Multistatic UWB Radar-Based Passive Human Tracking Using COTS Devices

Due to its high delay resolution, the ultra-wideband (UWB) technique has been widely adopted for fine-grained indoor localization. Instead of active positioning, UWB radar-based passive human tracking is explored using commercial off-the-shelf (COTS) devices. To extract time-of-flight (ToF) reflected by moving person, accumulated channel impulse responses (CIR) and corresponding variances are used train convolutional neural networks (CNN) model. Particle filter algorithm track person based on extracted ToFs all pairs links. Experimental results show that proposed CIR- variance-based CNN models achieve less than 30-cm root-mean-square errors (RMSEs). Especially, model robust scenario changing promising practical applications.