Vehicle-mounted CAN network intrusion detection system and method based on incremental learning

Abstract

The invention relates to the technical field of vehicle-mounted network security, in particular to a vehicle-mounted CAN network intrusion detection system and method based on incremental learning, and the system comprises an automobile terminal module, a roadside edge module, a cloud server and an incremental learning detection module. A mode of combining incremental learning and a deep neural network (DNN) is adopted, marked real vehicle data is used for training a basic model of the DNN, online detection and incremental learning are performed at an edge end, and unmarked data is adopted for updating a prediction model, so that the generalization ability of an intrusion detection model is improved, the calculation and storage burdens of a vehicle-mounted system are unloaded, and the prediction efficiency is improved. The method fully utilizes the calculation time efficiency of the edge end and the storage capability of the cloud end, can maintain the continuous updating of the model, timely and accurately detects the attack abnormity of the vehicle-mounted network, and provides a guarantee for the safe driving of the vehicle.

Description

technical field [0001] The invention relates to the technical field of in-vehicle network security of connected vehicles, in particular to an in-vehicle CAN network intrusion detection system and method based on incremental learning. Background technique [0002] With the application of information technology and sensor technology, the Internet of Vehicles and autonomous driving technologies are booming, and the new generation of vehicles is developing towards intelligence and networking. More and more functions such as adaptive cruise and automatic parking are integrated in the car. System, the car is no longer a simple, single-function means of transportation, it will realize the all-round connection and communication between the car and the car, the car and the person, the car and the transportation infrastructure and network, and provide users with safe, comfortable, intelligent and efficient. Driving experience and transportation services. However, while bringing conve...

AI Technical Summary

Problems solved by technology

However, in practical applications, as the driving environment of the car changes or new attack behaviors occur, the generalization ability of the existing model will decrease, which will affect the detection accuracy.

Images