SEPTEMBER 2018, VOL. 1, NO.

IEEE • Why Industry needs to AccelerAte Iot stAndArds

• drIllIng deep Into dIgItAl IndustrIAl trAnsformAtIon
WIll determIne Who survIves And thrIves
• connectIng the dots on Iot for the IndustrIAl World
Internet of Things Magazine • the future of Iot

A Publication of the IEEE Internet of Things

Initiative, a Multi-Society Technical Group
iot.ieee.org
 IoT and the Digital Revolution

2019 IEEE 5TH WORLD FORUM ON

THE INTERNET OF THINGS
15-18 April 2019 // Limerick, Ireland
Join us for the 2019 IEEE 5th World Forum on Internet of Things (WF-IoT 2019) in beautiful
Limerick, Ireland. The theme of the 4-day conference is “IoT and the Digital Revolution” in
recognition of strides and leadership that Ireland has shown in the deployment of “smart”
technologies, operating principles, and policies.

The program will feature the most outstanding participants from the research community,
public sector, and industry as well as a vast array of papers and presentations on the latest
technological innovations in the many fields and disciplines that drive the utility and vitality of
IoT solutions and applications.
TOPICAL TRACKS VERTICAL TRACKS
• Artificial Intelligence
• Agriculture
• Communications, Connectivity and
• Automotive and Transportation
5G Technologies
• Healthcare, Pharmaceuticals, and
• Cybersecurity, Data Security, and Privacy
Medical Devices
• Data and The Internet of Things
• Industrial Internet of Things
• Green Technologies – Environment,
• Maritime Internet of Things
Sustainability and the Circular Economy
• Smart Cities
• Sensors and Sensor Systems

The program will also include:

• Doctoral Symposium
IMPORTANT DEADLINES:
• Entrepreneurial and Innovation Workshop Technical Paper Submission:
• Industry Panels November 1, 2018
• Tutorials Early Bird Discount Registration:
• Special Sessions January 5, 2019
• Women in Engineering Program
Visit wfiot2018.iot.ieee.org for
• Workshops more information or to register.
• Young Professional Program

IENYIOT0002.indd 1 24/08/18 11:56 PM

 IEEE

Internet of Things Magazine

SEPTEMBER 2018, VOL. 1, NO. 1

2 Editorial and Introduction to the Issue: Risk and Rewards of

the Internet of Things
4 Policy and Regulatory Issues
Perspectives on IoT Policy from the U.S. NTIA.
6 IoT Standards
On a global basis, one of the challenges is to bring a necessary level of conformance that realistically
manages the risks of cyber-threats without impeding the functioning of the Internet of Things. This requires
reasonable standards.
8 Privacy and Security
When it comes to privacy and security risks, what is implicit across the myriad conceptualizations of IoT lies
the key to unearthing why IoT risk heralds a difference with a distinction compared to traditional offline and
online contexts.
12 Around the World of IoT
We live in a technology-oriented world where the business bottom-line numbers matter more than ever
before. The IoT world is in a transition period, where a technology has to convince the world that it can
deliver on its promises.

FEATURED VERTICALS
14 Why Industry Needs to Accelerate IoT Standards
Maciej Kranz
20 Drilling Deep into Digital Industrial Transformation Will
Determine Who Survives and Thrives
William Ruh
24 Connecting the Dots on IoT for the Industrial World
Karine Lavoie-Tremblay
28 The Future of IoT
Joern Ploennigs

34 Afterword and Introduction for Issue No. 2

EDITOR-IN-CHIEF COLUMN EDITORS Scanning the Literature

Keith Gremban, National Telecommunications and Around the World of IoT To Be Determined
Information Administration (NTIA) (USA) Raffaele Giaffreda, FBK CREATE-NET (Italy)
Bridging the Physical, the Digital, and the Social PUBLICATIONS STAFF
EDITORIAL BOARD Jun Zhang, University of Denver (USA) Joseph Milizzo, Assistant Publisher
Nuno Carvalho, Universidade de Aveiro (Portugal) IoT Standards Jennifer Porcello, Production Specialist
Shawn Chandler, Pacificorp (USA) Mike Violette, Washington Laboratories (USA) Catherine Kemelmacher, Associate Editor
Robert Douglass, Alta Montes (USA) Policy and Regulatory Issues Susan Lange, Digital Prod­uction Manager
Joern Ploennigs, IBM Research (Ireland) Doug Sicker, Carnegie Mellon University (USA)
Arjmand Samuel, Microsoft (USA) Privacy and Security
Sachin Seth, Tesla (USA) Erin Kenneally, Department of Homeland Security (USA)
Massimo Vecchio, FBK CREATE-NET (Italy) and International Computer Science Institute (USA)

