An Insight into the Displacement Evaluation during Real-Time Radar Measurements

Nowadays, many important radar sensor features rely on the ability to reconstruct the target phase history and show the displacement over time. Generally, in state-of-the-art papers, radar's raw data processing occurs offline, i.e., after the measurement. Although this is reasonable for research purposes and facilitates processing constraints, real-life applications require real-time motion pattern measurements with particular precautions, e.g., in the case of breathing activity monitoring. This paper provides some insights on how the target motion during real-time measurement can be properly elaborated. This task has been widely discussed in the literature, but many articles do not deal with the possible incorrect displacement estimation when offline algorithms are applied to real-time measurements, involving inappropriate unwrapping. This generates wrong displacement detection and phase drift that are often solved with post-processing algorithms and erroneously attributed to accumulated noise and hardware imperfections. Moreover, a few articles describe the peculiar requirements of real-time detection compared to offline measurement. This contribution aims to investigate the origin of possible errors and give a simple but effective solution to correctly estimate the target displacement. The reported algorithm can be applied to both FMCW and Doppler radars.