Efficient Decentralized Security Service Architecture for Industrial IoT
Thesis event information
Date and time of the thesis defence
Place of the thesis defence
L10, Linnanmaa campus
Topic of the dissertation
Efficient Decentralized Security Service Architecture for Industrial IoT
Doctoral candidate
Master of Science Tharaka Hewa
Faculty and unit
University of Oulu Graduate School, Faculty of Information Technology and Electrical Engineering, Networks and Systems
Subject of study
Communications Engineering
Opponent
Professor Karl Andersson, Luleå University of Technology
Custos
Professor Mika Ylianttila, University of Oulu
Efficient Decentralized Security Service Architecture for Industrial IoT
The current evolution of industrial systems is characterized by expectations of increased production efficiency, data security, regulatory compliance, scalability, and environmental sustainability. One of the key technologies driving these advancements is the Industrial Internet of Things (IIoT), together with 5th Generation(5G) and beyond networks. These enable seamless connectivity between infrastructure, machines, and people, facilitating rapid data exchange, automation, monitoring, and control of industrial systems.
In this thesis, the main contributions are threefold. First, the thesis proposes a novel decentralized service architecture to establish confidentiality, integrity, and authentication of cloud-integrated IIoT. Secondly, the research proposed a decentralized architecture incorporating Game Theory for efficient and secured network slice brokering and service-level agreement establishment. Finally, the thesis proposed a novel consensus mechanism for reliable IIoT data formulation. This thesis proposes to utilise reputation score as a numerical indicator for the IIoT data reliability in combination with BulletProof zero-knowledge proof to defend the data formulation IIoT from slowly adaptive adversaries to yield energy efficiency. Identifying the scalability limitations in the centralized security services, the thesis incorporated blockchain-based smart contracts as a decentralized service enabler that provides decentralization, lower latency, and transparency with cryptographically integrity-preserved ledger.
The proposed service architecture was implemented and evaluated with numerical and programmatic simulations. The thesis results were derived from the comparisons of partial implementations from state-of-art to distinguish the numerical advantages of the proposal. The proposed architecture has yielded significant efficiency improvements, including storage utilization(to 20% in IIoT authentication), latency(Up to 55% in IIoT authentication), resource offer pricing (Up to 21% in slice requests), and energy consumption(Up to 53% in reputation score verification) beyond key state-of-art. In addition, the proposed consensus protocol in the thesis was verified for robustness of chain growth in attack scenarios.
In this thesis, the main contributions are threefold. First, the thesis proposes a novel decentralized service architecture to establish confidentiality, integrity, and authentication of cloud-integrated IIoT. Secondly, the research proposed a decentralized architecture incorporating Game Theory for efficient and secured network slice brokering and service-level agreement establishment. Finally, the thesis proposed a novel consensus mechanism for reliable IIoT data formulation. This thesis proposes to utilise reputation score as a numerical indicator for the IIoT data reliability in combination with BulletProof zero-knowledge proof to defend the data formulation IIoT from slowly adaptive adversaries to yield energy efficiency. Identifying the scalability limitations in the centralized security services, the thesis incorporated blockchain-based smart contracts as a decentralized service enabler that provides decentralization, lower latency, and transparency with cryptographically integrity-preserved ledger.
The proposed service architecture was implemented and evaluated with numerical and programmatic simulations. The thesis results were derived from the comparisons of partial implementations from state-of-art to distinguish the numerical advantages of the proposal. The proposed architecture has yielded significant efficiency improvements, including storage utilization(to 20% in IIoT authentication), latency(Up to 55% in IIoT authentication), resource offer pricing (Up to 21% in slice requests), and energy consumption(Up to 53% in reputation score verification) beyond key state-of-art. In addition, the proposed consensus protocol in the thesis was verified for robustness of chain growth in attack scenarios.
Last updated: 23.1.2024