Through-the-wall high-dimensional imaging of human vital signs by combining multiple enhancement algorithms using portable LFMCW-MIMO radar

Abstract

Radar-based life imaging technology has been widely applied in both civilian and military applications. Wall penetration attenuation degrades the signal-to-clutter-and-noise ratio (SCNR) and radar portability restricts the spatial resolution of the high-dimensional image. In this study, we constructed a portable linear-frequency continuous-wave(LFMCW) multiple-inputmultiple-output(MIMO) radar system, and we presented a novel high-dimensional imaging framework. We combined the maximum spectrum peak, fourth-order cumulant (FOC) and 3D coherent factor (CF) to enhance vital signs based on attention mechanism. We improve resolution by deconvolution algorithm using the data from a human dummy. Visualization-enhanced 4D point clouds images were created by mapping micro-doppler-frequency to color. We employed a series of experiments to validate the proposed method, including human and beagle imaging in free space, and short-range and long-range (16 m) human imaging through a wall. These high-dimensional imaging results validated the portable radar system’s potential in detection, identification, interpretation of human targets.