Trust-Based Cluster Head Selection Using the K-Means Algorithm for Wireless Sensor Networks

Abstract

Clustering sensor nodes in a wireless sensor network is a key technique to reduce the energy consumption of sensors which extends the lifetime of the network. The head of the cluster plays a key role in a network and serves as a router. In addition, the head of the cluster is responsible for collecting and transmitting sensed information from their cluster members to a destination node or base station/sink. Hence, an efficient clustering approach is required to safely elect a cluster head. It remains a critical task for overall network performance. As a result, in this research approach, we proposed a scheme for selection of cluster head method based on a trust factor that ensures all nodes are trustworthy and authentic during communication. To achieve this, direct trust is calculated using parameters such as the residual energy and the distance between the nodes, along with the use of the k-means clustering algorithm. The simulation results show that in terms of network lifetime, packet delivery ratio, packet drop rate, and energy consumption, the proposed solution outperforms the LEACH protocol. In addition, this strategy can significantly improve performance while discriminating against the network's legitimate and malicious (or compromised) nodes.