DIB - A Novel Optimized VANET Traffic Management Using a Deep Neural Network

Abstract

The advancement of the Internet of Things (IoT) establishes the development of the Internet of Vehicles (IoV) and Intelligent Transportation Systems (ITS). An integral part of ITS is the vehicular ad hoc network (VANET) with smart vehicles (SV). In this research, a dynamic method of traffic regulation in VANET is presented using Deep Neural Networks (DNN) and Bat Algorithms (BA). With a reduced average delay, the former (DNN) is utilized to guide vehicles across very crowded routes to increase efficiency. In order to examine the traffic congestion status between network nodes, BA is integrated with the IoT and moved over VANETs. Experiments were conducted to test the effectiveness of the proposed method with various parameters such as average latency, packet delivery ratio (PDR) and throughput and the performance were compared with different machine learning (ML) algorithms. The simulation outputs proved that the proposed technique supports real-time traffic circumstances with less energy usage and delay than existing methods.