Design and Implementation of Identity Authentication Architecture System Fusing Hardware and Software Features

Distributed resource access in the new power system will also occur in large-scale growth, due to the variety, large-scale, and in a complex open physical environment, in the relatively weak security protection measures, it is very easy to become the entry point for illegal personnel to implement network attacks. This paper proposes a fusion of software and hardware identity authentication architecture system, software fingerprint part based on network probing extracted message features, statistical features and time difference sequence features extracted based on bypass listening using direct splicing fusion to generate software fingerprints of electric power intelligent terminals, and then so that a variety of feature information can be taken into account to improve the recognition of the fingerprint features, hardware fingerprints part to take the extraction of the signal leading code part of the signal for statistical feature analysis, and finally based on the integration of the signal leading code part of the signal for statistical feature analysis. The hardware fingerprint part takes the part of extracting the leading code of the signal for statistical feature analysis, and finally based on the integrated learning fusion of software and hardware fingerprints as the device fingerprint features, which can largely solve the security authentication problem of electric power equipment based on HPLC and Modbus communication, and achieve more than 95% recognition ability under most of the machine learning classification methods.