IEEE Internet of Things Magazine • September 2018 1

MESSAGE FROM THE EDITOR-IN-CHIEF

Editorial and Introduction to the Issue:

Risk and Rewards of the Internet of Things

W elcome to the inaugural issue of IEEE Internet of

Things Magazine (IoTM)! IoTM is sponsored by the
IEEE Internet of Things (IoT) Initiative, a multi-disci-
plinary initiative with 19 member societies/councils. The mis-
sion of the IEEE IoT Initiative is to “serve as the gathering place
for the global technical community working on the Internet of
of the car and cargo, and inspect the state of the rails and
roadbed.
• In the agricultural domain, many growers have installed
distributed networks of sensors that provide real-time data
and historical trends for multiple parameters such as soil
moisture, local temperature, humidity, and others. Growers
Things; to provide the platform where professionals learn, share use the data to optimize watering, fertilizer application, and
knowledge, and collaborate on this sweeping convergence more. Some agribusinesses now employ networked autono-
of technologies, markets, applications, and the Internet, and mous systems to observe crop status and apply treatments.
together change the world.” • In the industrial domain, miniaturized sensors integrated
The IoT is one of the most important, exciting, and trans- into critical equipment monitor performance parameters
formational technology developments today. IoT is global in to proactively diagnose maintenance issues, enable trend
impact, multi-disciplinary in nature, and spans virtually all indus- analysis of equipment performance, and optimize overall
try segments. The IoT is a topic that is growing in interest, yet system operations. The cost of sensors and sensor inte-
no concise definition of IoT is universally accepted within the gration is more than offset by the savings in maintenance
community of practitioners. A document produced by the IEEE costs and increase in equipment up-time.
IoT Initiative states that “Despite the diversity of research on The growth in IoT has been nothing less than astounding, and
IoT, its definition remains fuzzy.” The document attempts to is forecast to continue. For example, Cisco projects over 10 bil-
address this issue, but concludes with three pages that define lion machine-to-machine IoT devices will be connected in 2019.2
IoT with a list of characteristics of IoT systems.1 Very generally, Intel predicts 200 billion connected devices in 2020.3 McKinsey
the IoT refers to collections of things, such as devices, vehicles, Global Institute estimates the economic impact of IoT to be as
sensors, and actuators, for example, that interconnect and com- much as $11.1 trillion per year by 2025.4 Many corporations
municate over a network to perform some application. IoT inte- now have IoT divisions led by corporate vice-presidents.
grates the physical and the digital, providing unique solutions IoTM was proposed to meet the needs of industry, govern-
that exploit sensing, aggregation of data from multiple sources, ment, and academic practitioners, who are working to design
common infrastructure such as communications, computing and deploy IoT applications every day. Through a mix of articles
and storage, to remotely monitor and control physical systems. and regular columns, IoTM will present IoT solutions, report
IoT applications range from chemical plant control to personal on IoT experiences throughout the community, analyze IoT
health monitoring, from automobile engines to transportation deployments from a business perspective, and point readers
systems, from electrical generators to smart grids, and more. to relevant literature, events, and activities. The objective is
Some example IoT applications, which are representative for IoTM to be a forum for practitioners to share experiences,
topics for IoTM articles, include: develop best practices, and establish guiding principles for tech-
• In the transportation domain, a railroad company uses IoT nical, operational, and business success.
technology to instrument thousands of miles of track, as The articles in our inaugural issue present examples of suc-
well as every locomotive and railcar. Customized sensor cessful IoT deployments, as well as recommendations for indus-
packages placed at various locations along the track mea- try growth, and some perspectives on the future of IoT.
sure various externally observable properties of every car IoT is growing rapidly, and more and more organizations are
as the train passes by. Communication between sensor appreciating the financial benefits of investing in IoT. However,
stations enables observations of trends in the data so that rapid growth is bringing challenges to the IoT community. The
proactive maintenance can be scheduled in advance and article “Why Industry Needs to Accelerate IoT Standards” dis-
performed when the train reaches a location with a main- cusses two key issues, security and standards, in the context of
tenance yard. Similarly, customized sensors on board each representative IoT deployments and makes some recommenda-
car monitor various parameters that indicate the stability tions for the IoT community to follow.

