Target track classification for airport surveillance radar (ASR)

The aim of this paper is to identify false tracks that arise from weather and biological targets, and to increase the air traffic security and safety level by detecting aircrafts lacking secondary surveillance radar (SSR) data. In this paper a single-source target classification after tracking using support vector machines (SVM) is introduced, which gives each track an updated probability value based on its likelihood behavior to conform to aircraft and non-aircraft targets. We introduce various features and evaluate different combinations in order to achieve the highest clustering index. The experimental classification results with real radar data provide good evidence that machine target classification is viable, with the capability of being implemented in real time.