Intrusion Detection System for In-Vehicular Communication

摘要

Intelligent wirelessly communicating vehicles brought the concept of Vehicular Ad-hoc Networks. Advancement in vehicles and services for communication are provided by Intelligent Transportation Services. Each vehicle in connected network is individually responsible for task performance. In order to ensure the safety, comfort and Quality of Service of the network, it is important for In Vehicular Communication services to function properly. Controller Area Network is a standard protocol that is responsible for the communication between different Electronic Control Units (ECUs) inside the vehicle. Advancements in wireless technologies leads to various security and cyber threats. In order to effect the communication, the intruder can inject different messages to misguide the driver or ultimately leads to the system failure. In this paper, an Intrusion Detection System is proposed which is able to detect and classify the normal and attack messages. Two type of attacks i.e. DoS and Fuzzy types are considered in this work. We use two algorithms, Multilayer perceptron and Decision Tree for classification of normal and attack messages.