Specific Emitter Identification Based on Eye Diagram

Specific emitter identification (SEI) refers to identifying an individual emitter by extracting the features of individual differences reflected in the external structure of RF signals of the emitter. The RF fingerprint of transmitters are complex and non-linear, but the performance improvement of SEI is limited by the current cognitive level. Therefore, this paper proposes an SEI method based on the signal eye diagram, which uses CNN to realise the joint extraction of multiple complex features. Firstly, we construct the transmitter distortion model and analyse the visual representation of the distortion on the eye diagram. On this basis, an enhancement strategy is introduced to the original eye diagram. The enhanced eye diagram is used to train the network model and complete the SEI task. The proposed method overcomes the limitations of existing cognition and has high information integrity and low computational complexity. The simulation results show that the proposed method can significantly improve the performance of SEI, and the average recognition gain is about 15%.