Enhance data availability and network consistency using artificial neural network for IoT

Abstract

IoT networks have become famous and utilized in many industries such as agriculture, medical, manufacture, and others due to their efficiency and productivity. WSN is an IoT network used for smart farming and smart health monitoring. WSNs can self-manage, self-configure, self-diagnose, and self-heal, making them ideal for agricultural monitoring. A wireless sensor network collects data from numerous sensor nodes scattered across the physical world. WSN data processing is critical when a node fails for unknown reasons. Data handling is an essential aspect of WSN; once any node fails due to unknown reasons, data reliability and availability become crucial. Hence, limited battery energy, low bandwidth, limited computing capacity, and link failure affect network performance. Therefore, an effective cluster-based data aggregation with an appropriate routing must be designed in the media access control. The proposed hybrid artificial neural network and decision tree algorithm with cognitive radio is developed to select the cluster head. The higher amount of residual energy increases the number of packets received at the base station and aggregate the data from the normal sensor nodes. The on-demand routing protocol is designed to keep data in local storage for retransmission during link failure to obtain reliable data transmission. The proposed method performance is analyzed as residual energy, end to end delay, normalized overhead, packet delivery ratio, packet drop, and throughput. This proposed method is evaluated with the cluster-based data aggregation scheme to prove its efficiency. The proposed method residual energy is 8.3Joules for 50 nodes; it is high compared to the cluster-based data aggregation scheme. 2022, The Author(s), under exclusive licence to Springer Science+Business Media, LLC, part of Springer Nature.