Methodology to achieve accurate non cooperative target identification using high resolution radar and a synthetic database

In the last few years, there is a great interest in developing an identification system capable to make a reliable classification of aircrafts into different groups (friendly, hostile or neutral). Depending on the context in which these systems are deployed, incorrect identification may lead to serious problems, such as fratricide or engagement of civilian aircrafts. Different techniques have been researched to face this problem, but non-cooperative ones have awakened more interest because they do not require aircraft collaboration. Non Cooperative Target Identification (NCTI) using radar is a complex task, mainly due to the fact that a database of possible targets is needed. To populate this database, Radar Cross Section (RCS) predictions produced by computer simulation seem to be the most feasible way to perform this task, since measurements alone cannot cover the vast range of targets, configurations and required aspect angles. These predictions are typically performed in the frequency domain and a specific processing must be done to obtain both High Resolution Range Profiles (HRRPs) and 2D Inverse Synthetic Aperture Radar (2D-ISAR) images. This paper shows a methodology to face the NCTI task, which use both synthetic HRRPs and 2D-ISAR to achieve an accurate identification.