Investigation on Attack Detection in IoT Networks: A Study and Analysis of the Existing Machine Learning and Deep Learning Techniques

Despite revolutionizing fields like eHealth, smart cities, and autonomous systems, the Internet of Things (IoT) has presented serious security challenges due to its widespread usage. This study gives an extensive review of current deep learning (DL) and machine learning (ML) techniques aimed at improving IoT network security. It examines different methods for identifying and thwarting a range of security risks, including malware and botnet attacks, evaluating their effectiveness, computational needs, and dataset consumption. The outcome of the survey helps in understanding the drawbacks of the current existing approaches such as high complexity, unbalanced datasets, and scalability problems. Additionally, the study examines new developments in ensemble models and hybrid architectures to increase detection efficiency and accuracy. The study provides a thorough grasp of the state of IoT security solutions today and their suitability for dealing with changing cyber threats, thanks to this examination.