IEEE INTERNET OF THINGS MAGAZINE (ISSN 2576-3180) is published quarterly by The Institute of Electrical and Electronics Engineers, Inc. Headquarters address: IEEE, 3 Park Avenue, 17th Floor, New
York, NY 10016-5997, USA; tel: +1 (212) 705-8900. Responsibility for the contents rests upon authors of signed articles and not the IEEE or its members. Unless otherwise specified, the IEEE neither endorses
nor sanctions any positions or actions espoused in IEEE Internet of Things Magazine.
ANNUAL SUBSCRIPTION: US$53 print subscription; US$38 electronic subscription; US$624 non-member print subscription.
EDITORIAL CORRESPONDENCE: Editor-in-Chief, Keith Gremban, e-mail: [email protected].
COPYRIGHT AND REPRINT PERMISSIONS: Abstracting is permitted with credit to the source. Libraries are permitted to photocopy beyond the limits of U.S. Copyright law for private use of patrons:
those post-1977 articles that carry a code on the bottom of the first page provided the per copy fee indicated in the code is paid through the Copyright Clearance Center, 222 Rosewood Drive,
Danvers, MA 01923. For other copying, reprint, or republication permission, write to Director, Publishing Services, at IEEE Headquarters. All rights reserved. Copyright © 2018 by The Institute of
Electrical and Electronics Engineers, Inc.
POSTMASTER: Send address changes to IEEE Internet of Things Magazine, IEEE, 445 Hoes Lane, Piscataway, NJ 08855-1331. Printed in USA. Periodicals postage paid at New York, NY and at additional
mailing offices. Canadian Post International Publications Mail (Canadian Distribution) Sales Agreement No. 40030962. Return undeliverable Canadian addresses to: Frontier, PO Box 1051, 1031
Helena Street, Fort Eire, ON L2A 6C7.
SUBSCRIPTIONS: Orders, address changes — IEEE Service Center, 445 Hoes Lane, Piscataway, NJ 08855-1331, USA; tel: +1 (732) 981-0060; e-mail: [email protected].
ADVERTISING: Advertising is accepted at the discretion of the publisher. Address correspondence to: Advertising Manager, IEEE Internet of Things Magazine, IEEE, 3 Park Avenue, 17th Floor, New
York, NY 10016-5997.
SUBMISSIONS: The magazine welcomes high-quality articles on IoT technology and end-to-end IoT solutions. Submissions will normally be approximately 4500 words, accompa-
nied by up to six figures and/or tables, with up to fifteen carefully selected references. Electronic submissions are preferred and should be submitted through Manuscript Central:
https://mc.manuscriptcentral.com/iotmag. All submissions will be peer reviewed.

2 IEEE Internet of Things Magazine • September 2018

 MESSAGE FROM THE EDITOR-IN-CHIEF
Keith Gremban ([email protected]) is the Direc-
The pace of technological change is driving the rise and fall tor of the Institute for Telecommunication Sciences
of companies and even entire industries. IoT is proving to be (ITS), which is the research and engineering laborato-
one of those transformational technologies that may determine ry for the National Telecommunications and Informa-
the winners and losers in many sectors. The article “Drilling tion Admin-istration (NTIA). Keith has been involved
in systems engineering and advanced technology
Deep into Digital Industrial Transformation Will Determine Who devel-opment for over thirty years. Prior to arriving
Survives and Thrives” uses case studies to present the argument at ITS, he was a Program Manager at the Defense
that one type of transformation in particular is required for any Advanced Research Projects Agency (DARPA) where
organization that wants to thrive in the 21st century. he managed a portfolio of programs in the areas of
wireless communications and electronic warfare.
IoT provides a mechanism for organizations to do more with Prior to DARPA, He worked at a variety of compa-
less, while providing agility and responsiveness to customer nies and research institutes, managing and leading
needs. The article “Connecting the Dots on IoT for the Industri- research and systems engineering pro-jects, includ-
al World” presents some examples of the application of dense ing a diverse collection of unmanned systems and
command-and-control applications. He received his Ph.D. and M.S. in Computer
sensing and near-real-time data analytics to aircraft system main- Science from Carnegie Mellon University, and his M.S. in Applied Mathematics
tenance and operations. and B.S. in Mathematics from Michigan State University.
Our closing article “The Future of IoT” presents a panel dis-
cussion among three experts in IoT regarding the risks and FOOTNOTES
rewards of IoT. Additionally, the panel discusses the real and 1 IEEE_IoT_Towards Definition_Internet_of_Things_Revision1_27May15.pdf,
potential impact to IoT of some of today’s most anticipated https://iot.ieee.org/images/files/pdf/IEEE_IoT_Towards_Definition_Internet_
of_Things_Revision1_27MAY15.pdf
technologies. 2 http://www.cisco.com/c/en/us/solutions/collateral/service-provider/visual-net-
In closing, the Editorial Board would like to thank the authors working-index-vni/VNI_Hyperconnectivity_WP.html
that contributed to this inaugural issue. We hope that the com- 3 http://www.intel.com/content/www/us/en/internet-of-things/infographics/

