Periodic Reporting for period 2 - IoT-NGIN (Next Generation IoT as part of Next Generation Internet)
Período documentado: 2022-04-01 hasta 2023-09-30
Internet of Things (IoT) is one of the next big concepts to support societal changes and economic growth, being one of the fastest growing ICT segments. A specific challenge is to leverage existing technology strengths to develop solutions that sustain the European industry and values.
To address this, IoT-NGIN introduces novel research and innovation concepts, to establish itself as the “IoT Engine” that will fuel the Next Generation of IoT as a part of the European Next Generation Internet. First, IoT-NGIN uncovers a patterns-based meta-architecture that encompasses evolving, legacy, and future IoT architectures. In addition, it optimizes IoT/M2M and 5G/MCM communications, including secure-by-design micro-services to extend the edge cloud paradigm. Furthermore, it enables user and self-aware, autonomous IoT systems through privacy-preserving federated ML and ambient intelligence, with Augmented Reality (AR) support for humans. Finally, IoT-NGIN researches towards distributed IoT cybersecurity and privacy, for example, using Self-Sovereign Identities and interconnected Distributed Ledger Technologies (DLTs) to implement Meta-Level Digital Twins.
IoT-NGIN will bring significant benefits to the European Digital Society by contributing to the long-term evolution of next-generation IoT infrastructures and service platforms via a set of innovative and secure building blocks, which will increase the safety and wellbeing of the European citizens, reduce the environmental impact and accelerate safety at the industry space. Moreover, IoT-NGIN will support start-up and SMEs to build innovative sustainable new business via 2 open calls, releasing cascading funding opportunities.
To address this, IoT-NGIN introduces novel research and innovation concepts, to establish itself as the “IoT Engine” that will fuel the Next Generation of IoT as a part of the European Next Generation Internet. First, IoT-NGIN uncovers a patterns-based meta-architecture that encompasses evolving, legacy, and future IoT architectures. In addition, it optimizes IoT/M2M and 5G/MCM communications, including secure-by-design micro-services to extend the edge cloud paradigm. Furthermore, it enables user and self-aware, autonomous IoT systems through privacy-preserving federated ML and ambient intelligence, with Augmented Reality (AR) support for humans. Finally, IoT-NGIN researches towards distributed IoT cybersecurity and privacy, for example, using Self-Sovereign Identities and interconnected Distributed Ledger Technologies (DLTs) to implement Meta-Level Digital Twins.
IoT-NGIN will bring significant benefits to the European Digital Society by contributing to the long-term evolution of next-generation IoT infrastructures and service platforms via a set of innovative and secure building blocks, which will increase the safety and wellbeing of the European citizens, reduce the environmental impact and accelerate safety at the industry space. Moreover, IoT-NGIN will support start-up and SMEs to build innovative sustainable new business via 2 open calls, releasing cascading funding opportunities.
Aiming to support a wide range of architecture designs in the next generation of IoT, the IoT-NGIN meta-architecture has been defined, analyzing more than 100 state-of-the-art IoT platforms and frameworks.
IoT-NGIN has provided 5G enhancements via research on relay selection strategies to support 5G network coverage extension via Device-to-Device (D2D) communications, a prototype 5G Core with Time Sensitive Networking (TSN), as well as implementation of the Network Slice Management System (NSMS) following the 3GPP specifications [TS 28.531]. 5G capabilities exposure APIs for facilitating management of 5G services, network and device, as well as a secure edge cloud framework for execution of microservices at the edge have been developed.
Moreover, IoT-NGIN developed a hybrid edge-cloud MLaaS (Machine Learning as a Service) framework, in line with BDVA SRIA4.0. The platform covers the whole ML lifecycle, supporting ML training via privacy-preserving Federated Learning, XAI online learning and reinforcement learning and polyglot model sharing.
Furthermore, IoT-NGIN provides enhancements towards Human-Centric IoT through Ambient Intelligence and Tactile Internet focused on enhanced object recognition, based on both visual and non-visual methods, FIWARE-based Digital Twinning of IoT devices and pervasive security via a flexible identity management and access control service, as well as innovative interfaces via Augmented Reality (AR) tools.
