Research On Malicious Attack Detection And Protection Key Technology In The Internet Of Vehicles

Scholars have proposed many methods for the detection and prevention of malicious attacks,but these methods do not necessarily apply to the Internet of vehicles.There are two main reasons: First of all,we need to make it clear that the Internet of vehicles malicious attacks mainly originate from three major active attacks,namely Denial of Service,Tamper,and Spoofing.The traditional detection methods have not been optimized for these three attacks in Internet of vehicles,so the direct use on the Internet of Vehicles will reduce the detection accuracy.Secondly,attackers generally conduct malicious attacks on the Internet of Vehicles through malicious viruses.the current mainstream response to malicious attacks is passive defense,and its disadvantage is that the security vendor update always lags behind the virus update,this also reflects the lag of passive defense in dealing with malicious attacks.The main work of this thesis is divided into two parts: the first part is to improve the traditional K-nearest neighbor algorithm,a new detection algorithm CW-KNN is obtained,and finally use the constructed malicious attack simulation data set of Internet of Vehicles for simulation experiments.The second part is to improve the traditional infectious disease model,and introduce two factors that reflect individual awareness differences to obtain a new prevention model IOV-SIRS,and then carry out balance solution and stability analysis on the new model and finally carry out simulation experiments.The experiment in this thesis proves that the new detection algorithm is superior to the traditional intrusion detection algorithm in the detection of Internet of vehicles malicious attacks.The new model can evaluate and calculate the spread trend of malicious attack virus,and provide corresponding strategies before the outbreak of virus,so as to minimize the harm caused by the new attack.At the same time,the experiment also proves that using the complex network thought to predict the virus transmission trend can effectively control the virus transmission on the Internet of vehicles.