Trustworthy Routing in VANET: A Q-learning Approach to Protect Against Black Hole and Gray Hole Attacks

Vehicular Ad-Hoc Networks (VANETs) are very promising in the context of intelligent transportation systems. VANETs are vulnerable to various types of attacks, including black hole and gray hole attacks, which can disrupt or intercept communication and compromise the security and reliability of the network. To address this issue, we propose a method for detecting and preventing malicious vehicles activity in VANETs by utilizing a trustworthy routing technique based on Q-learning (QL-TRT). Our approach, which is formulated based on the Markov Decision Process, allows the vehicle to choose the best neighbor for routing and avoid malicious neighbors. To evaluate the trustworthiness and reliability of the links between pairs of vehicles, we consider factors such as packet forwarding ratios, energy consumption, and expected transmission time. Q-learning is used to learn the trust value of links and select the most trusted route from source to destination. The evaluation results demonstrate the effectiveness of QL-TRT in detecting black hole and gray hole attacks, while ensuring the communication performance in VANETs.