Context aware cyber physical security for smart homes

Majib, Yasar 2025. Context aware cyber physical security for smart homes. PhD Thesis, Cardiff University. Item availability restricted.

	PDF - Accepted Post-Print Version Restricted to Repository staff only until 16 January 2027 due to copyright restrictions. Available under License Creative Commons Attribution Non-commercial No Derivatives. Download (3MB)
	PDF (Cardiff University Electronic Publication Form) - Supplemental Material Restricted to Repository staff only Download (38kB)

Abstract

The Internet of Things has transformed traditional homes into smart homes by integrating cyber systems with the physical environment, enabling enhanced automation and data collection from diverse sensors. While these advancements offer numerous benefits, they also introduce vulnerabilities that pose substantial security and privacy risks to smart home residents. Anomaly detection techniques are crucial for monitoring and mitigating these risks by identifying irregularities in both cyber, e.g., network traffic and physical (e.g., environmental sensor readings) layers. Therefore, this thesis aims to develop and evaluate novel frameworks for detecting behavioural anomalies, particularly unknown types, using fused cyber-physical data in smart homes. Key contributions include a comprehensive cyber-physical smart home dataset, a thorough survey and decision framework, and the development of two novel detection systems (RT-SHAD and CP-BASH). First, this thesis presents a comprehensive survey of anomaly detection methods in smart home environments, categorising anomalies into known and unknown types. Based on the analysis, it proposes a decision-making framework to guide the implementation of effective anomaly detection systems in smart homes. The second, RT-SHAD, is a real-time smart home anomaly detection solution which compares resource-constrained and resource-rich devices for the feasibility of anomaly detection setup within a smart home to preserve residents’ privacy. By implementing RT-SHAD, it is found that a resource-constrained device such as RaspberryPi can be used for the purpose with minor delay. The third, CP-BASH, is a two-stage approach to detect behavioural anomalies in smart homes using fused cyber-physical data on a singular timeline of over 27 activities from a real smart home equipped with various IoTs. The system effectively detects activities by applying and comparing various machine learning techniques on fused and individual data streams. It predicts anomalous behaviours, with the optimal approach achieving 98.52% composite accuracy in detecting abnormal behaviour in a smart home environment

Item Type:	Thesis (PhD)
Date Type:	Completion
Status:	Unpublished
Schools:	Schools > Computer Science & Informatics
Subjects:	Q Science > QA Mathematics > QA75 Electronic computers. Computer science
Date of First Compliant Deposit:	16 January 2026
Date of Acceptance:	12 January 2026
Last Modified:	20 Jan 2026 14:58
URI:	https://orca.cardiff.ac.uk/id/eprint/183948

Actions (repository staff only)

Edit Item