Improving performance and data transmission security in VANETs

This article proposes a new approach to achieve fast and reliable transfer of data and uses machine learning techniques for data processing to improve the performance and data transmission security of the vehicular network. The proposed approach is the combination of 5G cellular network and alternative data transmission channels. The data collection experiment took place within different areas of the city of Berlin over a 3-month time period and involved the use of 5G technologies. The study carried out the analysis and classification of big data with the help of position-based routing protocols and the Support Vector Machine algorithms. The said techniques were employed to detect non-line-of-sight (NLoS) conditions in real time, which ensure the secure transmission of data without the loss or degradation of network performance. The novelty of the work is that it tackles various traffic scenarios (the extent of road congestion can affect the quality of big data transmission) and offers a way to improve big data transfer using the Support Vector Machine technology. The study results show that the proposed approach is effective enough with big data and can be employed to improve the performance of urban VANET networks and data transmission security. The study results can be useful in developing high-performance 5G-VANET applications to improve traffic safety in urban vehicular environments. © 2021 Elsevier B.V.