Abstract:

According to the characteristics of moving radar signals, an ambiguity function (AF) representative-cut based feature extraction and optimization algorithm is presented for unintentional modulation recognition of moving radar emitters. It considers near-zero Doppler cuts of AF as a representative feature set of radar emitters and extracts the stable features of moving radar emitters effectively. Further, the direct discriminant ratio criterion is employed to preserve the most discriminating features by ranking the kernel points along the representativecut. Experimental results on real data of moving radar emitters, as well as synthetic-radar-data from USA Naval Research Laboratory, demonstrate that the proposed approach not only mitigates the computational problem in existing AF based methods, but also boosts recognition accuracy greatly.