Font Size: a A A

Research On Enhanced Radio Frequency Fingerprint-based Inherent Identity Authentication Methods Of Wireless Devices

Posted on:2024-01-28Degree:DoctorType:Dissertation
Country:ChinaCandidate:X F LiFull Text:PDF
GTID:1528307340473954Subject:Computer system architecture
Abstract/Summary:PDF Full Text Request
Using the embedded Radio Frequency(RF)features in wireless signals to construct the RF fingerprint of signal-sources and serving for non-intrusive and continuous identity authentication is a key technology in the field of wireless network security,and such an approach holds great potential for applications in the Internet of Things,industrial network,and integrated aerospace networks,and etc.Existing researches on RF fingerprint-based inherent identity authentication have focused on the construction process of RF fingerprint from RF features.However,issues such as the limited variety of RF features and underutilization of RF features in terms of time and space properties have resulted in weak discrimination ability of the constructed RF fingerprint for different wireless devices,which will affect the accuracy of device identity authentication.In response to these challenges,this dissertation proposes some RF fingerprint enhancement mechanisms to improve the accuracy of RF fingerprint-based device identity authentication by expanding the variety of RF features and improving the time and space properties of RF features based on RF fingerprints.The research contents and innovations of this dissertation are summarized as follows:(1)By focusing on the limited variety of existing RF features,this dissertation proposes a new RF feature called ”fractal dimension” for direct sequence spread spectrum communication systems.The calculation method of fractal dimension through the samples of wireless signals is proposed,and a mathematical model is established to analyze the relationship between random hardware imperfections and the calculated fractal dimension,which provides a theoretical demonstration that the proposed fractal dimension can characterize the hardware clock jitter characteristics of wireless devices.The proposed fractal dimension feature is also combined with five existing RF features,including carrier frequency offset,to construct the RF fingerprint and serve for identity authentication on a closed device set.An experiment using twenty wireless network interface cards demonstrates that,by involving the proposed fractal dimension feature into RF fingerprint construction,the identity authentication accuracy can be improved from 94% to 98.5%.(2)By focusing on the underutilization of RF features in terms of time property,this dissertation proposes a RF watermark-based RF fingerprint construction scheme,which can combine multiple frames signal to improve the time property of RF features,for carrier sensing multiple access/collision avoid communication systems.This dissertation proposes the generation and embedding methods of RF watermark to form the unique and unforgeable mapping relationship among frames signal from the same signal source.Based on such relationship,the receiver can combine multiple frames signal from the same signal source to improve the time property of RF features,enhancing the constructed RF fingerprint,and improve the accuracy of identity authentication.Simulation results show that,by involving the proposed combination scheme of multiple frames signal in the process of RF fingerprint construction,the authentication accuracy can be improved from 90% to 97%.(3)By focusing on the underutilization of RF features in terms of space property,this dissertation proposes a cooperative transmission-based RF fingerprint construction scheme,which can combine the RF features estimate at two nodes to improve the time property of RF features,for multiple access communication systems.This dissertation proposes the twonodes cooperative transmission scheme,in which a neighbor node monitors the signal from the target device,estimates the RF features,and forwards to receiver,to form the replicas of RF features estimate of the signal signal source at different locations.The receiver combines these replicas to improve the space property of RF features,enhance the constructed RF fingerprint,and improve the identity authentication accuracy.Simulation results demonstrate that,by involving the proposed cooperative transmission scheme in the process of RF fingerprint construction,the accuracy can be improved by an average level of 9.2%.
Keywords/Search Tags:Identity Authentication, Radio Frequency Fingerprint, Fractal Dimension, Radio Frequency Watermark, Cooperative Transmission
PDF Full Text Request
Related items