Deep Learning-Based Intrusion Detection Systems for Intelligent Vehicular Ad Hoc Networks

Abstract

In recent years, malware classifies as a big threat for both internet and computing devices that directly related with the in-vehicle networking security purpose. The main perspective of this paper is to study use of intrusion detection system in in-vehicle network security using deep learning (DL). In this topic, possible attacks and required structure and the examples of the implementation of the DL with intrusion detection systems (IDSs) is analyzed in details. The limitation of each DL-based IDS is highlighted for further improvement in the future to approach assured security within in-vehicle network system. Machine learning models should be modified to gain sustainable in-vehicle network security. This modification helps in the quick identification of the network intrusions with a comparatively less rate of false-positives. The paper provides proper data; limitation of previously done researches and importance of maintaining in-vehicle network security.