Fault tolerant cluster-based routing in wireless sensor network using novel intelligent deep learning methodology

Abstract

Wireless Sensor Networks (WSNs) are highly susceptible to cluster head (CH) failures due to limited node energy and unstable communication links, resulting in data loss and reduced network reliability. This work presents a fault-tolerant cluster-based routing model that integrates an Improved Graph Neural Network (IGNN) with Multi-Objective Crested Porcupines Optimization (MOCPO) for efficient recovery from CH failures. The model constructs a virtual CH at the sink node by organizing the residual energy and connectivity information of failure-free CHs. The proposed IGNN uses essential parameters such as node position, Euclidean distance, residual energy, connectivity degree, packet delay, and fault-tolerance score to identify optimal fault-recovery paths. MOCPO optimizes IGNN parameters under a multi-objective fitness function incorporating energy consumption, end-to-end delay, connectivity measure, and fault-tolerance metrics. Simulation results with 1000 sensor nodes, 10 faulty CHs, packet size of 500 bytes, and 200 optimization iterations show that the proposed IGNN-MOCPO model reduces energy consumption by 32.72% and end-to-end delay by 33.14%, while significantly improving packet delivery ratio, residual energy, and network lifetime compared to existing FTCR schemes. These results demonstrate the efficiency, stability, and adaptability of the proposed framework under dense and fault-prone WSN deployments.