Fuzzy‐based cluster head selection and cluster formation in wireless sensor networks. Issue 6 (1st November 2019)

• Record Type: Journal Article
• Authors: Shivappa, Nagesha; Manvi, Sunilkumar S.
• Is Part Of: IET networks. Volume 8:Issue 6(2019)
• Page Start: 390
• Page End: 397
• Publication Date: 2019-11-01
• Subjects: wireless sensor networks -- pattern clustering -- fuzzy set theory; respective cluster head -- cluster centre -- optimal clustering mechanism -- node death -- wireless sensor networks -- bandwidth requirements as well as offers improved network lifetime -- centralised cluster head selection -- distributed cluster formation scheme -- fuzzy clustering -- fuzzy inference system; Computer network architectures -- Periodicals; Computer network protocols -- Periodicals; Information networks -- Periodicals; Telecommunication systems -- Periodicals; 004.605
• DOI: 10.1049/iet-net.2018.5102
• Languages: English
• ISSNs: 2047-4954

Abstract

Abstract : Clustering is one of the mechanisms for routing in wireless sensor networks which reduces the energy and bandwidth requirements to improve network lifetime. A centralised cluster head selection and distributed cluster formation scheme by using fuzzy techniques is proposed in this study. The fuzzy c‐means is used by the sink to find cluster centres and their associated member nodes. A cluster head is selected from each cluster centre associated member nodes based on the fitness of the nodes which is computed using residual energy and distance to the cluster centre as inputs to fuzzy inference system. Nodes associate with a cluster head depending on their proximity, thereby forming clusters. Proximity will be found out by node with the assistance of signal strength parameter from cluster head which broadcast selection advertisement. This completes a round of cluster formation. At the end of each round, the nodes transmit their fitness value to their respective cluster head which in turn is forwarded to the sink. The sink selects a node with highest fitness value from each cluster centre associated members as cluster head for the next round. The scheme is compared with self‐management model and optimal clustering mechanism to demonstrate the improved performance.