Intelligent Transportation Systems for Deep Learning-Driven Vehicular Ad hoc Network

Abstract—Numerous studies demonstrate that the vehicular ad hoc network (VANET) depends on various characteristics and intermediate connections. It offers real-time automatic reaction and acute traffic analysis, but more studies are still needed to determine how best to use it in various situations. The primary goals of this VANET system are to distinguish between specific agents and identify collision remnants, which is still a research area in terms of scalability, optimization strategies, and efficient data aggregation. Due to problems with distance disintegration, temporal channel deterioration, and signal distortion, analysis was not feasible until recently. Therefore, this research will carry out a comparative review of available studies related to Intelligent Transportation Systems that use deep learning applications in VANETs, such as the recurrent neural network model, cybersecurity, decision-making, and collision avoidance, as well as future work, so it can have a more concise understanding of the topic.