Technologies Research For Specific Emitter Identification

Specific Emitter Identification(SEI)is a technology that identifies different emitters using the unique characteristics of distorted signals.The individual characteristics of the emitter utilized by SEI,also known as Radio Frequency Fingerprint(RFF),are caused by the production,processing and debugging of the hardware device.It is the inherent characteristic of wireless devices and is difficult to be tampered with.SEI has played an increasingly important role in both military and civil applications.Example applications include battlefield spectrum management,cognitive radio and wireless network security,etc.With the development of modern communication system,SEI techniques will have broad application prospects in various fields and breed more new application modes.There have been many SEI algorithms developed in recent years.To evaluate the performance of SEI algorithms,researchers always compare their study with the state-ofart techniques.However,due to the factors such as SEI scenarios,data acquisition devices,public datasets and different data pre-processing effects,the lateral comparison is not entirely "fair" and "scientific".The evaluation criteria of SEI are lacking.To address those problems,this thesis investigates the identification accuracy of the SEI system.Firstly,this thesis discussed the distortion model with In-phase/Quadrature(I/Q)imbalance and Power Amplifier(PA)nonlinear distortion.Based on the distortion model,the Euclidean distance distributions of the transmitted signals from different emitters with I/Q imbalance and PA nonlinear distortion are derived.Then,a theoretical performance bound for dual-target emitter identification was proposed based on the Naymen-Person(NP)rule,which provided evaluation criteria for related SEI evaluation problems.Finally,according to the Euclidean distance distribution,this thesis studied the minimum Euclidean distance distribution between multiple emitters,and proposed an upper bound for multi-target emitter identification to match the practical communication scenario better.This further improved the theoretical basis for the performance evaluation of the SEI system based on distortion features.The analytical method and derivation framework can also be extended to different modulation schemes and distortion models.