Multisensor systems and countermeasures

A methodology for the simulation of a multisensor system exposed to countermeasures has been developed in the project. The aim is to investigate how a multi-sensor system is affected by countermeasures. In this project the target recognition function of a combined radar-/IR-seeker that is exposed to countermeasures has been studied. Features extracted from simulated high range resolution radar data and IR images are fused before classification (feature fusion) or used for fusion of individual sensor decisions (decision fusion). Sensor data are collected during a flight of the seeker approaching ground vehicle targets moving on a grass field. The countermeasure is smoke that attenuates both IR and radar signals. The correct classification rate is always better for the fused sensor data than for any single sensor as long as the confidence level of the data is considered. Disurbances that are not indicated might cause severe degradation of the system performance. In this study, decision fusion turned out more robust than feature fusion, which in turn, is more robust than classification using individual sensor data.