Interoperability and Open-Source Solutions for the Internet of Things

Inhaltsverzeichnis

1. Frontmatter

2. Semantic Interoperability

   1. Frontmatter

   2. Semantic Interoperability as Key to IoT Platform Federation

      Michael Jacoby, Aleksandar Antonić, Karl Kreiner, Roman Łapacz, Jasmin Pielorz

      Abstract

      Semantic interoperability is the key technology to enable evolution of the Internet of Things (IoT) from its current state of independent vertical IoT silos to interconnected IoT platform federations. This paper analyzes the possible solution space on how to achieve semantic interoperability and presents five possible approaches in detail together with a discussion on implementation issues. It presents the H2020 symbIoTe project as an example on how semantic interoperability can be achieved using semantic mapping and SPARQL query re-writing. We conclude that the found approaches together with the proposed technologies have the potential to act as corner stone technologies for achieving semantic interoperability.

   3. Overcoming the Heterogeneity in the Internet of Things for Smart Cities

      Aqeel Kazmi, Zeeshan Jan, Achille Zappa, Martin Serrano

      Abstract

      In the past few years, the viability of the Internet of Things (IoT) technology has been demonstrated, leading to increased possibilities for novel human-centric services in the smart cities. This development has resulted in numerous approaches being proposed for harnessing IoT for smart city applications. Having received a significant attention by the research community and industry, IoT adaptation has gained momentum. IoT-enabled applications are being rapidly developed in a number of domains such as energy management, waste management, traffic control, mobility, healthcare, ambient assisted living, etc. On the other hand, this high-speed development and adaptation has resulted in the emergence of heterogeneous IoT architectures, standards, middlewares, and applications. This heterogeneity is hindrance in the realization of a much anticipated IoT global eco-system. Hence, the heterogeneity (from hardware level to application level) is a critical issue that needs high-priority and must be resolved as early as possible. In this article, we present and discuss the modelling of heterogeneous IoT data streams in order to overcome the challenge of heterogeneity. The data model is used within the VITAL project which is an open source IoT system of systems. The main objective of the VITAL platform is to enable rapid development of cross-platform and cross-context IoT based applications for smart cities.

3. Interoperable Architectures and Platforms

   1. Frontmatter

   2. An Architecture for Interoperable IoT Ecosystems

      Stefan Schmid, Arne Bröring, Denis Kramer, Sebastian Käbisch, Achille Zappa, Martin Lorenz, Yong Wang, Andreas Rausch, Luca Gioppo

      Abstract

      The Internet of Things (IoT) is maturing and more and more IoT platforms that give access to things are emerging. However, the real potential of the IoT lies in growing IoT cross-domain ecosystems on top of these platforms that will deliver new, unanticipated value added applications and services. We identified two crucial aspects that are important to grow an IoT ecosystem: (i) interoperability to enable cross-platform and even cross-domain application developments on top of IoT platforms as well as (ii) marketplaces to share and monetize IoT resources. Having these two crucial pillars of an IoT ecosystem in mind, we present in this article the BIG IoT architecture as the foundation to establish IoT ecosystems. The architecture fulfills essential requirements that have been assessed among industry and research organizations as part of the BIG IoT project. We demonstrate a first proof-of-concept implementation in the context of an exemplary smart cities scenario.

   3. A Sensor Observation Service Extension for Internet of Things

      Argyrios Samourkasidis, Ioannis N. Athanasiadis

      Abstract

      This work contributes towards extending OGC Sensor Observation Service to become ready for Internet of Things, i.e. can be employed by devices with limited capabilities or opportunistic internet connection. We present an extension based on progressive data transmission, which by-design facilitates selective data harvesting and disruption-tolerant communication. The extension economizes resources, while respects the SOS specification requirement that the client should have no a-priori knowledge of the server capabilities. Empirical experiments in two case studies demonstrate that the extension adds little overhead and may lead to significant performance improvements in certain cases, as for irregular timeseries. Also, the proposed extension is not invasive and backwards compatible with legacy clients.

   4. Requirement-Based Deployment of Applications in Calvin

      Ola Angelsmark, Per Persson

      Abstract

      In order for IoT application developers to deliver on the promise of IoT, new tools and methodologies addressing the challenges associated with development of highly distributed systems running on non-reliable and heterogeneous hardware are required. Some of the main characteristics of cloud computing that has been a driving force for its success, are resource pooling, elasticity and the capacity for combining unrelated services. We believe that a similar approach is needed for IoT as well. In this paper, we show how Calvin, an open source peer-to-peer platform for distributed applications, tackles many of the problems inherent in IoT. By only loosely associating the functionality of a device with a semantics, and having implementations based on conventions rather than pre-defined terminology, it is possible to let a system of Calvin runtimes autonomously handle deployment decisions, and respond to changing requirements. We will discuss how to develop and deploy dynamic and adaptive IoT-applications based on capabilities and requirements, and how to resolve requirements by automatically combining information from multiple sources based on encapsulated domain knowledge.

