An energy-efficient cluster routing for internet of things-enabled wireless sensor network using mapdiminution-based training-discovering optimization algorithm

In recent decades, the Internet of Things (IoT)-enabled Wireless Sensor Network (IWSN) facilitates to development of numerous real-time applications. IWSN has become significant expertise in acquiring a better quality of service, long-term consistency, and low-cost management. Nevertheless, the sensor nodes of IWSN typically have restricted battery energy and are vulnerable to several intrusion attacks. To address the constraints of IWSN, an energy-efficient clustering and rapid intrusion detection system have been proposed. A novel MapDiminution-based Training-Discovering Optimization method is employed in the proposed framework to obtain optimal cluster routing path from each cluster to sink. Once the route is determined, the MapDiminution model invokes the task scheduling process in which each cluster member is managed with the queuing framework. This optimum path and scheduling process reduces the energy consumption in IWSN. Afterward, the Hybrid classifier can be formulated by integrating Artificial Neural Network (ANN) with Simulated Annealing (SA). The weights of ANN are optimized through the SA where the different types of intrusion attacks are then classified based on received information from the cluster nodes. The simulation results expose that the proposed framework achieves a lesser energy intake of 0.01 J and a higher detection accuracy of 97.57% as compared to the existing methods.