IWDSA: a hybrid Intelligent Water Drops with a Simulated Annealing for the localization improvement in wireless sensor networks

• Main Authors: Gumaida, Bassam, Ibrahim, Adamu Abubakar
• Format: Article
• Language: English
• Published: Telkom University 2024
• Subjects: Q350 Information theory
• Online Access: http://irep.iium.edu.my/113095/1/113095_IWDSA.pdf; http://irep.iium.edu.my/113095/; https://journals.telkomuniversity.ac.id/ijait/article/view/6456

Improving localization accuracy and reducing development costs are pivotal keys and main issues in managing and administrating wireless sensor networks (WSNs). This paper considers a modern and qualified algorithm that leverages advanced optimization techniques to localize nodes deployed in outdoor environments. The proposed algorithm, named Intelligent Water Drops with Simulated Annealing (IWDSA), combines two powerful optimization methods: Intelligent Water Drops (IWD) and Simulated Annealing (SA). IWD is a qualified stochastic optimization tool adept at minimizing objective functions. In IWDSA, SA is integrated to enhance solution quality and prevent IWD from getting trapped in local minima. This paper ensures that internal distances between nodes are calculated using Received Signal Strength Indicator (RSSI) measurements. The paper aims to achieve two primary goals. First, it addresses the challenge of low accuracy in RSSI measurements by employing IWDSA. Second, it aims to achieve highly accurate localization of unknown sensor nodes in WSNs. IWDSA enhances localization precision due to its flexible implementation of IWD and SA, combined with the cost-free utilization of RSSI. Simulation results demonstrate the reliable performance of the proposed algorithm in solving the low accuracy of RSSI measurements and localizing unknown nodes with high accuracy. Additionally, simulation results confirm that the proposed algorithm IWDSA exhibits outstanding performance compared to other algorithms utilizing optimization techniques, including genetic algorithms, bat algorithms, ant colony optimization, and swarm optimization. This exceptional performance is evident across various evaluation metrics, including localization error, localization rate, and simulation runtime