Specific Emitter Identification Method Based On Amplifier Behavior Modeling

Specific Emitter Identification(SEI)refers to the technology of measuring the characteristics of the received electromagnetic signal and determining the individual radiation source that generates the signal based on the existing prior information.It is of great military significance to correctly identify detected radiation sources in a complex and changing electromagnetic environment.At the same time,with the advent of the era of 5G and the Internet of Things,the security of wireless networks has also ushered in huge challenges.Specific Emitter Identification technology can be applied to wireless access authentication,electromagnetic environment supervision and other projects,which can effectively improve the security of wireless networks.In this paper,the power amplifier was used as an example to conduct in-depth research on individual identification methods of emitters.The specific research content is as follows:Firstly,the inevitable nonlinearity and memory effects of the power amplifier are analyzed,and the basic theory of modeling the nonlinear behavior of the power amplifier is studied.Behavioral modeling mainly involves memoryless Saleh model,memoryless polynomial model and memory polynomial model.For the modeling of amplifier behavior,the measured data of BLT53 A amplifier was collected as the data basis for modeling.Use three models to model the behavior of amplifier data,and compare and analyze to get an optimal model.The model parameters are changed to simulate the approximate amplifier in practical applications as the research object of identification,and the model is used to generate a simulation data set for individual identification.Secondly,this paper extracts the nonlinear characteristics of the power amplifier signal,and then uses a classifier to identify and analyze it.The validity of the method is verified using the generated simulation data set,and the method can effectively distinguish eight individuals.On this basis,the data of eight actual amplifiers of the same model and the same batch were collected and identified using a method based on nonlinear feature extraction.The influencing factors of the identification were also analyzed,including the modulation method and classification of the signal and FFT points.The experimental results show that when the signal-to-noise ratio is 20 d B,the recognition rate of the eight amplifiers can reach 85%.Finally,the method of individual identification of radiation source based on the combination of pictures and deep learning is studied.Use the Contour Stella Image as a bridge between the signal and the picture.After obtaining the picture dataset,use deep learning methods to classify.This paper uses Alex Net,Squeeze Net,Res Net-18 and VGG16 networks for classification and recognition.The experimental results show that this method has greatly improved the classification and recognition of equipment compared with the method based on nonlinear feature extraction.When the signal-to-noise ratio is 5d B,the recognition rate is greater than 90%.