Bistatic aspect diversity for improved SAR target recognition

This paper analyzes the potential for improvement in the performance of automatic target recognition (ATR) for synthetic-aperture radar (SAR) with bistatic aspect diversity. Initial assessments using decision-level fusion of monostatic observations with bistatic observations provide promising results. Data was generated using three civilian vehicle facet files and an electromagnetic scattering simulator. Classification was performed using normalized cross-correlation template matching and majority voting. Results showed an increase in the probability of correct classification with decision-level fusion of bistatic observations over classification using single observations.