A Cluster-Based Routing Strategy for Extending the Lifetime of Internet of Things Networks Using Quantum Dilated Convolutional Neural Networks with Emperor Penguin Optimizer

Distributed IoT devices that can sense, process, and exchange data make up Wireless Sensor Networks (WSNs). However, due to their resource constraints, WSNs have trouble operating efficiently and conserving energy. Improved solutions are needed since current clustering methods frequently lead to short network lifespans, uneven load distribution, and excessive end-to-end latency. Using Multi-Hop Low Energy Adaptive Clustering Hierarchy (MH-LEACH) to optimize Cluster Head (CH) selection based on residual energy, distance, overhead, and trust, a Quantum Dilated Convolutional Neural Network with Emperor Penguin Optimizer (QDCNNet-EPO) is proposed as a solution to these problems. Performance is improved by utilizing EPO to adjust the QDCNNet's hyper parameters. With an energy usage of only 0.25 J, this method produces excellent results, such as 30% packet loss, 350 kbps throughput, and a much longer network lifetime, ranging from 98% (100 nodes) to 94% (500 nodes). With possible future applications in dynamic IoT and social IoT contexts, the suggested approach guarantees energy efficiency and balanced load distribution.