Periodic Reporting for period 2 - RAIS (RAIS: Real-time Analytics for the Internet of Sports)
Berichtszeitraum: 2021-01-01 bis 2023-06-30
New emerging sensing technologies of the Internet of Things (IoT) in our global interconnected world present us with unprecedented amounts of data, available at highly heterogeneous and distributed data sources. IoT is a greenfield market. New players, with new business models, approaches, and solutions, appear and overtake incumbents. Existing IoT solutions are expensive because of the high infrastructure and maintenance costs associated with centralized clouds, large server farms and networking equipment. The sheer amount of communication that will have to be handled when IoT devices grow to tens of billions will increase those costs substantially. Even if the unprecedented economical and engineering challenges are overcome, cloud servers will remain a bottleneck and point of failure that can disrupt the entire network. This makes Privacy-Preserving Decentralized Big Data Analytics a main challenge for the success of future IoT initiatives that capitalize on mining data coming from personal devices by uncovering
the collective behaviour and latent phenomenon arising in such systems.
The main objectives of the RAIS project were to lay the groundwork and develop essential key technologies for such Privacy-Preserving Decentralized Big Data Analytics in future large-scale distributed IoT systems, as well as provide comprehensive training to a new generation of 14 early-stage researchers (ESRs) in key technical subjects of Big Data Analytics on the Edge, Data Stream Processing, Distributed & Decentralized Machine Learning, and Security & Privacy. The project specifically focused on the Internet of Sports (IoS) as an application domain heavily impacted by IoT.
The project successfully reached all the targeted research and ESR training objectives.
the collective behaviour and latent phenomenon arising in such systems.
The main objectives of the RAIS project were to lay the groundwork and develop essential key technologies for such Privacy-Preserving Decentralized Big Data Analytics in future large-scale distributed IoT systems, as well as provide comprehensive training to a new generation of 14 early-stage researchers (ESRs) in key technical subjects of Big Data Analytics on the Edge, Data Stream Processing, Distributed & Decentralized Machine Learning, and Security & Privacy. The project specifically focused on the Internet of Sports (IoS) as an application domain heavily impacted by IoT.
The project successfully reached all the targeted research and ESR training objectives.
RAIS consortium conducted research through four technical workpackages: WP1 Distributed Sensing Infrastructure & Networking, WP2 Security, Privacy and Trust, WP3 Data Mining & Edge Analytics and WP4 Predictive Analytics for IoS.
In particular, WP1 focused on designing the main components for a novel overlay network infrastructure comprising peers that provide the run-time environment, the platform, and the communication functionality required for basic collective sensing services and for deploying new applications. Four RAIS fellows made contributions on the tasks of WP1 and produced 4 peer-reviewed publications.
WP2 focused on designing a suite of security and user-centric privacy-preserving mechanisms for securing IoT-generated data against some of the most challenging threats while, at the same time, keeping the utility of the data for the intended consumer. The fellows designed services both for standard centralized architectures as well as for more challenging distributed use cases. Blockchain has been used as an underlying trust framework to enable accountability and transparency among the collaborating parties. Four RAIS fellows made contributions on the tasks and produced 18 peer-reviewed publications.
WP3 focused on developing novel ML algorithms for Human Activity Recognition (HAR), designing of novel and lightweight Graph Representation Learning (GRL) algorithms for linked data as well as building Federated and Decentralized ML Solutions for massively distributed data. Several ESRs contributed to the workplackage tasks and produced 18 peer-reviewed publications.
WP4 focused on leveraging data analytics, machine learning, and behavioral science to advance understanding of user exercise habits, social contagion, anomaly detection, well-being, and sustained engagement. The resulting technologies directly contribute to practical and responsible applications and interventions that promote healthier lifestyles, enhance performance, and improve overall well-being. Four fellows were involved on the tasks and produced 17 peer-reviewed publications.
The RAIS fellows underwent extensive "hands-on" research training and gained valuable exposure to non-academic environments through industrial secondments. Overall, the project facilitated a wide range of network-wide events, including 17 Interactive Online Seminars, 3 summer schools, 1 entrepreneurship event, 1 hackathon, 2 workshops, 1 conference and 21 secondments, fostering teamwork among the fellows while also supporting their individual development as experts.
In particular, WP1 focused on designing the main components for a novel overlay network infrastructure comprising peers that provide the run-time environment, the platform, and the communication functionality required for basic collective sensing services and for deploying new applications. Four RAIS fellows made contributions on the tasks of WP1 and produced 4 peer-reviewed publications.
WP2 focused on designing a suite of security and user-centric privacy-preserving mechanisms for securing IoT-generated data against some of the most challenging threats while, at the same time, keeping the utility of the data for the intended consumer. The fellows designed services both for standard centralized architectures as well as for more challenging distributed use cases. Blockchain has been used as an underlying trust framework to enable accountability and transparency among the collaborating parties. Four RAIS fellows made contributions on the tasks and produced 18 peer-reviewed publications.
WP3 focused on developing novel ML algorithms for Human Activity Recognition (HAR), designing of novel and lightweight Graph Representation Learning (GRL) algorithms for linked data as well as building Federated and Decentralized ML Solutions for massively distributed data. Several ESRs contributed to the workplackage tasks and produced 18 peer-reviewed publications.
WP4 focused on leveraging data analytics, machine learning, and behavioral science to advance understanding of user exercise habits, social contagion, anomaly detection, well-being, and sustained engagement. The resulting technologies directly contribute to practical and responsible applications and interventions that promote healthier lifestyles, enhance performance, and improve overall well-being. Four fellows were involved on the tasks and produced 17 peer-reviewed publications.
The RAIS fellows underwent extensive "hands-on" research training and gained valuable exposure to non-academic environments through industrial secondments. Overall, the project facilitated a wide range of network-wide events, including 17 Interactive Online Seminars, 3 summer schools, 1 entrepreneurship event, 1 hackathon, 2 workshops, 1 conference and 21 secondments, fostering teamwork among the fellows while also supporting their individual development as experts.
Within RAIS, the intersectoral collaboration between academia, entrepreneurs, business developers, and industry strengthened novelty and impact by ensuring that relevant needs have been addressed, and that research results are both economically and technically feasible.
One of the key outcomes of the the RAIS project was a novel decentralized data framework PDS2, which ensures that the users can maintain full control of their data and do not need to compromise their privacy, while being rewarded for the value that their data generates. In order to achieve this, the consortium had to develop a number of novel technologies within the areas of privacy-preserving computation, decentralized machine learning, blockchain etc., which were produced by all RAIS fellows in all technical workpackages, and published in prestigious venues. The project also produced and shared with the public LifeSnaps, a new public 4-month multi-modal dataset capturing unobtrusive snapshots of people's lives in the wild, that can enable future research in other disciplines and from different perspectives.
One of the key outcomes of the the RAIS project was a novel decentralized data framework PDS2, which ensures that the users can maintain full control of their data and do not need to compromise their privacy, while being rewarded for the value that their data generates. In order to achieve this, the consortium had to develop a number of novel technologies within the areas of privacy-preserving computation, decentralized machine learning, blockchain etc., which were produced by all RAIS fellows in all technical workpackages, and published in prestigious venues. The project also produced and shared with the public LifeSnaps, a new public 4-month multi-modal dataset capturing unobtrusive snapshots of people's lives in the wild, that can enable future research in other disciplines and from different perspectives.