A Scalable, Lightweight AI-Driven Security Framework for IoT Ecosystems: Optimization and Game Theory Approaches

Abstract

The rapid growth of IoT has introduced significant security challenges, particularly in scalability, real-time threat detection, and resource management. Traditional security models struggle with an increasing number of interconnected devices, often reacting to threats rather than proactively mitigating them. This study proposes a three-layer security framework that combines artificial intelligence-based intrusion detection, blockchain for decentralized trust management, and edge computing for efficient resource utilization. Machine learning enhances anomaly detection, blockchain ensures secure data integrity, and edge computing reduces latency. Optimization techniques improve the detection accuracy from 94.2% to 94.78%, reduce the response time by 14.98%, and optimize the energy consumption by 12.01%. Game theory models the interactions between attackers and defenders, whereas differential equations simulate system behavior under cyber threats. The performance evaluation demonstrates that the proposed framework provides a scalable, adaptive, and efficient IoT security solution, making it suitable for resource-constrained environments and real-time applications.

Keywords

• Artificial intelligence
• blockchain
• edge computing
• Internet of Things
• intrusion detection systems
• machine learning