Weighted inertia-based dynamic virtual bat algorithm to detect NLOS nodes for reliable data dissemination in VANETs

Abstract

Vehicular ad hoc network (VANET) is the significant network suitable for the deployment of risk-free environment that ensures least congestion and secure collaboration among the vehicular nodes of the network. The maintenance of connectivity among vehicular nodes is influenced by non line of sight (NLOS) nodes by introducing broadcasting storm and channel congestion during data dissemination. The problem of the NLOS node localization is considered as the optimization process for reducing the latency incurred in emergency information distribution. In this paper, a weighted inertia-based dynamic virtual bat algorithm (WIDVBA) is proposed for enhancing the characteristics of the traditional Virtual bat scheme that integrates the merits of simulated annealing (SA) and particle swarm optimization (PSO) for effective NLOS node localization. This proposed WIDVBA prevents the problem of premature convergence by incorporating the benefit of a weighted inertial factor compared to the traditional dynamic virtual binary bat-based NLOS localization approaches. This proposed WIDVBA dynamically increase and decrease the area of exploration and degree of exploration depending on the location of the NLOS nodes. The performance of WIDVBA studied using EstiNet 8.1 reveals that the number of NLOS nodes and time incurred in detecting NLOS nodes potentially increases with an increase in the neighbor awareness rate on par with the compared NLOS detection techniques.