Fuzzy-based Clustering Scheme with Sink Selection Algorithm for Monitoring Applications of Wireless Sensor Networks

Wireless sensor networks (WSNs) are predominantly used for monitoring applications. The sensor nodes are resource-constrained devices, and hence efficient energy utilization of these nodes is one of the major challenges. The communication distances directly impact on the energy consumption of the sensor nodes. Clustering methods are popularly used to reduce communication distances and prolong the network lifetime. Multi-sink deployment is another method to reduce communication distances. It also resolves congestion and hotspot issues. In multi-sink WSNs, the number of sinks to be considered is a challenging task as it affects the network topology, lifetime and deployment cost. In this research work, multi-sink deployment and clustering scheme with sink selection algorithm are jointly proposed to maximize the network lifetime and minimize the deployment cost. An iterative filtering model is proposed to estimate optimal number of sinks, while sink positions are determined based on Fuzzy logic inference system (FLIS). Fuzzy-c-means algorithm is used to form balanced clusters in the network. Cluster representative and sink selection processes are based on FLIS. The proposed optimal multi-sink deployment scheme reduces the deployment cost and the propagation delay of the system, while enhancing the network lifetime. The proposed scheme is also energy efficient in the case of higher node density. Hence, the proposed scheme can be suitably implemented for large-scale monitoring applications of WSNs.