An AI-Driven Intelligent Security Framework for IoT and Wireless Sensor Networks
An AI-Driven Intelligent Security Framework for IoT and Wireless Sensor Networks
Hooman Hosseinieh Farahani1 Somayeh Azizi2
1) Ph.D. Student in Computer Engineering, Department of Computer Engineering, Islamic Azad University, Arak Branch, Arak, Iran
2) Ph.D. in Software Engineering, Faculty Member, Department of Computer Engineering, Islamic Azad University, Arak Branch, Arak, Iran
IOI Code
:
AAHD-ABKCKB
Publication :
National Conference on Information Technology, Nanotechnology, Artificial Intelligence and Technological Futures Studies(itnaf.ir)
Abstract :
The rapid proliferation of the Internet of Things (IoT) and Wireless Sensor Networks (WSNs) as key infrastructures in smart cities, digital healthcare, and industrial systems has significantly expanded the attack surface and made these environments attractive targets for cyber adversaries. The inherent constraints of such networks, including limited computational power and energy resources, node heterogeneity, and distributed architectures, have substantially reduced the effectiveness of conventional security mechanisms and highlighted the need for intelligent, lightweight, and adaptive security solutions capable of operating under dynamic conditions.
This paper proposes an intelligent multilayer security framework based on artificial intelligence for threat detection and prediction in IoT and WSN environments. Adopting a data-driven approach, the proposed framework performs real-time network behavior analysis and identifies complex attack patterns. At its core, a purposeful integration of deep learning and machine learning techniques, including both supervised and unsupervised models, is employed to detect anomalies, reduce dependence on labeled datasets, and enable the identification of previously unseen zero-day attacks. Furthermore, intelligent feature extraction and dimensionality reduction mechanisms are incorporated to optimize computational cost and facilitate deployment on resource-constrained sensor nodes.
The main contributions of this study include the design of a hierarchical threat detection architecture, the integration of network traffic analysis with node behavior analysis, and the application of adaptive learning for continuous model updating against emerging threats. The proposed framework is capable of accurately detecting a wide range of common attacks, such as Denial-of-Service (DoS), Sybil attacks, and impersonation attacks. In addition, by exploiting spatiotemporal pattern analysis, it can predict threat evolution trends and provide proactive early warnings.
Experimental evaluations conducted on benchmark datasets and simulated scenarios demonstrate that the proposed method achieves significant improvements in detection accuracy, attack discovery rate, and false positive reduction compared with baseline approaches, while maintaining computational overhead and energy consumption at an optimized level. This balance between security effectiveness and resource efficiency makes the framework a practical and scalable solution for deployment in real-world IoT and WSN environments.
In conclusion, this paper presents a comprehensive, adaptive, and AI-driven security framework that contributes to enhancing the cyber resilience of IoT and WSN infrastructures and provides a promising foundation for the development of next-generation intelligent defense systems in connected ecosystems.
**Keywords:** Internet of Things (IoT); Wireless Sensor Networks (WSNs); AI-based Intrusion Detection; Machine Learning; Deep Learning; Anomaly Detection; Lightweight Security; Zero-day Attacks.
Keywords : internet of things (iot) wireless sensor networks (wsns) ai based intrusion detection machine learning deep learning anomaly detection lightweight security zero day attacks