A Novel QoS Based Secure Unequal Clustering Protocol with Intrusion Detection System in Wireless Sensor Networks

Wireless sensor network (WSN) becomes a hot research topic owing to its application in different fields. Minimizing the energy dissipation, maximizing the network lifetime, and security are considered as the major quality of service (QoS) factors in the design of WSN. Clustering is a commonly employed energy-efficient technique; however, it results in a hot spot issue. This paper develops a novel secure unequal clustering protocol with intrusion detection technique to achieve QoS parameters like energy, lifetime, and security. Initially, the proposed model uses adaptive neuro fuzzy based clustering technique to select the tentative cluster heads (TCHs) using three input parameters such as residual energy, distance to base station (BS), and distance to neighbors. Then, the TCHs compete for final CHs and the optimal CHs are selected using the deer hunting optimization (DHO) algorithm. The DHO based clustering technique derives a fitness function using residual energy, distance to BS, node degree, node centrality, and link quality. To further improve the performance of the proposed method, the cluster maintenance phase is utilized for load balancing. Finally, to achieve security in cluster based WSN, an effective intrusion detection system using a deep belief network is executed on the CHs to identify the presence of intruders in the network. An extensive set of experiments were performed to ensure the superior performance of the proposed method interms of energy efficiency, network lifetime, packet delivery ratio, average delay, and intrusion detection rate.