WOGRU-IDS — An intelligent intrusion detection system for IoT assisted Wireless Sensor Networks

Abstract

One of the key mechanisms of the current electronic and wireless frameworks is the assistance of Wireless Sensor Networks (WSN) in Internet of Things (IoT) networks. A WSN typically consists of multipurpose sensor hubs for data sensing, processing, and communication. These networks are more suited to conveying medical data from various geographical regions and sending private medical data to the network owner. However, the worry about various attacks on health care data normally grows daily. These assaults could quickly have adverse impacts on the WSN-IoT (Internet of Things) nodes. Additionally, the low detection rate, significant processing overhead, and resource limitations of current intrusion detection systems all contribute to an increase in false alarm rates when trying to identify various attacks. The unique Whale Optimized Gate Recurrent Unit (WOGRU) Intrusion Detection System (IDS) for WSN-IoT networks is proposed in this research in light of the aforementioned issues in order to effectively identify various attacks. The whale algorithm was used in the proposed framework to tune the hyperparameters of the deep long short-term memory in order to achieve low computational overhead and great performance. Last but not least, validations are carried out using the WSN-DS dataset, and the performance of the suggested work is evaluated using the parameters accuracy, recall, precision, specificity, and F1-score. Additionally, the comparison study was conducted using the current frameworks. The data demonstrates that the suggested framework had an average performance of 99.85 percent for the detection of flooding, scheduling, black hole, and gray hole attacks.