#	Pagina
attuale pagina	/open-h2020/projects/218387/results.html

Report

Teaser, summary, work performed and final results

Periodic Reporting for period 1 - Fed4IoT (Federating IoT and cloud infrastructures to provide scalable and interoperable Smart Cities applications, by introducing novel IoT virtualization technologies)

Teaser

Summary

\"The deployment of pervasive and large-scale IoT systems and applications poses several problems, including interoperability and expenditure ones.

Interoperability problems arise because there is not, yet, a wide-scale consensus about which is the ultimate IoT standard for devices and/or data. Although there is an ongoing, valuable, global standardization effort, especially on the data access perspective, ranging from oneM2M, to FIWARE NGSIv2, to ETSI Context Information Management and NGSI-LD (just to name a few of the most relevant initiatives), many IoT systems use proprietary interfaces, still. Lack of interoperability leads to vendor lock-in, which, on the one hand, makes it costly for a service provider to switch between competing vendors and, on the other hand, limits the type (and potentially the number) of IoT devices that can be integrated in a production system to those provided by a specific vendor (or supporting a specific interface).

Expenditure problems arise because it is expensive to deploy and manage the high volume of sensors that is necessary to create pervasive and intelligent applications, such as smart city ones. The cloud computing paradigm gradually solved this issue for the generic ICT domain, by means of virtualization technologies for sharing computing, storage, and networking resources that are made available by the cloud provider. In the IoT domain, such a decoupling between infrastructure providers and infrastructure users doesn\'t exist, yet. An IoT service provider is usually also the owner of the supporting IoT infrastructure (i.e. the sensors, actuators, the dedicated wireless networks, the data brokers, etc...), leading to a so-called \"\"IoT silo\"\", just to emphasize the closed and application-specific behavior of the overall system. The deployment and management cost of a silo might discourage many service providers or private users even for small-scale applications, and large-scale systems would definitely be affordable to a small number of large corporations only. Fair competition is thus prevented and, even worse, the innovation pace, which is fast when thousands of small stakeholders take the field, is slowed down, instead.

The Fed4IoT project focuses strong solutions to the expenditure and interoperability issues, by introducing novel \"\"IoT virtualization and data sharing\"\" technologies.
The project is organized in 5 work packages (WPs):
- WP1: Project Management, Dissemination and Exploitation
- WP2: Fed4IoT System
- WP3: Virtual IoT infrastructures
- WP4: IoT smart-city information/knowledge sharing
- WP5: Integration, deployment, validation and pilots
\"

Work performed

\"During the first year (Y1) of activity the project team has designed the Fed4IoT system architecture for IoT virtualization, shown in Fig. 1. This design leads to a proposed solution, which can be demonstrated in the 5 use cases of the project (Fig.2).
A summary of the work done in Y1 follows:

- Design and development of an initial proof-of-concept prototype of an IoT Virtualization Platform (VirIoT) show in Fig. 1, which enables users/tenants to set up their \"\"Virtual Silos\"\" formed by \"\"Virtual Things\"\" plus configurable IoT brokers (e.g. oneM2M Mobius, Orion NGSI v2, etc.) for data access. A Virtual Silo can also be chained to upstream IoT platforms, such as NodeRED or those offered by cloud providers (e.g. Azure IoT, AWS IoT, etc.).
- Development and research on specific components of the architecture including: i) standard data formats and API (NGSI-LD) used for internal/external data access, ii) technologies for simplifying and optimizing the design of \"\"ThingVisor\"\" components that generate Virtual Things\' data, by exploiting edge computing (e.g. FogFlow) and Information Centric Networking, iii) light virtualization/containerization technologies (UniKraft) for fast deployment and configuration through container orchestration systems such as Kubernetes.
- Design of use cases for testing both the architecture and the applications, including initial hardware deployment on pilot sites.
- Contribution to ITU and ETSI standardization activities and initial contacts with AIOTI and FIWARE groups.
- Dissemination activities in several IoT events (e.g. Global IoT Summit, SIDO, etc.) and 24 scientific publications.
- Identification of preliminary exploitation assets and plans.
\"

Final results

The main ambition of Fed4IoT is to devise a platform for IoT virtualization offering to users different and isolated IoT systems (Virtual Silos) formed by on-demand Virtual Things. Virtual Things are generated by the integration and processing of data coming from a federated set of real things.

While off-the-shelf IoT cloud platform mainly offer cloud services to IoT devices of customers, Fed4IoT is instead focused on offering Things-as-a-Service, by acquiring (control of) an ever-growing number of devices out there in the field, and by visualizing them to supply a scalable layer of horizontally share-able IoT resources to customers. Moreover, the Virtual Things rented by a customer can be, in turn, connected to upstream cloud service platforms as if they were real, un-shared, IoT devices. In this sense, the Fed4IoT services are complementary to most of the existing solutions and can interoperate with them in an extended IoT chain.

The project will show the effectiveness of the proposed solutions through a set of five concrete smart-city use-cases (Fig. 2) based on virtual things obtained by the sharing of a common set of real IoT devices.

Potential impacts includes:
1) Credible demonstrations based on cross-border business and/or societal applications of robust interoperable technologies identifying policy/legal obstacles (i.e., free flow of data, data protection, data portability etc.)
2) Concrete implementations of interoperable solutions that integrate IoT, Cloud and Big Data including security that are candidates for standardisation.
3) Facilitation of the development of cloud-enabled, secure and trustworthy IoT/big data applications (i.e., integrating intelligent security systems and virtualization technologies and devices/interfaces).
4) Promotion of the use of data related to Smart Cities and the creation of new increasingly efficient services in urban and regional administrative management.
5) Joint contributions to standardization activities (ETSI, ITU) under the cooperation of EU-Japan research institutes and IoT-related consortia (e.g. AIOTI and IoT Acceleration Consortium).