A Self-learning Clustering Protocol in Wireless Sensor Networks for IoT Applications

Abstract

The integration of Wireless sensor networks (WSN) and Internet of Things (IoT) perform many tasks control or monitor the surrounding area or the environment. The WSN-based IoT consists of many sensor nodes connect which transmit the collecting data of the environment to the manager through the Internet. The network topology requires high reliability connections while requires low energy consumption at the sink node and long network lifetime. In this paper, we introduce the self-learning clustering protocol to discover neighbors and the network topology. The cluster head is selected based on the information of the neighbors and the residual energy of the node. The maximum number of cluster members is set according to the network density. The proposed protocol can adapt the changing of the dynamic network with low energy consumption; therefore, ensuring the network connectivity. The simulation results show that the proposed clustering protocol performs well in terms of long network lifetime and high throughput while comparing to other clustering protocols.