A survey on Reducing Traffic Congestion by Disseminating Messages in Vehicular Ad Hoc Networks

Reducing the number of road accidents in the current scenario is a very challenging societal problem. If the information regarding the accidents is given to the vehicles approaching the area, the secondary accidents can be considerably reduced. Vehicular Ad Hoc Networks (VANETs), the network of vehicles that can communicate with each other play a vital role in the reduction of such accidents. Many survey and research papers published on this topic majorly emphasize on implementation of VANETs using simulators but this research work is focused on implementation using hardware components. Henceforth, this research work proposes a real-time system with vehicular nodes that detects an accident and disseminates the message. The system uses a deep learning object detection model, MobileNet SSD to detect the accident and then starts capturing the video of the accident location. Video capturing is done to increase the credibility and reliability of the information in the network. The video streams are transmitted among vehicles connected via the same network. Routing protocols are applied in the network to find the best route which helps in quick dissemination of the capture.