IntelliGuard: An AI-Powered Threat Detection System for Smart Home Operations
Sharmin Akter, Md. Maruf Hossain Munna, Kazi Redwan, Mustakim Ahmed, Md. Faruk Abdullah Al Sohan · 2025
The growing application of smart home technologies has exposed significant vulnerabilities in cybersecurity. This research proposes a lightweight, AI-powered Threat Detection System (TDS) aimed at improving the security of IoT devices in smart homes by combining Azure Cloud and Google Big-Query for efficient data storage and processing. The proposed system implements firewalls and Transport Layer Security (TLS) encryption to secure communication channels and safeguard critical data from unauthorized access. The TDS utilizes Google BigQuery to analyze extensive IoT data for trained threat detection and mitigation decisions. Some necessary algorithms, such as Deep Q-Networks (DQN) and autoencoders-reinforce its functionality. DQN separates devices exhibiting atypical behavior, and autoencoders identify anomalies in device activity, network traffic, and access patterns. The technology quarantines the damaged device to avoid further damage, using safe data processing in Azure Cloud under strict encryption procedures. The TDS continually analyses data streams in real-time, instantly notifying administrators through a mobile application when abnormalities arise and autonomously upgrading security policies to address evolving threats. This research shows the essential function of cloud-based security solutions and artificial intelligence in safeguarding smart home ecosystems against evolving cyber threats.