An Optimal Cluster Head Selection in Wireless Sensor Network Utilizing an Unsupervised Learning Algorithm

Energy consumption is one of the most vital factors to be considered when designing a wireless sensor network (WSN). In this paper, a new cluster head (CH) selection mechanism is introduced which selects the CH node such as it is close to the most active nodes, thus minimizing the energy consumption of these nodes particularly, and the entire network generally. This is attained through monitoring the traffic of the nodes and forming an unconstraint optimization problem that aims to find the optimal CH location that minimizes the nodes’ energy consumption. Furthermore, the proposed algorithm utilizes an unsupervised machine learning clustering algorithm in the clustering formation process. The conducted simulation results show that the proposed algorithm have achieved the goal of enhancing the network lifetime and outperformed other existing work in the literature.