STAP moving target position estimation accuracy improvement and false detection recognition using a priori road information

Abstract

We have recently presented a processor for traffic monitoring applications that combines the post-Doppler space-time adaptive processing (PD STAP) with a road map obtained from the freely available OpenStreetMap (OSM). In this paper, the positioning error model of this processor is presented and discussed. In fact, two error models are combined: one for the PD STAP detections and one for the OSM road points. The positioning error model is essential for obtaining robust and reliable results. It was tested using real 4-channel aperture switching radar data acquired by the DLR's airborne system F-SAR. The results reveal a powerful algorithm that recognizes and rejects many of the false detections, being suitable for traffic monitoring applications.