Reinforcement Learning Empowered IDPS for Vehicular Networks in Edge Computing

Abstract

As VANETs have been widely applied in various fields including entertainment and safety- related applications like autonomous driving, malicious intrusions into VANETs may lead to disastrous results. Hence, intrusion detection accuracy as well as efficiency is sensitive to the normal operation of VANETs. Regarding this, in this article we propose an architecture of IDPS for VANETs. One of the highlights of the architecture is that it applies RL throughout the architecture in order to deal with the dynamics of VANETs and to make proper decisions according to current VANETs states, aiming at high detection accuracy. On the other hand, the architecture is deployed in EC in an attempt to obtain low detection latency with high processing efficiency, since VANETs IDPS is sensitive to latency, especially for safety applications. A case study is conducted to assess the validity of the proposed VANETs IDPS in EC, with the results revealing that it holds the capacity to detect and prevent intrusion in VANETs in complex environments.