Work package (WP1) –ESR1 (BCU) sought to design IoT context-aware framework in heterogeneous sensors and network devices for smart water networks by bringing the analytics, context-awareness, and decision-making closer to the network edge to achieve nearly real-time water quality monitoring by using smart sensing and IoT technologies. This research focused on optimizing the deployment of water quality sensors to reduce the impact of water contamination on human health; Machine learning (ML) and edging computing were explored to develop scalability, reliability, and a context-aware water quality monitoring framework. WP1 generated 5 publications 4 technical reports and a Ph.D. thesis. ESR1 successfully obtained his Ph.D. degree in May 2022 at BCU. WP1 main achievements:
1. Proposed a sensor deployment method that facilitates near/real-time water quality monitoring by multi-objective optimisation within multi-contamination sources.
2. Analysed the impact on human health in simulating contaminated water networks based on the deployed sensors.
3. Developed machine learning and deep learning models to enhance water contaminant detection.
4. Developed the proposed context-aware framework integrating the results 1-3 for near/real-time monitoring in WDS
Work package 2 (WP2) –ESR2 (SLG-GR) investigated large-scale stream processing solutions to interlink data from IoTs and extract real-time information for SWN, covering the integration of heterogeneous data streams, and large-scale IoT stream processing, and real-time information processing and knowledge extraction. Data Information Interoperability Model (DIIM) can help water utilities to analyse data to help decision making. WP2 generated 3 technical reports and one publication on IEEE ICSC 2022. WP2 main achievements:
1. Reviewed interoperability challenges faced by IoT-enabled SWN.
2. Developed a Data Information Interoperability Model (DIIM) with a focus on the syntactic and semantic interoperability issues to enable and enhance the interoperability of IoT data.
3. Defined a real-world use case of water quality monitoring in the water distribution system with the evaluation of the DIIM.
4. Implement a showcase of DIIM in Python by Natural language processing techniques to find similarities between IoT and domain-specific ontologies.
Work package (WP3) –ESR3 (BCU) developed an architecture using data intelligence for Cyber-physical systems (CPS) in Smart Water Networks (SWN) to improve the security and resilience of water networks.WP3 investigated potential new data intelligent techniques to preserve the security and resilience of CPS in SWN. Attack detection and data validation model and toolkits were developed to detect and mitigate Cyber-Physical attacks in SWN using ML and blockchain. WP3 generated 5 research papers, 4 technical reports, and a Ph.D. thesis. ESR3 successfully obtained his Ph.D. degree in May 2022 at BCU. WP3 main achievements:
1. Reviewed the previous CPS attacks in SWN, their consequences, and learned lessons
2. Developed an attack detection simulation model on MATLAB using machine learning algorithms
3. Developed data validation simulation model in SWN considering the security requirements using consensus algorithms of blockchain technology
4. Developed an integrated attack detection simulation model using hybrid machine-learning and blockchain technology
IOT4Win delivered all S&T WPs; the research outcomes and achievements generated two Ph.D. completions. The research outputs have been published in scientific journals and conferences and published in open access. All ERSs have seconded with industrial partners and academia onsite and virtually. Four research training workshops for ESRs have been delivered by the IOT4Win consortium and other dissemination events were organized for industrial engagement and outreach.