A novel predictive localization algorithm for underwater wireless sensor networks

Abstract

The location of nodes is critical in underwater wireless sensor networks (UWSNs), which is an ocean monitoring platform. UWSNs are motivated by the popular usage of localization and play a major role in several technologies that depend primarily on innovations and localization of these nodes. Underwater node localization is a critical technology that enables the deployment of a variety of underwater applications. In this study, the underwater nodes are divided into two levels. Firstly, a clock asynchronous localization system (LS-AC) for base layer’s node localization is presented. In order to eradicate the original ranging strategy's dependence on active nodes and address the problem of energy consumption, LS-AC performs in-network situation-based monitoring by relying on asynchronous clocks. Secondly, we propose a backtracking search algorithm (OTKL-BSA) based on optimal topology and knowledge learning. It is used to address the issues associated with traditional algorithms' lack of diversity and the imbalance between exploration and exploitation. Thirdly, to solve the problems that the traditional gray wolf optimizer (GWO) is prone to falling into local optimal values and has a low search efficiency, this paper proposes a GWO scheme based on hunting step size (GWO-HSS). Finally, simulation results show that the proposed algorithm outperforms SLMP, MCL-MP, MP-PSO, and MGP in aspects of localization performance.