Background and Clutter Removal Techniques for Ultra Short Range Radar

Living-object-detection (LOD) in hazardous areas can save lifes and is the key feature of safety critical radar sensors. Suppression of bright and static clutter is needed in order to detect weak echoes from potentially moving living objects. We study the classical method of frame differencing (FD) as well as statistical background and clutter removal techniques to maximise the object detection probability. The investigated techniques are evaluated by simulation and qualified in terms of achievable signal-to-clutter ratio (SCR) and computational complexity. Subsequently, the techniques are implemented in a radar prototype and real life measurements are conducted. The prototype consists of a two channel C-band frequency modulated continuous wave (FMCW) quasi-monostatic radar system with closely placed omnidirectional monopole antennas. For the background and clutter removal a new modified method is proposed and qualified. The proposed method provides top SCR levels and outperforms FD by far in terms of efficiency.