Linearly decreasing inertia weight PSO and improved weight factor-based clustering algorithm for wireless sensor networks

• Published in: Journal of ambient intelligence and humanized computing Vol. 14; no. 6; pp. 6661 - 6679
• Main Authors: Choudhary, Shilpa, Sugumaran, S., Belazi, Akram, El-Latif, Ahmed A. Abd
• Format: Journal Article
• Language: English
• Published: Berlin/Heidelberg Springer Berlin Heidelberg 01.06.2023; Springer Nature B.V
• Subjects: Algorithms; Artificial Intelligence; Clustering; Communication; Computational Intelligence; Energy consumption; Engineering; Inertia; Literature reviews; Microelectromechanical systems; Nodes; Optimization algorithms; Optimization techniques; Original Research; Parameters; Particle swarm optimization; Protocol; Residual energy; Robotics and Automation; Sensors; User Interfaces and Human Computer Interaction; Wireless sensor networks; Wireless sensor networks; Clustering; Particle swarm optimization (PSO); Energy-efficient
• ISSN: 1868-5137; 1868-5145
• DOI: 10.1007/s12652-021-03534-w

In wireless sensor networks (WSNs), clustering is one of the most effective routing protocols. Most clustering algorithms include two stages of operations: cluster head selection and cluster formation. Cluster heads are selected from the sensor nodes based on several key parameters like residual energy of the cluster heads candidates, the distance of cluster heads from their cluster members, and the distance between cluster head and the base station. Cluster formation deals with the association of sensor nodes with one of the selected cluster heads. This paper presents an energy-efficient clustering algorithm with linearly decreasing inertia weight particle swarm optimization (PSO) and improved weight factor. The interia weight PSO is based on cluster head selection and the improved weight factor-based cluster formation. The merit of the proposed approach is the robust formulation of a linear weight factor that leads to efficient cluster formation. The efficacy of the proposed algorithm is verified via different scenarios with varying numbers of nodes and different positions of base stations. Results are compared with some of the existing algorithms, and it is found that the proposed approach outperforms other approaches in terms of various evaluation parameters.