Multiple dictionaries-based radar target identification via a likelihood ratio test

Target identification has been an active researching area in past decades. In this paper, we present an iteration procedure to optimize the size of the undercomplete dictionary when multiple undercomplete dictionaries are used to characterize the scattering signatures of a complex target. Furthermore, we extend the signature reconstruction and decision criterion with only single undercomplete dictionary to the case with multiple dictionaries for a more practical target identification application. The proposed approach is compared with the matching-score criterion-based approach and single dictionary-based approach using measured signatures of three aircraft models in an ultra wide-band chamber. Results show that the proposed approach can provide more promising identification accuracy due to a more effective representation to the complex scattering behaviors.