Skip to main content

Indeterminacy-aware prediction model for authentication in IoT.

Heydari, M., 2020. Indeterminacy-aware prediction model for authentication in IoT. Doctoral Thesis (Doctoral). Bournemouth University.

Full text available as:

HEYDARI, Mohammad_Ph.D._2020.pdf



The Internet of Things (IoT) has opened a new chapter in data access. It has brought obvious opportunities as well as major security and privacy challenges. Access control is one of the challenges in IoT. This holds true as the existing, conventional access control paradigms do not fit into IoT, thus access control requires more investigation and remains an open issue. IoT has a number of inherent characteristics, including scalability, heterogeneity and dynamism, which hinder access control. While most of the impact of these characteristics have been well studied in the literature, we highlighted “indeterminacy” in authentication as a neglected research issue. This work stresses that an indeterminacy-resilient model for IoT authentication is missing from the literature. According to our findings, indeterminacy consists of at least two facets: “uncertainty” and “ambiguity”. As a result, various relevant theories were studied in this work. Our proposed framework is based on well-known machine learning models and Attribute-Based Access Control (ABAC). To implement and evaluate our framework, we first generate datasets, in which the location of the users is a main dataset attribute, with the aim to analyse the role of user mobility in the performance of the prediction models. Next, multiple classification algorithms were used with our datasets in order to build our best-fit prediction models. Our results suggest that our prediction models are able to determine the class of the authentication requests while considering both the uncertainty and ambiguity in the IoT system.

Item Type:Thesis (Doctoral)
Additional Information:If you feel that this work infringes your copyright please contact the BURO Manager.
Uncontrolled Keywords:authentication; internet of things; uncertainty; trust; prediction models; risk; machine learning
Group:Faculty of Science & Technology
ID Code:34577
Deposited By: Symplectic RT2
Deposited On:22 Sep 2020 12:40
Last Modified:14 Mar 2022 14:24


Downloads per month over past year

More statistics for this item...
Repository Staff Only -