Violation Recognition Approaches for IoT Systems Employing Machine Learning and Deep Learning Techniques

The rise of networked tools in the Internet of Things (IoT) era has led to a surge in intrusion attempts. An Intrusion Detection System (IDS) acts as a smart auxiliary system, responsible for monitoring, detecting, and alerting about potential malicious activities. IDS is crucial for creating a robust security model. Although many studies review detection models, they often lack consistency in their advancements. Moreover, existing models have limitations that must be addressed to develop new security frameworks. We explore the significance of different techniques, tools, and methods based on Artificial Intelligence in IoT identification and prohibition systems, specifically highlighting the role of Machine Learning (ML) and Deep Learning (DL) approaches in interference perception and protection systems. This survey is beneficial for both industry and academia, helping to identify challenges and issues in current security models and to develop new security frameworks using efficient ML or DL methods.