Advancing Intrusion Detection Systems for IoT: Techniques, Challenges, and Future Directions

In recent times, a significant rise is observed in the usage of IoT devices, as a repercussion of which threats of cyberattacks have increased. Intrusion Detection system play an important role in identifying malicious activities in networks. This study examines various IDS approaches including hybrid approaches, and highlights key challenges such as heavy reliance on labelled data, constraints in identifying complex attack patterns and explainability in real time, as well as better solutions discussed in section two of this paper. This research derives potential enhancements such as integration of XAI with various models like CNN, Random Forest and federated learning aiming to enhance performance, generalization and adaptability in order to make IDS more scalable and prone to various types of attacks.