Research On The Secure Communication Scheme For ICV In-vehicle Network Based On The Random Characteristics Of Vehicle Power Supply

Recently,with the rapid development of automobile intellectualization and networking,the traditional automobile gradually exits the historical stage.Intelligent Connected Vehicle(ICV)has become a new development trend of automobile industry.It gives people a better experience in entertainment and comfort,but it also has a certain security threat.The attacker uses the remote interface of the vehicle to illegally invade the Electronic Control Unit(ECU)in the vehicle,and obtains the data transmitted on the bus through the security vulnerability in the ECU node communication mechanism,so as to achieve the effect of illegally controlling the vehicle.Therefore,it is an important work to study a safe and efficient protection mechanism of ICV network communication to ensure data security.At present,the existing ICV in-vehicle network security communication method is to protect the bus data with the security protection means in the information system,but the attacker will still bypass these protection methods to control the vehicle illegally.In addition,the bandwidth required by identity authentication,data encryption and other methods in these protection methods exceeds the resources provided by ICV,so it is difficult to apply them in the vehicle network with limited resources.In view of the above difficulties,from the point of view that ICV is a physical information system,this paper proposes an in-vehicle network communication scheme based on the random number extracted from the vehicle power supply voltage as the communication key.The main work of this paper is as follows:Firstly,in order to get an ideal random key,the voltage change process of vehicle power supply is analyzed from the vehicle itself.Aiming at the influence of external traffic environment on driving decision,a causal model of driving mode is built.On this basis,the generator model is further constructed to describe the change process of power supply voltage.Through Markov and maximum likelihood estimation,the voltage change is divided into three model events,and the state transition probability matrix of the voltage is obtained.It is concluded that the power supply voltage change process is a stochastic process with Markov characteristics.Secondly,according to the famous Kerckhoff hypothesis,the security of the cryptosystem does not depend on the security of the algorithm,but on the key.Therefore,with the help of the random characteristics of the continuous variation of vehicle power supply voltage,a random number extraction scheme is designed.In order to enhance the randomness of the sequence,multi information entropy fusion is used as the initial entropy source.Wavelet transform is used for decomposition and reconstruction to filter periodic signals.The security of sequence extraction is evaluated by conditional minimum entropy.Then,parallel extraction is used to improve the efficiency,and digital post-processing is used to improve the randomness of sequence.Finally,the random sequence is extracted by simulation experiment,and the randomness of the sequence is analyzed,and the linear complexity is compared with the existing sequence Gold and Kasami.The results show that the extracted sequence has good random performance.Thirdly,in order to ensure the security of communication between ECUs in the vehicle network,the random sequence extracted above is used as the communication key,and a one-time pad communication architecture is designed.The architecture mainly protects the network communication in the vehicle from three levels: identity authentication,session key distribution and one-time pad communication of ECU based on SM4.After the security theory analysis,and compared with the existing in vehicle network security communication scheme,the architecture designed in this scheme not only meets the security requirements of in vehicle network,but also has low bus load rate.