Wireless Sensor Network (WSN) architectures are widely used in a variety of practical applications. In most cases of application, the event information transmitted by a sensor node via the network has no significance without the knowledge of its accurate geographical localization. In this paper, a method based on Machine Learning Technique (MLT) is proposed to improve node accuracy localization in WSN. We propose a Single Hidden Layer Extreme Learning Machine (SHL-ELM) and a Two Hidden Layer Extreme Learning Machine (THL-ELM) based methods for nodes localization in WSN. The suggested methods are applied in different Multi-hop WSN deployment cases. We focused on range-free localization algorithm in isotropic case and irregular environments. Simulation results demonstrate that the proposed THL-ELM algorithm greatly minimizes the average localization errors when compared to the Single Hidden Layer Extreme Learning Machine and the Distance Vector Hop (DV- Hop) algorithm.
Location plays a crucial role in many applications of Wireless Sensor Networks (WSNs), and accurate sensor localization is an important aspect of the acquired data. While connectivity algorithms are commonly used for localizing multi-hop WSNs because their simplicity and acceptable accuracy, their effectiveness can be limited in two-dimensional (2D) or three-dimensional (3D) environments. An analytic model that incorporates hop size quantization and the Recursive Least Squares (RLS) method can be advantageous for Range-Free 3D wireless sensor networks (WSNs) in localization. This approach reduces computational complexity, memory requirements, and localization errors. The third dimension significantly impacts localization accuracy, necessitating the development of effective self-localization algorithms for 3D WSNs. This article introduces a novel probabilistic quantization technique for hop sizes in 3D-WSNs, specifically designed to address the uniform distribution of sensor nodes. The RLS method is employed as an adaptive filtering algorithm to recursively estimate the positions of sensor nodes in the system by minimizing the sum of squared errors between actual measured values and predicted values. Through extensive simulations conducted in isotropic settings under various conditions, the proposed algorithms are evaluated based on their average localization error performance. The simulation data clearly demonstrate that the suggested localization algorithm outperforms previous 3D-DV-Hop heuristics in terms of accuracy. The proposed localization method for 3D-WSNs successfully decreases the average localization error of nodes and achieves superior location accuracy when utilizing predicted hop quantization for hop-size estimation and the RLS algorithm for position estimation compared to competing approaches.
Localization is a crucial concern in many Wireless Sensor Network (WSN) applications. Moreover, getting accurate information about geographic positions of nodes (sensors) is very interesting to make the collected data useful and meaningful. The based connectivity algorithms aim to localize multi-hop WSN thanks to their advantages such as their simplicity and acceptable accuracy. However, the localization accuracy may be relatively low due to environment conditions. An Extreme Learning Machine technique (ELM) is given in this manuscript to minimize the localization error in Range-Free WSN. In this work, based on the Cascade-ELM, we propose a Cascade Extreme Learning Machine (Cascade-ELM) to improve the localization accuracy in Range-Free WSN. We applied the proposed methods in different scenarios of Multi-hop WSN. In our study, we focused on an isotropic and irregular environment. Simulation results prove that the proposed Cascade-ELM algorithm greatly optimizes the localization accuracy in comparison with other algorithms issued from smart computing techniques. Improved localization performances, when compared to previous works, are obtained for isotropic environments.
scite is a Brooklyn-based organization that helps researchers better discover and understand research articles through Smart Citations–citations that display the context of the citation and describe whether the article provides supporting or contrasting evidence. scite is used by students and researchers from around the world and is funded in part by the National Science Foundation and the National Institute on Drug Abuse of the National Institutes of Health.