Research On Key Technologies Of Malicious Node Detection In Internet Of Vehicles

In the intelligent transportation system,the Internet of Vehicles is an important part of it,especially for road safety and intelligent transportation management,which brings huge convenience to the society.Since malicious vehicle nodes can attack the normal nodes of the Internet of Vehicles by forging false information,collusion attacks,etc.,the network performance is seriously affected.Therefore,detecting malicious nodes in the Internet of Vehicles is very important to ensure the safety of the Internet of Vehicles system.This thesis conducts an in-depth study on the key technologies of malicious node detection in the Internet of Vehicles.The main contributions are as follows:1.Aiming at the problem of false information attacks by malicious nodes in the Internet of Vehicles,a malicious node detection mechanism based on repeated games and trust evaluation is proposed.Based on the repeated game model,the profit function of malicious nodes and normal nodes is obtained.In order to make it more suitable for the actual Internet of Vehicles,a discount factor is introduced to optimize the profit function model.Through multiple games,the maximum profit of each node is calculated.When the revenue of each node is maximized,the revenue of the node is transformed into the trust value of the node.The node trust value is compared with the threshold value to filter out malicious nodes.Simulation experiments show that the method proposed in this chapter improves the detection rate of malicious vehicle nodes,reduces the network packet loss rate,and effectively isolates malicious vehicle nodes.2.Aiming at the defect of ignoring the historical trust value of nodes in the current malicious node detection methods in the Internet of Vehicles,a method for detecting malicious nodes in the Internet of Vehicles based on the improved Bayes algorithm and adaptive adjustment is proposed.When calculating the direct trust value of two interactive nodes,increase the weight of the influence of the negative event.When recommending the trust value,the "recommended trust distance" is introduced as the recommended trust measurement method,and malicious recommendations are first excluded.Since the behavior of malicious nodes cannot be determined,an adaptive threshold adjustment method is introduced to prevent malicious nodes from evading detection.Simulation experiments show that this method can quickly detect malicious nodes and avoid node collusion attacks.3.Aiming at the task of detecting malicious nodes in the Internet of vehicles,this thesis designs and implements a detection system for malicious nodes in the Internet of vehicles.According to the two detection methods proposed in this thesis,a convenient and efficient detection system for malicious nodes is designed and implemented.It can provide users with vehicle networking node situation overview,malicious node analysis,malicious node prediction and other functions,so as to facilitate users to analyze and judge vehicle nodes.