A Comprehensive Survey of Contemporary Anomaly Detection Methods for Securing Smart IoT Systems

Attacks are growing more complex and dangerous as network capabilities improve at a rapid pace. Network intrusion detection is usually regarded as an efficient means of dealing with security attacks. Many ways have been presented, utilizing various strategies and focusing on different types of visitors. Anomaly-based network intrusion monitoring is an essential area of intrusion detection investigation and development. Despite extensive research on anomaly-based network detection, there is still a lack of comprehensive literature reviews covering current methodologies and datasets. Despite the substantial research into anomaly-based network intrusion detection algorithms, there is a dearth of a research evaluation of new methodologies and datasets. We explore and evaluate 50 highest publications on anomaly-based intrusion detection using an in-depth review of related literature techniques. Our work thoroughly explores the technological environment of the subject in order to help future research in this sector. Our examination is carried out from the relevant angles: application areas, data preprocessing and threat detection approaches, assessment measures, and datasets. We select unresolved research difficulties and underexplored research areas from every viewpoint recommendation of the study. Finally, we outline five potentially increased research areas for the future.