Research On Device Physical Layer Authentication Methods For MmWave MIMO Communication Systems

A wireless communication system that integrates millimeter wave(mm Wave)and multiinput multi-output(MIMO)technology,with its rich spectrum resources and antenna diversity,meets the requirements of 5G/6G wireless communication networks for ultra-high data transmission rates and high communication reliability,supporting diverse business,network and device heterogeneity,and dynamic application scenarios,such as smart cities,industrial internet,etc.However,due to the practical demand for fast and random access of diverse heterogeneous wireless devices to networks,as well as the natural openness of wireless transmission media,millimeter wave MIMO communication systems are highly susceptible to device identity spoofing attacks,resulting in severe security threats to millimeter wave MIMO communication systems.The existing identity authentication methods for wireless devices mainly rely on upper level cryptographic techniques.However,with the widespread use of millimeter wave MIMO technology in wireless communication systems,wireless communication devices are rapidly evolving towards heterogeneity,dynamism,and lightweighting,posing significant challenges to the key distribution and management,encryption and decryption processes in cryptographic authentication methods.As an effective supplement and enhancement to upper level cryptographic authentication methods,physical layer authentication is expected to become an effective authentication scheme for millimeter wave MIMO communication systems.Physical layer authentication uses physical layer features to provide identity identification for device authentication.Due to the lack of complex upper level information processing,its computational and communication costs are minimal.However,there are still some key issues that need to be addressed in the physical layer authentication methods for millimeter wave MIMO communication systems,such as the susceptibility of physical layer features to environmental noise interference,difficulty in accurately characterizing physical layer features in complex dynamic scenes,and limited authentication performance based on coarse-grained physical layer features.In response to the above issues,this paper studies a highly reliable,adaptive,and secure device physical layer authentication method for millimeter wave MIMO communication systems.The main research results are as follows:1.To tackle the problem of carrier frequency offset in single antenna communication systems being easily affected by environmental noise interference,which leads to poor authentication reliability,a highly reliable device authentication method that can effectively combat environmental noise interference is designed using MIMO antenna diversity technology.Specifically,using autocorrelation methods to accurately model the multi-dimensional carrier frequency offset of mobile MIMO communication devices,providing anti-interference and reliable identity identification for device authentication.Research on instantaneous carrier frequency offset tracking and prediction methods based on extended Kalman filtering technology,combined with the received signal to extract statistical features of carrier frequency offset.On this basis,using quantization theory and binary hypothesis testing methods,a highly reliable device identity authentication model based on MIMO hardware multidimensional carrier frequency offset features is constructed to achieve device identity legitimacy verification.Apply theories such as probability theory and random signal analysis to provide analytical expressions for false alarm rate and detection rate,and evaluate the performance of the improved reliable device authentication method.The simulation results show that MIMO diversity technology can effectively enhance the carrier frequency offset’s ability to resist environmental noise interference,improve the reliability of authentication methods,and provide basic theoretical and security technical support for further exploring device identity authentication in mobile millimeter wave MIMO communication systems.2.To address the issue of difficulty in accurately characterizing physical layer features in complex dynamic scenes,an adaptive device authentication method is designed using online machine learning algorithms to accurately capture sparse,weak,and time-dependent physical layer features of millimeter wave MIMO.Specifically,utilizing the full threedimensional characteristics of multiple domains,accurately modeling high-frequency and ultra wideband time-varying sparse channels and carrier frequency offset,providing multidimensional identity labeling for device authentication.Design a narrow beam channel estimation method based on compressive sensing algorithm and a dynamic carrier frequency offset tracking method based on Kalman filtering technology to extract physical layer features of millimeter wave MIMO.At the same time,using Gaussian kernel function method,a low computational complexity identity security authentication model without prior knowledge of features is constructed.Online machine learning algorithms are used to track actual time related features in real-time,achieving real-time updating of the authentication model and flexible and efficient verification of device identity legitimacy.Combining mathematical statistics and functional analysis theories,provide analytical expressions for false alarm rate and false detection rate,and evaluate the performance of adaptive authentication methods.The experimental results show that the designed online machine learning algorithm can effectively characterize time-dependent features,and the proposed authentication method has strong environmental adaptability.3.To address the issue of limited authentication performance based on coarse-grained physical layer features,design a strong security device authentication method that integrates finegrained multi-dimensional physical layer features of millimeter wave MIMO.Specifically,the low rank structure and sparse scattering characteristics of millimeter wave channels are utilized to accurately characterize the physical channels driven by geometric scenes,and the Wiener model is used to efficiently model the phase noise of millimeter wave MIMO,accurately capturing multi-dimensional and refined features of device identity.Design a fine-grained geometric channel and phase noise estimation method for millimeter wave MIMO based on maximum likelihood estimation technology and optimization methods,using the received signal to extract fine-grained multi-dimensional features.On this basis,a device identity strong security authentication model based on fine-grained and multi-dimensional features is constructed to efficiently resist multi angle identity forgery attacks by illegal devices.Combining matrix analysis theory and conditional probability methods,provide closed form expressions for false alarm rate and detection rate,and evaluate the representation effect of fine-grained multi-dimensional physical layer features on device identity.A large number of simulation results indicate that fine-grained multidimensional features can greatly increase the difficulty of illegal devices forging identities,effectively enhancing the security of device authentication methods.