An abnormal intrusion detection method for Internet of Vehicles based on the difference of traffic flow density

Abstract

The invention discloses a vehicle networking abnormal intrusion detection method based on traffic flow density difference, belonging to the field of vehicle networking and network intrusion detection.The invention is provided with an event analysis module in a vehicle-mounted unit and a roadside base station. Firstly, a distributed intrusion detection mechanism or a centralized intrusion detection mechanism is selected according to the difference of the actual traffic flow density in the road network. Then, the network information and traffic information in the vehicle node are obtained by the vehicle unit, and the anomaly intrusion detection in the vehicle network environment is realized by using the event analysis module of the vehicle computer or the roadbed computer and the weighted improved naive Bayesian algorithm to classify the information. The two detection mechanisms of the invention cooperate with each other so that the vehicle node can be detected under any moving speed, thus ensuring the integrity and high efficiency of the intrusion detection, and solving the problem that the traditional intrusion detection system is not adapted to the dynamic change of the vehicle networking communication and the network node moves quickly.

Description

technical field [0001] The invention relates to a vehicle networking technology and a network intrusion detection technology, in particular to a vehicle network abnormal intrusion detection method based on traffic flow density differences. Background technique [0002] With the intelligent development of the transportation field, the Internet of Vehicles technology has been considered as one of the core technologies of future intelligent transportation. At present, the development of the Internet of Vehicles has provided great convenience to our lives, including the real-time transmission of traffic information and the reduction of traffic congestion. At the same time, information security issues in the Internet of Vehicles have gradually emerged. For example: the hacking incident of Jeep, some vehicles equipped with the Connected Drive digital service system have security vulnerabilities, etc., and once the security problem of the Internet of Vehicles breaks out, it will se...

AI Technical Summary

Problems solved by technology

For example: the hacking of Jeep, some vehicles equipped with the Connected Drive digital service system have security vulnerabilities, etc., and once the security problems of the Internet of Vehicles break out, it will seriously threaten the safety of people's lives and property

[0006] There are many existing algorithms for intrusion detection, including: neural network, support vector machine, genetic algorithm, statistical algorithm, etc. For the neural network classification algorithm, the classification system can be continuously updated through machine learning during the implementation process, but the Its application in the Internet of Vehicles system is too costly and time-consuming, and it cannot guarantee real-time detection in the Internet of Vehicles environment with frequent topologies; for support vector machines, it has many advantages in solving small samples, nonlinear and high-dimensional pattern recognition. However, in the intrusion detection of the Internet of Vehicles, the amount of data is huge, and it mainly involves two types of recognition, so the application of support vector machines will also have some shortcomings; the naive Bayesian algorithm based on statistics is widely used. It has strong classification ability in two types of classification problems, but the classification accuracy of Naive Bayesian algorithm is low. In actual application, it should be improved and adjusted according to actual needs.

Images