munity of IoT practitioners will find the articles and columns to guide-to-iot.html
4 https://www.mckinsey.com/business-functions/digital-mckinsey/our-insights/the-
be interesting and useful, and provide support for Internet of internet-of-things-the-value-of-digitizing-the-physical-world
Things Magazine in the form of articles, columns, references,
and, of course, subscriptions.

Enjoy the Many Benefits of

IEEE EPS Membership
JOIN TODAY! EPS.IEEE.ORG

• Informative, Educational Conferences

• Career Development Tools
• Access to the latest technical information
• Networking opportunities
• Prestigious Awards and
Recognition

IENYIOT0006.indd 1 02/10/18 2:08 AM

IEEE Internet of Things Magazine • September 2018 3

 POLICY AND REGULATORY ISSUES
Policymakers face a conundrum — promoting the adoption of IoT services to reap its many benefits, while safeguarding societal
concerns. This will be a balancing act of oversight and regulation from policymakers to drive investment and consumer adoption while
ensuring that safety, security, and privacy frameworks are in place. This column will explore critical national and international IoT policy
and regulatory efforts as well as take a deeper dive into specific topics of interest.

Introduction Commerce and Administrator of the National Telecommunica-

tions and Information Administration (NTIA), David Redl, as our
Policymakers face a conundrum: promoting the first contributor to this Regulatory and Policy Column, discuss-
adoption of IoT services to reap its many ben- ing key findings of NTIA’s recent green paper on IoT. In future
efits, while safeguarding societal concerns. This issues, we will explore critical IoT policy and regulatory issues
will be a balancing act of oversight and regula- being considered in other parts of the world as well as take a
tion from policymakers to drive investment and deeper dive into specific topics of interest. It is difficult to over-
consumer adoption while ensuring that safety, state the impact that IoT will have on our society; therefore,
Douglas C. Sicker security, and privacy frameworks are in place. getting the policy and regulatory issues right is critical because
Meanwhile, industry must step up to lead and the policies will guide the ultimate success and direction of this
adopt best practices and standards, which will allow policymak- important digital evolution. The breadth and depth of policy
ers to forbear as appropriate. and regulatory issues facing IoT are surprisingly vast, ranging
In this inaugural issue of IEEE Internet of Things Magazine from safety, privacy, security and spectrum policies to issues of
(IoTM), we are fortunate to have the Assistant Secretary of infrastructure coordination, rights of way and acceptable use.

