Contextual Authentication Technology Based On Fingerprint Of Internet Of Things

In recent years,with the continuous development of technologies such as smart home,intelligent transportation,Internet of Things(IoT)applications have become more and more popular.Meanwhile,the security of the IoT has become increasingly prominent and must be highly valued.If the authentication problem of the IoT device can be improved from the source,illegal control can be effectively avoided,and the security of the IoT device can be effectively improved.This thesis proposes a context authentication method based on the fingerprint of IoT devices.Under the premise of the technology certification,only legitimate users and legitimate IoT terminal devices can access the network,and ensure that the terminal device can access the network when the identity of the terminal device is consistent with the user's pre-set identity.Therefore,problems such as interest disputes and network security caused by unauthenticated devices and illegal user access are avoided.This thesis firstly surveys the IoT and its security issues through the current research status of the IoT,and summarizes common identity authentication and device fingerprint authentication technologies.Secondly,a method for extracting device fingerprints is proposed.By analyzing the fingerprint features unique to the interactive traffic extraction device,as an important factor for context authentication,the device identity information can be effectively judged.Then the identity authentication method based on device fingerprint is proposed.This method constructs the recognition context by using six kinds of situation factors in two practical scenarios,and uses the decision tree classification algorithm to classify the context factors,and considers the plaintext transmission of the data to propose the plaintext.Encryption is used to reduce the risk of important data breaches.Finally,the design and implementation of the proposed situation authentication system is carried out,and the functions and relationships of each module are clarified.At the same time,the IoT device extracts the fingerprint features of related devices and combines the decision tree classification algorithm to verify the feasibility of the context authentication technology.The classification accuracy rate of the experiment is 95%,and the other 5% false positive rate is special but also meets the certification requirements.The results show that the context authentication technology based on the fingerprint of IoT devices can ensure only trusted IoT devices and trusted users access the network,so as to achieve two-way authentication of IoT devices.