The field of study for this work centres on enhancing security within the expanding domain of the internet of things (IoT), where the need for reliable detection of malicious activities is critical. As IoT integrates a wide array of applications and hardware, the inherent online nature of these technologies makes vital infrastructure susceptible to cyberattacks. Despite the involvement of a significant community in critical applications like CPSs, traditional computational methodologies in anomaly-based programs often prove insufficient. This study aims to identify and classify issues at both the network and host levels using advanced ML and DL models, which offer promising solutions. Specifically, the research employs the IoT-23 dataset to conduct a comprehensive analysis using algorithms such as DT, SVM, and ECLDNN. By evaluating the precision and energy efficiency of these classifiers, the study seeks to determine the most accurate and time-efficient solution for defect detection in IoT systems. This work advances the field by proposing and validating sophisticated ML and DL techniques that significantly improve the detection and classification of cyber threats, thereby enhancing the security of IoT infrastructure.
Vardhan et al. (Thu,) studied this question.