Contributions to IoT cybersecurity include novel techniques for identification of common vulnerabilities in IoT devices, GAN and ML based attack detection in FL systems and the Moving Target Defense (MTD) Honeypots framework.
Data sovereignty is addressed through the Decentralized Interledger Bridge (DIB), as well as contributions to Self-Sovereign Identity (SSI) technologies, such as Decentralised Identifiers (DIDs) and Verifiable Credentials (VCs). The novel concept of DLT-enabled Meta-Level Digital Twin (MLDT) has been realized as Semantic Twin (ST), which provides the description of Digital Twin and device data and metadata.
Pilot activities in each Living Lab, through the execution of 10 use cases by the project and 10 additional cases by Open Call projects, have validated the IoT-NGIN value in 4 application verticals, namely Smart City, Smart Agriculture, Industry 4.0 and Smart Energy.
Last, but not least, the project has already raised significant impact through 3 contributions in 5G SA6 accepted as standards, the foundation of Twinbase startup exploiting IoT-NGIN outcomes, as well as scientific contributions in notable journals and events.
IoT-NGIN has provided 5G enhancements via research on relay selection strategies to support 5G network coverage extension via Device-to-Device (D2D) communications, a prototype 5G Core with Time Sensitive Networking (TSN), as well as implementation of the Network Slice Management System (NSMS) following the 3GPP specifications [TS 28.531]. 5G capabilities exposure APIs for facilitating management of 5G services, network and device, as well as a secure edge cloud framework for execution of microservices at the edge have been developed.
Moreover, IoT-NGIN developed a hybrid edge-cloud MLaaS (Machine Learning as a Service) framework, in line with BDVA SRIA4.0. The platform covers the whole ML lifecycle, supporting ML training via privacy-preserving Federated Learning, XAI online learning and reinforcement learning and polyglot model sharing.
Furthermore, IoT-NGIN provides enhancements towards Human-Centric IoT through Ambient Intelligence and Tactile Internet focused on enhanced object recognition, based on both visual and non-visual methods, FIWARE-based Digital Twinning of IoT devices and pervasive security via a flexible identity management and access control service, as well as innovative interfaces via Augmented Reality (AR) tools.
Contributions to IoT cybersecurity include novel techniques for identification of common vulnerabilities in IoT devices, GAN and ML based attack detection in FL systems and the Moving Target Defense (MTD) Honeypots framework.
Data sovereignty is addressed through the Decentralized Interledger Bridge (DIB), as well as contributions to Self-Sovereign Identity (SSI) technologies, such as Decentralised Identifiers (DIDs) and Verifiable Credentials (VCs). The novel concept of DLT-enabled Meta-Level Digital Twin (MLDT) has been realized as Semantic Twin (ST), which provides the description of Digital Twin and device data and metadata.
Pilot activities in each Living Lab, through the execution of 10 use cases by the project and 10 additional cases by Open Call projects, have validated the IoT-NGIN value in 4 application verticals, namely Smart City, Smart Agriculture, Industry 4.0 and Smart Energy.
Last, but not least, the project has already raised significant impact through 3 contributions in 5G SA6 accepted as standards, the foundation of Twinbase startup exploiting IoT-NGIN outcomes, as well as scientific contributions in notable journals and events.
IoT-NGIN is likely to bring significant innovations and advance the state of the art in the development, deployment and operation of IoT ecosystems.
The project has provided significant contributions to enhancing the 5G communication, an essential factor in fulfilling the next generation IoT requirements for high-speed, low-latency, high-bandwidth and increased security and privacy communications.
With intelligence and AI being at the centre of next generation systems, IoT-NGIN innovates with the MLaaS framework with significant research contributions with the novel Online Learning framework, the Reinforcement Learning framework and the privacy-preserving Federated Learning techniques applicable to multi-node real settings, as well as the polyglot model sharing, while complying with BDVA SRIA.
