Author: Guido Dartmann
Central research topics of the research group VSKI are the investigation of distributed algorithms, protocols, distributed software systems for the utilization and resource-efficient processing of the large amounts of information, e.g., of the Internet of Things. An important requirement of research in this area is a resource-efficient implementation of distributed systems and artificial intelligence (AI) methods. In addition, AI methods that make diverse application areas and systems more resource-efficient are to be researched. Due to the diversity of applications, the research topics of the WG are broad and practical: Algorithms for distributed infomation systems, federated learning for cyber-physical systems (CPS), efficient architectures for edge computing, architectures and concepts for big data analysis in the Internet of Things. The team is interdisciplinary in nature and bridges with other research groups and departments. The technologies developed and researched in the team are particularly related to current application areas (sustainability, resource efficiency, industry, logistics, economics, health, and natural sciences).
A current challenge is the distributed and at the same time efficient implementation (in hardware and software) of complex applications, in which e.g. algorithms of multimodal sensor signal processing and machine learning for sensor data streams in the edge domain have to be implemented in an energy-efficient way. Current and future applications consist of heterogeneous IoT end nodes that need to generate and interpret high-dimensional data sets. Such systems are increasingly organized in a decentralized manner and require concepts for distributed algorithms with resource-efficient local data preprocessing (edge computing). For efficient processing of data streams in such systems, we explore distributed information processing methods and their decentralized implementation in protocols, software and hardware.
Due to the major challenges regarding the resource efficiency of future systems, the team has to address a large breadth of the subject. In the future, this will enable the addressing of many tenders for sovereign third-party funded projects as well as the understanding of the second important task of technology transfer. The upper figure shows the structure of the working group. The team's research topics should be application-driven and have links to other specialist groups. In particular, cooperation in sovereign joint projects enables the team to integrate relevant practical experience from industry in order to solve relevant problems. From the interaction between fundamentals and application, three central research areas are defined for the research group, which are shown in the figure.
Research area 1 - Distributed Systems:
Here we explore concepts of (decentralized) implementation of CPS. Relevant research areas are protocols for distributed systems, smart contracts and IT security through zero trust applications.
An important result of this research is the implementation of energy-efficient smart contract systems for logistics, mobility and IoT. For this purpose, the research group developed the concept of Cypher Social Contracts, which is used within the open source application Fides (https://gitlab.rlp.net/l.creutz/fides) and has been made available to the community in several publications:
- L. Creutz and G. Dartmann, "Cypher Social Contracts A Novel Protocol Specification for Cyber Physical Smart Contracts," 2020 International Conferences on Internet of Things (iThings) and IEEE Green Computing and Communications (GreenCom) and IEEE Cyber, Physical and Social Computing (CPSCom) and IEEE Smart Data (SmartData) and IEEE Congress on Cybermatics (Cybermatics), 2020, pp. 440-447, doi: 10.1109/iThings-GreenCom-CPSCom-SmartData-Cybermatics50389.2020.00083.
- L. Creutz, J. Schneider and G. Dartmann, "Fides: distributed cyber-physical contracts," 2021 Third IEEE International Conference on Trust, Privacy and Security in Intelligent Systems and Applications (TPS-ISA), 2021, pp. 51-60, doi: 10.1109/TPSISA52974.2021.00006.
- L. Creutz, K. Wagner and G. Dartmann, "Cyber-Physical Contracts in Offline Regions," 2022 IEEE International Conferences on Internet of Things (iThings) and IEEE Green Computing & Communications (GreenCom) and IEEE Cyber, Physical & Social Computing (CPSCom) and IEEE Smart Data (SmartData) and IEEE Congress on Cybermatics (Cybermatics), 2022, pp. 461-469, doi: 10.1109/iThings-GreenCom-CPSCom-SmartData-Cybermatics55523.2022.00097.
The research was and is supported by the third-party funded projects:
- LandLights (BMVD-funded): Core topics are smart contracts and smart logistics.
- APEROL (BMVD-funded): Core topics are research demonstrators for smart logistics
Research Field 2 - Artificial Intelligence
This area addresses algorithmic methods for decentralized processing of information from IoT systems as well as algorithms for edge computing. A particular focus here is the exploration of federated learning methods for energy-efficient IoT end nodes. In addition, we have been researching data analysis methods in the application area of medicine. Several relevant research papers were produced here:
- Peine, A. Hallawa, J. Bickenbach, G. Dartmann, L. B. Fazlic, A. Schmeink, G. Ascheid, C. Thiemermann, A. Schuppert, R. Kindle, L. Celi, G. Marx, and L. Martin. Development and validation of a reinforcement learning algorithm to dynamically optimize mechanical ventilation in critical care. www.nature.com/articles/s41746-021-00388-6.pdf npj Digit. Med, 2021
- L. Begic Fazlic, A. Halawa, A. Schmeink, R. Lipp, L. Martin, A. Peine, M. Morgen, T. Vollmer, S. Winter, and G. Dartmann. A novel hybrid methodology for anomaly detection in time series. International Journal of Computational Intelligence Systems, 15(1):1-16, 2022.
The research work was and is supported by the third-party funded projects:
- COSY (BMBF-funded): Demonstrators for IoT Data Analysis
- IMEDALytics (BMBF-funded): Decision support systems for medicine
- Claire (EIT-funded): Benchmarking for AI algorithms
- AI Pilot (BMEL-funded): Distributed and incremental learning
- SAVE (BMWK-funded): Semantic learning from heterogeneous data streams
Research Field 3 - Engineering:
Here we address the resource-efficient implementation of CPS in software and hardware and in particular the development of research demonstrators. In the future, circuit boards for possible implementations of intelligent IoT end nodes with resource-efficient data preprocessing equipped with intuitive development tools will also be created here.
- G. Dartmann (Ed.), H. Song (Ed.) A. Schmeink (Ed.). Big Data Analytics for Cyber-Physical Systems. 2019. paperback ISBN: 9780128166376
- M. Hauck, R. Machhamer, L. Czenkusch, K. Gollmer, and G. Dartmann. Node and block-based development tools for distributed systems with AI applications. IEEE Access, 7:143109-143119, 2019.
The research work was and is supported by the third-party funded projects:
- IoT Pilot (BMEL-funded): Development tools for IoT systems
- AI-Pilot (BMEL-funded): Software and hardware for edge AI, AI demonstrators.
- PINOT (BMEL-funded): Laboratory demonstrator for an electronic nose
- AI-Map (BMWK-funded): Automation of machine learning for metal oxide sensors.
More details of my research group at University of Applied Sciences Trier.