Lamarckian learning‐based whale optimization algorithm to localize non‐line‐of‐sight nodes for achieving reliable data dissemination in vehicular ad‐hoc networks

Abstract

SummaryNon‐line‐of‐sight (NLOS) nodes in vehicular ad‐hoc networks (VANETs) are responsible for introducing channel congestion and broadcast storm during data dissemination for sustaining reliable connectivity between vehicular nodes. In this paper, an Integrated Lamarckian Learning and Whale Optimization Algorithm‐based NLOS Localization Technique (ILL‐WOA‐NLOS‐LT) is proposed for NLOS localization in order to improve high reliability and low latency during the emergency message sharing. This proposed ILL‐WOA‐NLOS scheme inherits an optimization process that concentrates on the objective of minimizing latency during warning message delivery. It included the merits of Lamarckian evolution‐based learning strategy for strengthening and speeding up the rate of local search. It possesses maximum probability of acquiring higher adaptability through active learning for improving the global convergence speed to attain better localization of NLOS nodes in emergency situations. It also incorporated a better tradeoff between exploitation and exploration for effective NLOS localization with reduced error. The simulation experiments of the proposed ILL‐WOA‐NLOS scheme conducted through EstiNet 8.1 confirmed its predominance in achieving an excellent mean emergency message dissemination rate of 12.96% mean NLOS node localization rate of 14.21% and the mean neighborhood awareness rate of 13.62% with different number of vehicular nodes, compared to the benchmarked localization approaches considered for investigation. The localization error of the proposed ILL‐WOA‐NLOS scheme was also identified to be significantly reduced by 12.82% compared to the baseline schemes used for investigation.