Furthermore, IoT-NGIN has provided contributions towards ambient intelligence-based IoT systems with enhanced device discovery methods, flexible Digital Twin services, pervasive security with privileged access control and AR tools in innovative applications.
As cybersecurity evolves also in the IoT domain, IoT-NGIN provides both passive and active protection against cyberattacks in federated learning and IoT systems, in line with ENISA’s threat modelling methodology for AI.
Moreover, the novel DIB solution supports high-throughput atomic transactions across different types of DLTs/blockchains. The Semantic Twin solution, integrating SSI and DID, supports secure and auditable data access across IoT devices, Digital and Semantic Twins.
IoT-NGIN is thus likely to accelerate the adoption of use of edge computing, supporting shifting significant amount of data persistence and computation at the edge. Moreover, the project developments provide the basis for the transition to swarm computation in the longer term.
The research and innovation outcomes are fully aligned with the Digital Europe objectives, especially in the context of AI (aligning to the AI Act), both providing advanced ML tools and methods and gracefully integrating them with both current state of the art and new solutions. Also, the project aligns to the Digital Europe requirements for increased cybersecurity and data privacy, considering and further protecting current and future IoT systems. IoT-NGIN provides advancements towards Connectivity & Infrastructure, through its contributions in the 5G domain. The project aligns to the objectives of the European Green deal, supporting development of resource-efficient applications, at high level of automation and autonomy, which leads to reduced consumption of energy or water resources.
Last, but not least, IoT-NGIN benefits the European Society, supporting its digital transformation via its open-source tools for building the next generation IoT, as well as supporting the development of intelligent, secure, privacy-aware, fast and reliable IoT systems and applications, which will ease and enhance the work and life of European citizens.
The project has provided significant contributions to enhancing the 5G communication, an essential factor in fulfilling the next generation IoT requirements for high-speed, low-latency, high-bandwidth and increased security and privacy communications.
With intelligence and AI being at the centre of next generation systems, IoT-NGIN innovates with the MLaaS framework with significant research contributions with the novel Online Learning framework, the Reinforcement Learning framework and the privacy-preserving Federated Learning techniques applicable to multi-node real settings, as well as the polyglot model sharing, while complying with BDVA SRIA.
Furthermore, IoT-NGIN has provided contributions towards ambient intelligence-based IoT systems with enhanced device discovery methods, flexible Digital Twin services, pervasive security with privileged access control and AR tools in innovative applications.
As cybersecurity evolves also in the IoT domain, IoT-NGIN provides both passive and active protection against cyberattacks in federated learning and IoT systems, in line with ENISA’s threat modelling methodology for AI.
Moreover, the novel DIB solution supports high-throughput atomic transactions across different types of DLTs/blockchains. The Semantic Twin solution, integrating SSI and DID, supports secure and auditable data access across IoT devices, Digital and Semantic Twins.
IoT-NGIN is thus likely to accelerate the adoption of use of edge computing, supporting shifting significant amount of data persistence and computation at the edge. Moreover, the project developments provide the basis for the transition to swarm computation in the longer term.
The research and innovation outcomes are fully aligned with the Digital Europe objectives, especially in the context of AI (aligning to the AI Act), both providing advanced ML tools and methods and gracefully integrating them with both current state of the art and new solutions. Also, the project aligns to the Digital Europe requirements for increased cybersecurity and data privacy, considering and further protecting current and future IoT systems. IoT-NGIN provides advancements towards Connectivity & Infrastructure, through its contributions in the 5G domain. The project aligns to the objectives of the European Green deal, supporting development of resource-efficient applications, at high level of automation and autonomy, which leads to reduced consumption of energy or water resources.
Last, but not least, IoT-NGIN benefits the European Society, supporting its digital transformation via its open-source tools for building the next generation IoT, as well as supporting the development of intelligent, secure, privacy-aware, fast and reliable IoT systems and applications, which will ease and enhance the work and life of European citizens.