Perspectives on IoT Policy stakes in the benefits and the risks. In other words, IoT brings
new challenges to ongoing and longstanding policy debates.
For example, IoT raises particular cybersecurity challenges, but
from the U.S. NTIA it would be counter-productive to consider them outside of the
context of broader cybersecurity conversations or to reinvent
by David J. Redl the wheel where progress has been made elsewhere.
Assistant Secretary for Communications and Information and This insight is reflected in the policy paper titled Fostering the
Advancement of the Internet of Things (https://www.ntia.doc.
Administrator, National Telecommunications and Information gov/files/ntia/publications/iot_green_paper_01122017.pdf)
Administration, U.S. Department of Commerce that was released by the Department of Commerce’s Internet
Policy Task Force. Informed by public comments from industry,
The Internet of Things (IoT) is not the future — it is the present. academia, civil society, individuals and a public workshop, the
From smart home devices that open blinds, brew coffee, and paper concludes that the United States’ policy approach that
turn on the news in the morning to industrial applications rev- helped lead to the global success of the Internet continues
olutionizing supply chain management, connected devices are to be the best way to respond to innovative technologies. It
changing how we live, work, and play. also lays out the following four areas that could help guide the
At the National Telecommunications and Information Department’s efforts to encourage IoT growth and innovation
Administration (NTIA), we understand the potential benefits of in a manner that is inclusive and widely accessible, and is within
IoT and are dedicated to ensuring that we have the policies and a stable, secure and trustworthy environment:
infrastructure in place to support the innovation behind these Enabling Infrastructure Availability and Access: IoT only
advances. In depth technical understanding is vital to well-run works so long as there is connectivity, and that connectivity
policy making processes, and it is equally important for the is dependent upon telecommunications infrastructure, both
technical community to grapple with the wider impacts of its wireline and wireless. NTIA plays important roles, as our Office
work. IEEE IoT Magazine is therefore a welcome addition to the of Telecommunications and Information Applications works to
conversation, and I appreciate the opportunity to contribute to ensure that unserved communities gain access to the broad-
the inaugural edition. band connectivity necessary for IoT applications, and our Office
NTIA is the Executive Branch agency located within the U.S. of Spectrum Management both manages Federal spectrum
Department of Commerce that is principally responsible for resources and works with the Federal Communications Com-
advising the President on telecommunications and information mission to identify additional spectrum for commercial use.
policy issues. NTIA’s programs and policymaking focus large- Our contribution to wireless connectivity also includes the work
ly on expanding broadband Internet access and deployment of the Institute for Telecommunication Sciences (ITS), NTIA’s
in America, increasing the use of spectrum by all users, and research lab, which is exploring what the real spectrum needs
ensuring that the Internet remains an engine for continued are going to be with the onset of IoT. These efforts put us at the
innovation and economic growth. This broad mandate includes forefront of working to help enable access to robust and inno-
emerging technologies, such as those powering IoT. Within the vative IoT solutions for everyone.
Administration’s interagency processes, NTIA is a primary voice Crafting Balanced Policy and Building Coalitions: The
at the table focused on innovation and economic growth. advancement and adoption of IoT will also be affected by the
Neither the fundamental technologies nor the policy chal- policies that are in place to help encourage trust while safe-
lenges of IoT are novel. IoT differs in the sheer number of devic- guarding innovation. This will require close collaboration across
es that will be connected to the Internet and to each other, the government with industry and civil society to take on issues,
the variety of industries that are newly integrating connectivity such as privacy, cybersecurity, and intellectual property, among
into their products and business processes, and the increased others, that will shape the IoT ecosystem. The United States
has a successful track record in achieving this balance, but it
Editor’s Note: Text appearing in bold indicates a live link in the online version. will take dedication and cooperation across sectors to ensure

