AI-Based Intrusion Detection & Prevention Models for Smart Home IoT Systems: A Literature Review

Abstract

As smart home IoT devices are being increasingly adopted, a lot of cybersecurity concerns have arisen and hence, more need for advanced security measures. This study presents an AI based Intrusion Detection and Prevention Systems (IDPS) for smart home environments. AI driven IDPS utilizes ML and DL techniques to increase threat detection accuracy, and reducing false positives through adaptive security mechanisms. The results show that hybrid detection models that integrate the signature and the anomaly detection algorithms establish a robust countermeasure to known and unknown cyber threats. But true to the work’s objectives, there are also key challenges that must be addressed for deployment in the real world, such as computational overhead, privacy considerations and adversarial attacks. Future work can be oriented toward developing lightweight AI models, integrating explainable AI (XAI), or looking into techniques that can be used for keeping the data private, for example, using techniques from federated learning. Moreover, such threat intelligence sharing frameworks on blockchain can further strengthen security in the connected smart home ecosystems. Future results from these advancements will lead to more resilient and efficient cybersecurity systems for smart home IoT systems.