A New Method of Specific Emitter Identification

Specific Emitter Identification (SEI) techniques have been widely used as a tool of spectrum management operation to enhance the security by matching the received signal with the specific emitter. However, there are few methods focusing on the unintentional modulation (UIM) of signals. In this letter, we propose a novel method which extracts the UIM part in the frequency spectrum and instantaneous phase from the original signals as UIM radio frequency fingerprinting, and then use the non-Gaussian measuring tools to extract the dimension-reduced secondary features. The real-world signals are used to evaluate the performance of the proposed method and the Support Vector Machine (SVM) classifier and the Gaussian Mixture Model (GMM) classifier are applied to make the classification. Experiment results verified that the proposed method has remarkable performance and less computational complexity, which is relatively practical.