OHRDBN: An optimized hybrid model of RBF and DBN for obstacle‐aware routing with optimal path selection in VANET sector

Abstract

SummaryVehicular ad hoc networks (VANETs) are increasingly essential for communication architecture utilized in daily human life. The presence of obstacles in the terrain area, as well as the vehicular nodes present in the network, is considered an essential constraint since it reduces the efficacy of the mobile ad hoc network (MANET). To attain the VANET requirements, a novel routing scheme is designed to identify the obstacle. The main aspect of the proposed work is to detect the obstacle and determine the path availability. Initially, the characteristics of vehicle nodes are gathered, which include the packet loss, received signal strength indicator (RSSI), connectivity, energy level, and so on. In the process of data transmission, the proposed model is intended to detect the obstacles and determine the available path by using an optimized hybrid model (OHL) combining radial basis function (RBF) and deep belief network (DBN). The parameter optimization in the hybrid classifier model is done with the help of the adaptive chimp optimization algorithm (ACOA). Once it checks the path availability, then the path optimization is accomplished by using the same ACOA. Finally, the objective function for the proposed work is constructed by considering various constraints like distance, energy, link availability, life span, link duration, and mobility. The performance of the implemented system is analyzed, and its results are validated through different measures. The outcome of the suggested model demonstrates that the implemented system tends to exploit effective routing with obstacle detection when compared to various other existing approaches.