Building the Internet of Things
Guest Editors’ Introduction • Vincenzo Piuri and Roberto Minerva • July 2015
Translations by Osvaldo Perez and Tiejun Huang
Listen to the Guest Editors’ Introduction
The definition of the Internet of Things (IoT) is somehow elusive because it refers to so many disciplines, technologies, and application domains. Essentially, the IoT envisions systems of networked sensors and smart objects that work together to make an environment intelligent, usable, and programmable.
IoT technology encompasses sensors, circuits, embedded systems, communications, intelligent interfaces, energy management, data management, data fusion, knowledge management, real-time systems, distributed processing, system design, and sophisticated software techniques that deal with big data. The IoT also ranges from the industrial Internet, which could profoundly change how goods are produced, to e-health, which offers new ways of caring for humans and treating diseases.
IoT systems cover a broad spectrum of architectures, ranging from small structures with few sensors and aggregators in a limited area that provide services and applications to a handful of people, to pervasive systems involving millions of sensors, complex distributed information processing, and actuators setting the framework of a smart city.
IoT systems’ administrative boundaries will determine striking differences: a home monitor system will be controlled and managed differently than logistic systems that must integrate functions and services offered by various administrative domains. IoT systems also differ from a business perspective: sensing wellness devices differ dramatically from enterprise security and monitoring systems. IoT systems and solutions thus face increasingly larger numbers of sensors, devices, administrative domains, protocols, and interfaces.
In This Issue
Providing an integrated view is difficult because the IoT could touch just about everything. In “Enabling the Internet of Things,” Roy Want, Bill Schilit, and Scott Jenson provide a key to uniformly represent real-world resources and objects and interact with them through Web-derived technologies. They show how the IoT could allow mobile devices to monitor and control just about anything.
Irene Bojanova, George Hurlburt, and Jeffrey Voas take this to the next level with “Imagineering an Internet of Anything.” They lead us to a near future when all objects will be connected and programmable through a wealth of applications. To reach this next stage will require the capability of mitigating technological differences and integrating functions and data. The IoT will also require the ability to move data and to integrate it for coordinated processing. In the video “About IPV6 and Its Importance for IoT,” Antonio Skarmet, TPC Chairman of the IEEE World Forum on IoT conference, presents how IPv6 could be instrumental in addressing and routing information among IoT objects. Such IP communication will help enable the desired integration and broad span of IoT solutions.
Current trends toward aggregating computing resources and functionalities in datacenters could be a cornerstone for implementing large IoT solutions. “Elastic Infrastructure to Support Computing Clouds for Large-Scale Cyber-physical Systems,” by Douglas Schmidt, Jules White, and Christopher Gill, discusses possible solutions for the IoT that exploit the cloud’s elasticity and the flexibility to cope with specific requirements such as aggregating processing power in different geographical areas and dealing with the dynamic demand of resources.
Current trends and studies related to virtualization could support the quest for integration and abstraction. Sanjay Madria, Vimal Kumar, and Rashmi Dalvi implement virtual sensors as the core of a sensor-cloud architecture in “Sensor Cloud: A Cloud of Virtual Sensors.” These virtual sensors could enable and help create a multiuser environment on top of resource-constrained physical wireless sensors and help support multiple applications. Virtual sensors could be reused in different contexts with different properties to accommodate heterogeneous applications’ needs.
A massive number of devices characterizes large IoT systems, including potentially unstable wireless devices operating in areas without comprehensive coverage. Competing with these and other issues will require flexible management and operating platforms. Traditional mechanisms won’t be able to handle the forecasted number of sensors and devices; their sheer quantity will require autonomous configuration and system upgrades implemented in flexible middleware platforms. In “Distributed and Managed: Research Challenges and Opportunities of the Next Generation Cyber-physical Systems,” Gabor Karsai and his colleagues present these and other challenges that the distribution of resources pose, and they propose a middleware architecture to cope with them.
The IoT is also strongly related to the capabilities of measuring the physical world. Many measures will be needed to control phenomena in a smart city. These could refer to both physical objects and humans using devices and sensors. Because IoT systems will collect large amounts of data and the analysis of such data could lead to identifying patterns of behavior or even specific people, the IoT also presents privacy and security challenges. In “Big Data Privacy in the Internet of Things Era,” Charith Perera and his colleagues place the use of big data techniques derived from IoT systems into a user-centered perspective while presenting current challenges.
In the video “The Disappearing of IoT” Antonio Manzalini, chairman of the IEEE SDN Initiative, discusses how “softwarization” is going to transform the Telecommunications infrastructure into a “continuum” of logical resources, extended from the end-Users’ terminals, through the Network, up to the Data Centers. Not only the border between the network and the Cloud will disappear, but also terminals, intelligent machines, smart things, robots, etc. will become like edge nodes capable of storing data locally and even executing network functionality and service components. Eventually this “continuum” will absorb also the Internet of Things, and it will become so embodied into reality to disappear.
Software and middleware platforms will play key roles in enabling the IoT promise. Software, APIs, and development environments will also prove indispensable for coping with integration challenges. The ICT industry is moving quickly into this domain with plenty of projects developing potential platforms (for instance, see the European Community’s attempt to cluster its IoT efforts, http://www.internet-of-things-research.eu). In addition, Web companies such as Google (with Brillo), hardware vendors such as ARM, and Intel, telecommunications vendors such as Huwei, and device manufacturers such as Samsung are demonstrating that they understand middleware’s importance.
Still, major issues remain, such as the heterogeneity of devices, protocols, and networks; different APIs and control interfaces; management functionalities; the service life cycle of “things;” various communication paradigms; and fair mechanisms for data acquisition. Besides the challenge of the middleware platform, other challenges include potentially invasive user-interaction mechanisms, integration with the apps ecosystem, and security threats. Finally, the IoT’s business dimension must be addressed in terms of new business models, a large and integrated ecosystem, and a revisited value chain.
V. Piuri and R. Minerva, “Building the Internet of Things,” Computing Now, vol. 8, no. 7, July 2015, IEEE Computer Society [online]; http://www.computer.org/web/computingnow/archive/building-the-internet-of-things-july-2015.
Vincenzo Piuri is a professor of computer engineering at Politecnico di Milano. His research interests include intelligent systems, machine learning, industrial applications, pattern analysis and recognition, intelligent measurement systems, the theory and industrial applications of neural networks, signal and image processing, biometrics, and digital- and signal-processing architectures. Piuri received a PhD in electronic engineering from Politecnico di Milano. He’s an IEEE Fellow, an ACM Distinguished Scientist, and a senior member of the International Neural Network Society. He’s the 2015 IEEE Vice President for Technical Activities and editor in chief of the IEEE Systems Journal. Contact him at firstname.lastname@example.org.
Roberto Minerva works in the Research Coordination group at Telecom Italia’s TILab, creating advanced scenarios derived from the application of emerging information and communications technologies with innovative business models, especially in the areas of the Internet of Things (IoT), distributed computing, programmable networks, personal data, self-organizing networks, software-defined networks, and network function virtualization. Minerva received a PhD in computer science and telecommunications from Telecom Sud Paris. He’s the chairman of the IEEE IoT Initiative. Contact him at email@example.com.