Hybrid Multipath Routing Cluster head prediction based on SDN-enabled IoT and Heterogeneous context-aware graph convolution network

Abstract

Intelligent technology and devices have affected nearly every aspect of human lives. The type of connection and means of communication among such a huge number of devices has led to development of Internet of Things (IoT), broad sector that has considerably increased awareness of the problem of energy management and lengthened the lifespan of networks. The design of networks has been made more difficult by complex communications. In this paper, a Hybrid Multipath Routing Cluster head prediction based on software defined networking enabled IoT and Heterogeneous context-aware graph convolution network Using Eagle Swarm Optimization (SDN-IOT-HCAGCN) is proposed. In this method, cluster is formed by coati optimization and cluster head selection by Heterogeneous context-aware graph convolution network, after that Eagle Swarm Optimization is used for multipath routing. The simulations are executed in MATLAB. The proposed technique is evaluated through different metrics, likes delay, number of dead nodes, packet delivery ratio, number of alive node, energy consumption and network life time. The proposed SDN-IOT-HCAGCN model provides 34.49%, 26.44% and 20.51% highest number of alive sensor nodes, 39.45%, 22.44% and 30.24% lowest number of delay when analyzed with the existing models, like Intelligent Energy-Aware Routing Protocol in Mobile IoT depending upon SDN (SDN IEAR-IOT-SDN), Edge-based Optimal Routing in SDN-enabled Industrial Internet of Things (EOR-SDN-IOT), Cooperative Self-scheduling routing protocol basis IoT communication for increasing lifetime duty cycled energy efficient protocol in SDN controlled embedded network (CSR-IOT-SDN) respectively.