A Genetic-Muted Leader Scheme with CNN-Based IDS for Industrial Internet of Things Networks

The industrial industry has adopted web and cloud-based technologies widely, which has accelerated the IoT ecosystem’s expansion. Industrial Internet of Things (IIoT) has established a large network due to the sheer volume of information and linked equipment. IIoT networks are made to be vulnerable in terms of cybersecurity. So as to ensure the security of the IIoT networks, intrusion detection systems (IDS) must be developed. This study suggests an IDS-based on a Genetic-Muted Leader Scheme (GMLS)-CNN for usage in IIoT networks. The data collection and preprocessing steps improve the detection accuracy. The next step is to classify data using convolutional neural networks (CNNs), with GMLS used to fine-tune the parameters for the most accurate results. The two datasets such as UNSW-NB15 and X-IIoTID and named as TD were used to distinguish normal and problematic data, which allowed the study to evaluate several categorization techniques. Among the other schemes, the GMLS-CNN model had the maximum accuracy rate for IDSis 98.7% (X-IIoTID) and 95.21% (UNSW-NB15). Additionally, the model’s success in correctly detecting attacks based on the various types of attacks reported in the datasets was assessed.