The “Internet of Things” (IoT) concept has been described by some as amorphous, blurred, clumsy, confusing, fuzzy, meaningless, nebulous, or vague . Others – such as the overwhelming majority of US consumers – never heard of it.
Yet, the growing number of articles, conferences, workshops, and reports on the topic around the world clearly indicates that something is happening, at least brewing, and invites us to notice and seek some kind of understanding. In short, the societal changes that IoT is on the cusp of bringing about are so broad and deep that we cannot ignore them.
My goal here is to contribute to the definitional debate with the intent to offer a clear view of what IoT is and is not.
IoT is a concept that has evolved over time
A complete IoT history would have to include early telemetry applications going back to the 19th century. Such a survey would also have to make room for wireless sensor networks first developed for military purposes in the 1950s, and other technologies the likes of packet-based, data-only networks such as Mobitex launched by Ericsson in Sweden in 1986 and particularly suited for machine-to-machine (M2M) applications. A brief overview of the Internet of Things is offered on the Postscapes website, which looks at IoT “from its first glimpses into today's trending topic”.
The “Internet of Things” term, which seems to have come to life in the Radio-Frequency Identification (RFID) space, has been catching on throughout the world to encapsulate a major change in the communications space. Over time, it has engulfed an increasingly broader range of technologies.
Note that in 1998, before IoT had a name, Roy Want, Mark Weiser, and Elizabeth Mynatt, distinguished researchers in the Ubiquitous Computing field working at the Xerox Palo Alto Research Center (PARC), had captured early on the essence of today’s IoT scope and transformational revolution in the title of a brief paper, “Activating Everyday Objects”. They presented the paper at the Defense Advanced Research Projects Agency (DARPA)/National Institute of Standards and Technology (NIST) Smart Spaces Workshop.
While they focused on the existing infrastructure and tools of everyday offices using technical approaches available then, their premise remains at the heart of present-day IoT: “The confluence of technological advances in wireless technology, mobile computing, novel displays, and sensors in addition to the decreasing cost of computing power provides the opportunity for utilizing computational capabilities in an increasing number of specialized, yet networked devices.” Their presentation, striking for its foresight, is still readily available on the Internet .
When the Internet began deploying its wings, i.e., when commercial Internet Service Providers (ISPs) started to emerge in the late 1980s, it was essentially used by people to communicate with other people either through exchanging information (documents) or communicating instantly with each other. This is undeniably the domain of the Internet of People.
Through converging and timely trends, it became possible to consider interconnecting the rest of the planet, i.e., anything and everything through the Internet (or more precisely the “Internet Protocol”). The word “things” in the term Internet of Things includes a wide range of categories such as products, equipment, digital constructs (e.g., pictures), animals, or even any element of the human body; essentially anything that is not a living human.
A seminal milestone took place on June 6, 2012 with the official arrival of the sixth version of the Internet Protocol (IPv6) which uses a 128-bit address, providing, therefore, approximately 340 trillion trillion trillion Internet addresses (let’s notice in passing that IPv6, developed by the Internet Engineering Task Force (IETF), embraces a lot more, and is not limited to a number of Internet addresses). The previous version (i.e., IPv4; IPv5 was never introduced for public use) used a 32-bit scheme generating “only” about 4.3 billion addresses. IPv6 is taking us to a universe heretofore out of reach.
As a result, the Internet of Things is about a communications space where “things” can be identified and reached the same way individuals can on the Internet of People (Figure below ). Any IoT solution always involves three foundational constituents: an endpoint (e.g., sensor or actuator), some kind of wireless and/or wired connection, and a software application, which interfaces with the endpoint.
However, the entire IoT value chain stretches far beyond the core foundation. It is increasingly complex as it extends to a variety of elements that are interwoven and that require various sets of skills and expertise. The IMAGE techno-management model (Figure below ) emphasizes the critical importance of the “gluing” elements, without which any IoT solution or IoT as a whole cannot work and thrive.
The IoT is comprised of two segments: the “Industrial Internet of Things” and the “Consumer Internet of Things.”
Click on image to enlarge.
The Industrial Internet of Things
The first and oldest one enables Machine to Machine communications or M2M, increasingly referred to as the “Industrial Internet of Things” (or Industrial IoT), that require little or no human interaction. Advanced manufacturing, factory automation, gas turbine and jet engine remote monitoring, precision agriculture, Supervisory Control and Data Acquisition (SCADA) systems, supply chain intelligence, and vehicle fleet tracking are a few examples of domains that rely on Industrial IoT technologies.
A detailed description of the Industrial Internet of Things at work at GE is presented by Jon Gertner in a June 18, 2014 Fast Company article on Behind GE’s Vision for the Industrial Internet of Things.
