Vehicular Ad-hoc Network (VANET) connects vehicles to improve road safety and service efficiency. However, VANET communication is vulnerable to various attacks. In particular, malicious vehicles can freely send false messages to other vehicles in the network, leading to serious consequences, such as traffic congestion and accidents. Therefore, to ensure road safety, it is critical to detect such false messages promptly and accurately. In this paper, we propose a novel machine learning (ML)-based method for misbehavior detection. In our approach, we first convert the received message data into images, then feed the images into a well-trained CNN model. The images that contain false messages can be accurately identified. As one single image is converted from several messages, our approach allows to detect false messages with less time than the state-of-the-art solutions, where detection is conducted directly over messages. Thus, the time efficiency of detection is significantly improved.
Poster: Machine Learning Based False Position Detection Using Data-to-Image Transformation
01.05.2024
1265632 byte
Aufsatz (Konferenz)
Elektronische Ressource
Englisch
British Library Online Contents | 2003
|Poster: Urban Traffic Monitoring via Machine Learning
IEEE | 2019
|ELIMINATION OF FALSE POSITIVES IN VIV IDENTIFICATION USING MACHINE LEARNING
British Library Conference Proceedings | 2019
|Seed Viability Detection Using Computerized False-Color Radiographic Image Enhancement
British Library Conference Proceedings | 1993
|