In the realm of wireless communication engineering, the localization of
unidentified nodes in wireless sensor networks (WSN) is crucial. The localization of
unknown nodes in wireless sensor networks (WSN) is an important problem in wireless
communication engineering. Improving localization performance while extending the
lifespan of energy-limited sensor nodes remains a key objective. The coordinates of
unidentified nodes can typically be found using sensors that have the anchor nodes'
known coordinates. However, it takes a lot of time to use bio-inspired algorithms.
Therefore, it remains a difficult problem to find the best network configuration for
node localization within a short period. This article describes an effective method for
utilising an Elephant Swarm Water Search Algorithm (ESWSA) model to assess the
ideal network parameters that lead to a low Average Localisation Error (ALE). The
average localisation error (ALE) and the input variables’ transmission range (TR), node
density (ND), anchor ratio (AR), and iterations (IT) are nonlinearly related in this study
using an ANOVA-based model. In the experimental section, a metaheuristic
optimization technique was applied to predict the ALE for a given set of input
variables, producing an accuracy of about 97.843% and 98.109% for cross-validation
and testing the dataset.
Keywords: ALE, ANOVA, ESWSA, Metaheuristic optimization technique, WSNs.