Incidentally, in the enterprise IoT space, not only is the purpose of those technologies intuitively understood, but “it is not too hard to see how value is created and captured” (see Ben Rooney’s article on Internet of Things Poses Big Questions in The Wall Street Journal – Tech Europe of July 8, 2013).The Consumer Internet of Things
The second and most recent(headline-grabbing) segment is about communications between things andpeople, a.k.a. the “Consumer IoT”, that necessitate human involvement.Evidently, both are often linked. In many cases, but not necessarilyalways, M2M (viz. the Industrial IoT) enables the Consumer IoT. It hasalso been argued that the reverse may also be true (see Jim Campbell’sarticle Consumer market will drive industrial Internet of things for manufacturing in Control Engineering of March 6, 2014).
Home,health, and fitness-focused applications such as baby and elderlymonitors, connected washers and dryers, smart door locks, smartgardening systems, smart watches, water monitors, pet collars and, moregenerally, the whole gamut of “wearables” are some examples of IoTdevices and services in the Consumer space (see the well-documentedNovember 2013 report on the Internet of Things by the Center for DataInnovation). Often, the smartphone serves as a “control tower” withinthe Consumer IoT.
While the Industrial IoT is alive and well,the Consumer IoT is in its infancy. It is not clear yet how it willevolve. For Jean-Louis Gassée, the former president of the AppleProducts Division, the Consumer IoT is just a “basket of remotes”, i.e.,”a metaphor for the unanswered management challenges in the Consumer IoT space “.
The two sides of IoT require different approaches
Emphasizingthe dichotomous nature of the Internet of Things (enterprise andconsumer) is not a futile exercise. Related technological and marketingrequirements as well as market challenges and issues differ profoundlyand must be thoroughly grasped (see this interesting O’Reilly Radararticle of February 25, 2014 by Varun Nagaraj, The Industrial IoT is not the same as the Consumer IoT).
Forexample, time synchronization and concurrency are critical technicaldimensions of the Industrial IoT. On the other hand, privacy, currentlyunder intense scrutiny in Europe and the United States, is a pivotalaspect of the Consumer IoT. These are not the only elements withdissimilar perspective or treatment. As the Internet of Things’disruption unfolds, new business models, educational programs, andprofessional training will be needed, which will not necessarily be thesame in the Business-to-Business (B2B) market and theBusiness-to-Consumer (B2C) market.
The Internet of Things technologies
Let’sbe clear, IoT as such is not an input but an output and is more of amarketing metaphor that highlights the massive economic and societaltransformation presently underway.
The Internet of Things, according to the definition of the International Telecommunication Union ,is “a global infrastructure for the information society.” It is in theprocess of being built and shaped by a host of enabling technologies:sensors and actuators, wireless and wired networks, high performancecomputing, energy harvesting and storage, encryption (cybersecurity),modeling & simulation, software analytics, etc.
Consequently,IoT is not a new science or technology, as it is sometimes erroneouslyportrayed, but, rather, an infrastructure architected and supported bywhat can rightfully be called “Internet of Things technologies.”
Among the IoT enabling ingredients, it makes sense to single out “Cyber Physical Systems” (or CPS), which are, according to the US National Science Foundation (NSF), “engineered systems that are built from, and depend upon, theseamless integration of computational algorithms and physicalcomponents.” It may be argued that CPS as a body of knowledgeconstitutes IoT’s scientific and technological foundation.
Therecently launched “Industry 4.0” project in Germany under the FederalGovernment’s High Tech strategy reflects the tight connection betweenthe Internet of Things and Cyber Physical Systems: “Driven by theInternet, the real and virtual worlds are growing closer and closertogether to form the Internet of Things… The Industry 4.0 project takes into account important aspects of location from atechnological, industrial and social perspective. Germany already plays aleading role in the field of (software-intensive) embedded systems,particularly in the car industry and engineering. So-called CyberPhysical Systems (CPS) are becoming increasingly important in thiscontext, that is to say the networking of embedded ICT systems both withone another and with the Internet. Along with increased automation inindustry, the development of intelligent monitoring and autonomousdecision-making processes is particularly important in order to be ablesteer and optimize both companies and entire value-adding networks inalmost real time.”
In sum, it is notinaccurate to equate the Internet of Things to a world of (more or less)automated connections and processes, where everything and everyone is apotential node, which is at the service of businesses, consumers andgovernments alike and aims, in the final analysis, at benefittinghumanity.
This is a seismic transformation since “things” onearth far outnumber humans. Because anything can be identified andreached at any level (including micro- and nanometer scale), it is notsurprising that the members of the IoT universe are thought to be in thetrillions.
Given the magnitude of this ongoing metamorphosis, itwill take time for it to effectively permeate all layers of the economyand society. This revolution in the making will not only rest onassociated progress in science, technology, and business practices, butalso on a legal, regulatory and policy environment that allows IoT toflourish.
Alain Louchez is the Managing Director of the Center for the Development and Application of Internet of Things Technologies (CDAIT) at the Georgia Institute of Technology (Georgia Tech). He chaired the Telecommunication Union (ITU) conference, on Internet of Things: Trends and challenges in standardization held in Geneva at the ITU headquarters in Geneva, Switzerland, onFebruary 18, 2014, and will be chairing the IoT standards session at theSecond Annual Internet of Things Global Summit in Washington, DC, USA, on October 27-28, 2014.