4. Business Models and Security

   1. Frontmatter

   2. Business Models for Interoperable IoT Ecosystems

      Werner Schladofsky, Jelena Mitic, Alfred Paul Megner, Claudia Simonato, Luca Gioppo, Dimitris Leonardos, Arne Bröring

      Abstract

      The Internet of Things (IoT) is growing and more and more devices, so-called “things”, are being connected every day. IoT platforms provide access to those “things” and make them available for services and applications. Today, a broad range of such IoT platforms exist with differing functional foci, target domains, and interfaces. However, to fully exploit the economic impact of the IoT, it is essential to enable applications to interoperate with the various IoT platforms. The BIG IoT project aims at enabling this interoperability and supporting the creation of vibrant IoT ecosystems, which facilitate the development of cross-platform and cross-domain applications. While the value of interoperability for the overall economy is well understood and cannot be underestimated, some stakeholders may still need to find their business value in interoperable IoT ecosystems. Thus, this paper identifies the different stakeholders of such ecosystems, and analyses how these stakeholders can enhance their existing business models when taking part in an interoperable IoT ecosystem.

   3. On the Road to Secure and Privacy-Preserving IoT Ecosystems

      Juan Hernández-Serrano, Jose L. Muñoz, Arne Bröring, Oscar Esparza, Lars Mikkelsen, Wolfgang Schwarzott, Olga León, Jan Zibuschka

      Abstract

      The Internet of Things (IoT) is on the rise. Today, various IoT platforms are already available, giving access to myriads of things. Initiatives such as BIG IoT are bringing those IoT platforms together in order to form ecosystems. BIG IoT aims to facilitate cross-platform and cross-domain application developments and establish centralized marketplaces to allow resource monetization. This combination of multi-platform applications, heterogeneity of the IoT, as well as enabling marketing and accounting of resources results in crucial challenges for security and privacy. Hence, this article analyses the requirements for security in IoT ecosystems and outlines solutions followed in the BIG IoT project to tackle those challenges. Concrete analysis of an IoT use case covering aspects such as public, private transportation, and smart parking is also presented.

   4. Attribute-Based Access Control Scheme in Federated IoT Platforms

      Savio Sciancalepore, Michał Pilc, Svenja Schröder, Giuseppe Bianchi, Gennaro Boggia, Marek Pawłowski, Giuseppe Piro, Marcin Płóciennik, Hannes Weisgrab

      Abstract

      The Internet of Things (IoT) introduced the possibility to connect electronic things from everyday life to the Internet, while making them ubiquitously available. With advanced IoT services, based on a trusted federation among heterogeneous IoT platforms, new security problems (including authentication and authorization) emerge. This contribution aims at describing the main facets of the preliminary security architecture envisaged in the context of the symbIoTe project, recently launched by European Commission under the Horizon 2020 EU program. Our approach features distributed and decoupled mechanisms for authentication and authorization services in complex scenarios embracing heterogeneous and federated IoT platforms, by leveraging Attribute Based Access Control and token-based authorization techniques.

5. Platform Performance and Applications

   1. Frontmatter

   2. Data Ingestion and Storage Performance of IoT Platforms: Study of OpenIoT

      Alexey Medvedev, Alireza Hassani, Arkady Zaslavsky, Prem Prakash Jayaraman, Maria Indrawan-Santiago, Pari Delir Haghighi, Sea Ling

      Abstract

      Internet of Things is a very active research area with great commercialisation potential. The number of IoT platforms is already exceeding 300 and still growing. However, performance evaluation and benchmarking of IoT platforms are still in their infancy. As a step towards developing a performance benchmarking approach for IoT platforms, this paper analyses and compares a number of popular IoT platforms from data ingestion and storage capability perspectives. In order to test the proposed approach, we use the widely used open source IoT platform, OpenIoT. The results of the experiments and the lessons learnt are presented and discussed. While having a great research promise and pioneering contribution to semantic interoperability of IoT silos, the experimental results indicate OpenIoT platform needs more development effort to be ready for any substantial deployment in commercial IoT applications.

   3. Apps for Environments: Running Interoperable Apps in Smart Environments with the meSchup IoT Platform

      Thomas Kubitza

      Abstract

      With Apps a popular concept was introduced allowing end-users to easily extend their devices such as smartphones or computers with specific functionality. Two million Apps have ever since found their way into each of the popular App-stores Google Play and Apple Store. We argue that the App-concept is not only well applicable to single devices but also to complete environments equipped with smart networked things. In the moment when Apps can be easily downloaded and executed in home, office and industry environments a wide new applications space will be opened up. In this work we introduce the concept of Smart Space Apps that can be downloaded from a cloud-based App-store into a smart environment where they dynamically utilize the capabilities of available smart things to optimally achieve the purpose they were installed for. We introduce a unified schema for the access of sensors and actuators of heterogeneous devices from within Smart Space Apps and describe the middleware and runtime that implements this approach. We explain how Apps are packaged into an exchangeable format and published within a cloud-based App-store. Multiple application use cases are shown and challenges of this novel approach are discussed.

   4. Semantic Interoperability at Big-Data Scale with the open62541 OPC UA Implementation

      Julius Pfrommer

      Abstract

      The OPC Unified Architecture (OPC UA) is a protocol for Ethernet-based communication in industrial settings. At its core, OPC UA defines a set of services for interaction with a server-side information model that combines object-orientation with semantic technologies. Additional companion specifications use the OPC UA meta-model to define domain-specific modeling concepts for semantic interoperability. The open62541 project is an open source implementation of the OPC UA standard. In this work, we give a short introduction to the core concepts of OPC UA and how the measures taken to scale OPC UA to Big-Data scale reflect in the architecture of open62541.

6. Backmatter