4 IEEE Internet of Things Magazine • September 2018

 POLICY AND REGULATORY ISSUES

that this continues for IoT and other emerging technologies. for Economic Co-operation and Development, and the Inter-
NTIA’s Office of Policy Analysis and Development and Office net Governance Forum. We engage as well on the bilateral
of International Affairs are key players in helping to shape these and regional level, representing U.S. positions and promoting
policies. dialogue. We actively support NIST’s Global Cities Challenge
Promoting Standards and Technology Advancement: The work, which has this year added consideration of cybersecurity
Department of Commerce, through ITS and the National Insti- issues that smart cities face.
tute of Science and Technology (NIST), is committed to ensur- While I highlight cybersecurity, this is only one aspect of IoT
ing that the necessary technical standards are developed and in which NTIA is engaged. All of our diverse efforts are predi-
in place to support global IoT interoperability, and that the cated on stakeholder engagement and reliance on the expertise
technical applications and devices to support IoT continue to of practitioners. So we want to hear from you about the chal-
advance. We remain steadfast in our support of industry-driv- lenges that you face and your thoughts on potential solutions. It
en, consensus-based, voluntary, global standards. NTIA helps is only through your active participation and sharing of knowl-
support these efforts in a number of international organizations. edge that we will be able to craft the informed policy solutions
Encouraging Markets: Finally, the Department of Com- that a vibrant IoT ecosystem requires. NTIA is excited to work
merce is working to promote IoT through the use of its own IoT with you to ensure that IoT continues to fulfill its extraordinary
devices, iterative enhancement, and novel deployment of the potential.
technologies. We will also be working with our global partners
to help translate the economic benefits and opportunities of David J. Redl was sworn in as Assistant Secretary for Commu-
nications and Information at the Department of Commerce in
IoT, expanding the market not only domestically but world- November 2017. He serves as Administrator of the National
wide. Telecommunications and Information Administration (NTIA),
To show how this work is being translated into practice, the Executive Branch agency that is principally responsible for
I want to spend some time drilling down into how NTIA is advising the President on telecommunications and information
policy. He is a lawyer and communications policy expert with
approaching what commenters on our report cited most fre- more than a decade of experience in government and the pri-
quently as a challenge to the advancement of IoT: cyberse- vate sector. He was previously the chief counsel at the U.S. House of Representa-
curity. NTIA is taking a multipronged approach to addressing tives Committee on Energy and Commerce. In that role, he served as principal legal
cybersecurity concerns, takings steps to work with stakeholders advisor to the chairman and members of the Energy and Commerce majority on
communications and technology matters. Prior to his time with the committee, he
and within the government to craft balanced solutions that can was director of regulatory affairs at CTIA, a trade association that represents the U.S.
affect real-world change. wireless communications industry. He earned his J.D. from the Catholic University
First, NTIA has convened cybersecurity multi-stakeholder of America with a certificate from the Institute for Communications Law Studies,
processes that have brought together representatives of indus- and he is a graduate of Pennsylvania State University with degrees in journalism and
political science. He is admitted to the New York and District of Columbia bars. He
try, civil society, academia and the security research community lives in Falls Church, Virginia, with his wife, Amy, and their son, Benjamin.
to tackle difficult policy issues. The second of these processes
focused on patching and upgrading IoT devices to limit vul- Douglas C. Sicker ([email protected]) is currently the Lord Endowed Chair in Engi-
nerabilities. This process resulted in three finished work prod- neering, department head of Engineering and Public Policy, director of CyLab
Security and Privacy Institute, and a professor of engineering and public poli-
ucts — a guide for consumers, a technical breakdown of the cy with a joint appointment in the School of Computer Science and courtesy
patching process and a discussion of incentives and barriers appointment in the Heinz College at Carnegie Mellon University. He is also the
to patching — as well as a compendium of standards and best Executive Director of the Broadband Internet Technical Advisory Group (BITAG).
practices, which will be maintained as a living document. These Previously, he was the DBC Endowed Professor in the Department of Computer
Science at the University of Colorado at Boulder with a joint appointment in, and
documents, as well as more information about our processes, directorship of, the Interdisciplinary Telecommunications Program. He recently
can be found on our website here: https://www.ntia.doc.gov/ served as the chief technology officer and senior advisor for Spectrum at the
other-publication/2016/multistakeholder-process-iot-security. National Telecommunications and Information Administration (NTIA). He also
NTIA has also worked closely with NIST and with the served as the chief technology officer of the Federal Communications Commis-
sion (FCC), and prior to that he served as a senior advisor on the FCC National
Department of Homeland Security to produce a report on how Broadband Plan. Earlier he was director of Global Architecture at Level 3 Commu-
to combat distributed threats on the Internet, which can be nications, Inc. In the late 1990s, he served as Chief of the Network Technology
found here: https://www.commerce.gov/page/report-presi- Division at the FCC. He is an active member of ACM, AAAS, and the Internet
dent-enhancing-resilience-against-botnets. This report benefited Society. He has served as an advisor to the Department of Justice, the Federal
Trade Commission, the FCC, and the Department of State; the Chair of the FCC
greatly from stakeholder input through both comments and a Network Reliability and Interoperability Council steering committee; an advisor on
workshop and lays out actionable steps that can be taken to the Technical Advisory Council of the FCC, and chair of a recent National Acade-
address this abuse of Internet-connected devices. my study on the Boulder Department of Commerce Laboratories. He has chaired
NTIA is further engaged at the international level on devel- numerous conferences as well as served on many program committees and sever-
al National Academy studies. He has published extensively in the fields of wireless
oping cybersecurity policies, working in forums such as the systems, network security, and network policy, and has received funding from
International Telecommunications Union, the Organization NSF, DARPA, FAA, Cisco, Intel, IBM, and other sources.

IEEE Internet of Things Magazine • September 2018 5

 IOT STANDARDS
This column will look at different segments of the IoT market as it relates to implementation and use of standards. Each column will
select a particular vertical, and lay out the relevant standards and technologies that affect the evolving IoT hyperspace. The pace of the
columns will start broadly with the vision of narrowing the subject of subsequent articles toward more specific applications of standards,
whether in the development, application, test, or commissioning of IoT technologies.

