Date of Award
2023
Document Type
Thesis
Publisher
Edith Cowan University
Degree Name
Master of Computing and Security by Research
School
School of Science
First Supervisor
David Cook
Second Supervisor
Leisa Armstrong
Abstract
Modern technology has proliferated into just about every aspect of life while improving the quality of life. For instance, IoT technology has significantly improved over traditional systems, providing easy life, time-saving, financial saving, and security aspects. However, security weaknesses associated with IoT technology can pose a significant threat to the human factor. For instance, smart doorbells can make household life easier, save time, save money, and provide surveillance security. Nevertheless, the security weaknesses in smart doorbells could be exposed to a criminal and pose a danger to the life and money of the household. In addition, IoT technology is constantly advancing and expanding and rapidly becoming ubiquitous in modern society. In that case, increased usage and technological advancement create security weaknesses that attract cybercriminals looking to satisfy their agendas.
Perfect security solutions do not exist in the real world because modern systems are continuously improving, and intruders frequently attempt various techniques to discover security flaws and bypass existing security control in modern systems. In that case, threat modelling is a great starting point in understanding the threat landscape of the system and its weaknesses. Therefore, the threat modelling field in computer science was significantly improved by implementing various frameworks to identify threats and address them to mitigate them. However, most mature threat modelling frameworks are implemented for traditional IT systems that only consider software-related weaknesses and do not address the physical attributes. This approach may not be practical for IoT technology because it inherits software and physical security weaknesses. However, scholars employed mature threat modelling frameworks such as STRIDE on IoT technology because mature frameworks still include security concepts that are significant for modern technology. Therefore, mature frameworks cannot be ignored but are not efficient in addressing the threat associated with modern systems.
As a solution, this research study aims to extract the significant security concept of matured threat modelling frameworks and utilise them to implement robust IoT threat modelling frameworks. This study selected fifteen threat modelling frameworks from among researchers and the defence-in-depth security concept to extract threat modelling techniques. Subsequently, this research study conducted three independent reviews to discover valuable threat modelling concepts and their usefulness for IoT technology. The first study deduced that integration of threat modelling approach software-centric, asset-centric, attacker-centric and data-centric with defence-in-depth is valuable and delivers distinct benefits. As a result, PASTA and TRIKE demonstrated four threat modelling approaches based on a classification scheme. The second study deduced the features of a threat modelling framework that achieves a high satisfaction level toward defence-in-depth security architecture. Under evaluation criteria, the PASTA framework scored the highest satisfaction value. Finally, the third study deduced IoT systematic threat modelling techniques based on recent research studies. As a result, the STRIDE framework was identified as the most popular framework, and other frameworks demonstrated effective capabilities valuable to IoT technology.
Respectively, this study introduced Defence-aware Threat Modelling (DATM), an IoT threat modelling framework based on the findings of threat modelling and defence-in-depth security concepts. The steps involved with the DATM framework are further described with figures for better understatement. Subsequently, a smart doorbell case study is considered for threat modelling using the DATM framework for validation. Furthermore, the outcome of the case study was further assessed with the findings of three research studies and validated the DATM framework. Moreover, the outcome of this thesis is helpful for researchers who want to conduct threat modelling in IoT environments and design a novel threat modelling framework suitable for IoT technology.
DOI
10.25958/22en-mk10
Recommended Citation
Godakanda, M. L. (2023). Optimising a defence-aware threat modelling diagram incorporating a defence-in-depth approach for the internet-of-things. Edith Cowan University. https://doi.org/10.25958/22en-mk10