity. This paper proposes a signal processing method for estimating accurately the Doppler velocity of targets from such sub-Nyquist radar data. The hypothesis tested in the paper is the possibility of circumventing the velocity ambiguity by exploiting the high range resolution of ultra-wideband radar sensors, more specifically incorporating the texture method applied to the signal envelope intensity profile. We employ a millimeter-wave radar sensor operating at 60 GHz to measure a rotating cylinder and a walking human to investigate the applicability of the proposed approach. The experimental results indicate that the proposed method can estimate the Doppler velocity accurately without aliasing from sub-Nyquist data, demonstrating that the use of the texture method is promising for resolving the ambiguity caused by sub-Nyquist sampling. The advantage of the proposed method is that the ambiguity can be resolved simply with the signal processing technique, and the approach does not require any custom hardware, such as nonuniform samplers. An important future study based on this paper is the application of the proposed approach to develop a low-cost radar system with a low sampling rate that can measure fast-moving targets.