IoT Standards Matters The IEEE is, obviously, an active contributor to the standards
world, with many of the IEEE outputs being elevated to Nation-
al Standards under the American National Standards Institute
by Mike Violette (ANSI), which is celebrating 100 years since its founding in May
Washington Laboratories, USA 1918, and whose mission is to “administer and coordinate the
U.S. voluntary standards and conformity assessment system.”
The universe of standards issues that affect IoT development A sister initiative under the aegis of the IEEE is the 5G Initia-
and implementation is vast and complicated, involving many tive, and the overlap of 5G and IoT is significant. One resource
organizations, technologies and interests. This article discuss- that has been developed under that initiative is the IEEE 5G
es the intersection of three elements of the IoT Standards Eco- Standards Database, found here: https://5g.ieee.org/standards/
system, namely: the principal organizations that drive Global standards-database. This database is a collaborative effort aimed
Standards Development, Standards for Functionality and Com- at collecting, in one place, standards that have “something to do
patibility, and Standards for Security and Privacy. These topics are with 5G,” and is a crowd-sourced kind of effort. Contributions
evolving rapidly and, like the Internet of Things, constantly mor- are welcome and encouraged from interested parties.
phing as technology solutions are developed and implemented. Clearly, the efforts undertaken to create this broad range of
Not being able to swallow the whole thing, our plan is to eat specifications and methods over the past decades has led to
the standards watermelon a bit at a time over the next several fundamental changes in our way of life. It is certainly expected
issues of IEEE Internet of Things Magazine. to proceed apace and shows no sign of easing as various influ-
So, for this inaugural article, I’ve chosen to briefly introduce encers cooperate and compete with each other for standards
the standards players by some admittedly subjective, broad, territory.
divisions: The Standards Development Organizations (SDO) The interaction between the various SDOs are varied and
Space, the Government Space and the Industry Space. A complex and involve a mix of face-to-face get-togethers requir-
sample of some of the players will be described and further ing many gallons of coffee as well as significant virtual efforts,
explored in future installments. where draft versions of documents whirl about on the Internet
as part of consensus efforts that aim to include many voices.
SDOs Often, it works well. Not infrequently the process may not work
Standards Development Organizations SDOs or Standards Set- so well and there are often winners and losers in the standards
tings Organizations SSOs operate, in general, according to certain race (think the long-ago VHS vs BETA square-off).
defined processes. Many organizations operate through a con-
sensus process that is characterized by openness, transparency, Standards for Functionality
balance, and due process or mechanisms for ensuring adherence A second layer of this standards-dive includes functionality. This
to organizational procedures, including provision for appeals.” 1 is a pretty critical aspect of the standards biome, and winners
The global standards eco-system is varied and vast with a mix of and losers are also defined in this space.
players with various, and sometimes opposing, agendas. With so many players in the IoT space, it’s a little dizzying to
Figure 1 represents a cross-section of standards-setting bod- determine the dominant drivers for IoT functionality; in fact, to
ies that are involved in various IoT standards (among other be honest, the task is a bit overwhelming. Major players in the
things, such as EMC, Electrical Safety, Radio/Wireless and industry, from software and hardware developers to govern-
Cyber-Security) involved with formally-recognized national ment to network operators, have already staked out decades of
standards bodies, committees and global organizations as well operating methods that drive IoT functionality.
other “fora and consortia.” This space includes consensus-driven efforts as well as pri-
The availability of the standards depends largely on the fund- vate and proprietary standards. One organization that has an
ing strategies. Many SDOs charge for their standards (IEC, ISO, open, consensus-based structure is the Internet Engineering
IEEE, etc.) while others distribute the standards for free (ETSI, Task Force (IETF), which is a “large open international commu-
e.g.) as part of government-sponsored efforts to promulgate nity of network designers, operators, vendors, and researchers
the information. In many of the standards-development models, concerned with the evolution of the Internet architecture.”
much of the work is performed by volunteers who presumably This particular group has over 100 working groups aimed at
have a stake in the outcome of the end-product, whether pro- improving “the smooth operation of the Internet.” The work-
moting a certain technology, protecting a bit of existing “real ing groups have defined areas that range from HTTP to Video
estate” or staying connected for other reasons. (I have met Codecs to IPv6 Operations to Software Updates for Internet
many standards “nerds” who do this for fun!) of Things. This group is open to anyone who is interested in
How big is the standards universe? Just a brief perusal of contributing to the IETF’s core mission and there are no mem-
the IEEE Standards Association website2 shows a “partial list- bership fees. https://www.ietf.org/about/participate/.
ing” of IEEE standards that are related to IoT. This listing has The development and improvement in standards and specifi-
over 70 standards, from base standards like IEEE 802.3-2012 cations are conducted through mail lists and collaborative tools,
“IEEE Standard for Ethernet” to more esoteric and specific IEEE while face-to-face meetings serve the purpose of putting faces
1609.11™-2010 — “IEEE Standard for Wireless Access in Vehic- with email addys.
ular Environments (WAVE) — Over-the-Air Electronic Payment The IETF’s IoT efforts include melding IoT needs with existing
Data Exchange Protocol for Intelligent Transportation Systems standards, such as the first WG chartered in 2005 (6LoWPAN)
(ITS).” which defined methods for adapting IPv6 to IEEE 802.15.4
(wireless personal area networks (WPANs)). Other wireless
Editor’s Note: Text appearing in bold indicates a live link in the online version. technologies are similarly adapted, including flavors of Blue-

