An Artificial Intelligence Network based-Host Intrusion Detection System for Internet of Things Devices

• Published in: 2023 4th International Conference on Electronics and Sustainable Communication Systems (ICESC) pp. 656 - 661
• Main Authors: Jain, Mr Ashish, Rao, B. Srinivasa, Chattopadhyay, Saumitra, Kumar, Aniruddh, Muthuraman, M. S., Manjula, A.
• Format: Conference Proceeding
• Language: English
• Published: IEEE 06.07.2023
• Subjects: Analytical models; Artificial Intelligence; Convolutional Neural Network; Deep Learning; Industries; Internet of Things; Intrusion detection; Intrusion Detection System; Measurement; Neural networks; Smart cities; Smart homes
• DOI: 10.1109/ICESC57686.2023.10193232

Internet of Things (IoT) is currently employed in almost all the areas, including applications in smart cities, smart homes, e-Wellbeing, and others. Due to its wider utilization, IoT security has become a serious concern. A secure Intrusion Detection System (IDS) for the Internet of Things is often built using artificial intelligence (AI) and its subsets, deep learning (DL), and machine learning (ML). Industrial IoT devices, which are readily available, are regularly used by researchers and industry experts. This research study investigates the possibility of deploying a DL-Based Host-IDS (DL-HIDS) on specific commercial IoT devices. In this study, an optimized Convolutional Neural Network (O-CNN) based on DL is used. The proposed model's efficiency is evaluated by utilizing performance metrics like recall, precision, accuracy, and f1score. The proposed model's effectiveness is verified by analyzing the promising results obtained from the implementation of the proposed DL-HIDS on various existing models.