Article: Feedback-based trust module for IoT networks using machine learning Journal: International Journal of Wireless and Mobile Computing (IJWMC) 2024 Vol.27 No.1 pp.78 - 91 Abstract: The increasing emergence of IoT in healthcare, industrial automation, manufacturing, infrastructure, business and the home undoubtedly provides more conveniences in different aspects of human life. Any IoT security and trust solution must consider the heterogeneity and mobility of IoT devices, and the features of other devices connected to the network. Traditional security measures face computation, storage and communication challenges in securing IoT, and the increasing reliance on IoT services and applications has exacerbated this issue. It is crucial to identify a security management model to handle these constraints. Trust as a security regime in the IoT has far-reaching implications. This paper proposes a novel trust-based security mechanism by combining machine learning and traditional summations to determine a node's trustworthiness in an IoT environment in milliseconds. Based on this, a computational trust management model is designed by aggregating direct and feedback-based trust information. The model parameters like the trust breach window and trust re-initiation point are selected based on extensive experimentation to dynamically calculate each node's trust value to reach the proposed model's peak performance. The model's effectiveness is verified through simulation. In 500 iterations, the proposed model achieved 97% accuracy with a run time of 0.37 seconds. Inderscience Publishers - linking academia, business and industry through research

Title: Feedback-based trust module for IoT networks using machine learning

Authors: Sania Iqbal; Shaima Qureshi

Addresses: Department of Computer Science and Engineering, National Institute of Technology, Srinagar, Jammu & Kashmir, India ' Department of Computer Science and Engineering, National Institute of Technology, Srinagar, Jammu & Kashmir, India

Abstract: The increasing emergence of IoT in healthcare, industrial automation, manufacturing, infrastructure, business and the home undoubtedly provides more conveniences in different aspects of human life. Any IoT security and trust solution must consider the heterogeneity and mobility of IoT devices, and the features of other devices connected to the network. Traditional security measures face computation, storage and communication challenges in securing IoT, and the increasing reliance on IoT services and applications has exacerbated this issue. It is crucial to identify a security management model to handle these constraints. Trust as a security regime in the IoT has far-reaching implications. This paper proposes a novel trust-based security mechanism by combining machine learning and traditional summations to determine a node's trustworthiness in an IoT environment in milliseconds. Based on this, a computational trust management model is designed by aggregating direct and feedback-based trust information. The model parameters like the trust breach window and trust re-initiation point are selected based on extensive experimentation to dynamically calculate each node's trust value to reach the proposed model's peak performance. The model's effectiveness is verified through simulation. In 500 iterations, the proposed model achieved 97% accuracy with a run time of 0.37 seconds.

Keywords: internet of things; machine learning; decision trees; trust management; network security.

DOI: 10.1504/IJWMC.2024.139801

International Journal of Wireless and Mobile Computing, 2024 Vol.27 No.1, pp.78 - 91

Received: 03 Dec 2022
Accepted: 23 Aug 2023
Published online: 05 Jul 2024 *

Full-text access for editors Full-text access for subscribers Purchase this article Comment on this article

Title: Feedback-based trust module for IoT networks using machine learning

Keep up-to-date