6 IEEE Internet of Things Magazine • September 2018

 IOT STANDARDS

Figure 1. A cross-section of standards-setting bodies that are involved in various IoT standards Figure credit:
Dr. Jochen Friedrich, IBM Europe.

tooth and Digital Enhanced Cordless Telecommunications

(DECT) Ultra Low Energy (ULE) cordless phones. The intent is
to make efficiency and reliability a prime feature of the IoT.
Naturally, the carriers and big data folks have evolved their
own practices and standards. Verizon, for example, has a strong
interest in connectivity for Machine-to-Machine (M2M) connec-
tions and the revenue that equates to the many billions of over-
the-air connections that are growing on its networks. The major
themes of its IoT services include the role of M2M connectivity
in Smart Cities, route planning and dispatch for delivery ser-
vices, mobile commerce and asset tracking.
From Verizon’s report, “State of the Market: Internet of
Things 2017: Making Way for the Enterprise,” 3 the subject of
standards comes up on page 1: “An absence of industry-wide IoT Figure 2. NIST cybersecurity framework.
standards, coupled with security, interoperability and cost consid-
erations make up over 50 percent of executive concerns around icies and Procedures, System, and Component. The structure
IoT, according to Verizon’s survey.” is meant to guide the development of a compliant and secure
Clearly, major players are working in many spaces to make process, whether it involves software, hardware or a mix.
this uncertainty diminish, and the importance of security is at One aspect of the IEC 62443 structure includes the opportu-
the top of the list. nity for systems to be evaluated under a Conformity Assessment
process and certification by Certification Bodies accredited for
StanDarDS fOr SEcurity anD PriVacy the discipline. Certification has most broadly been applied to
The National Institute of Standards and Technology (NIST) devices, either evaluation for conformance with safety require-
issued a voluntary “Cybersecurity Framework”4 that “consists ments, spectrum use and related physical conformance. The
of standards, guidelines, and best practices to manage cyberse- IEC 62443 Conformance Certification reviews the processes by
curity-related risk.” This guidance is broad and intended to be which an organization has assured that their processes, code
flexible and adaptable to the many different needs of the IoT and security measures properly implement the applicable IEC
space. The word “standard” appears over 30 times in the docu- 62443 requirements.
ment, and thus highlights the importance echoed in the words On a global basis, one of the challenges is to bring a neces-
of the Verizon report. sary level of conformance that realistically manages the risks of
The framework lays out tools to reach certain desirable out- cyber-threats without impeding the functioning of the Internet
comes and is designed to give a methodology for managing of Things. This requires reasonable standards and, for the fore-
and reducing cyber threats. The key part of the framework seeable future, will be an active area of development across the
relies on implementing these five functions: Identify, Protect, entire IoT space.
Detect, Respond and Recover, and can be further broken down
into Categories and Subcategories with references that may Michael Violette ([email protected]) is president of Washington
Laboratories and director of the American Certification Body. He
be used to inform an entities’ structure. Some examples of has over 25 years of experience in the field of EMC evaluation
Categories include: “Asset Management, Access Control and and product approvals, and has overseen the development of
Detection Processes.” engineering services companies in the United States, Europe, and
How an organization implements this framework depends Asia. He is a Professional Engineer, registered in the State of Vir-
ginia. He has given numerous presentations on compliance top-
on the needs and implementation of whatever operations it is ics and is a regular contributor to technical and trade magazines.
engaged in. Coupled with this methodology is a need for Risk
Assessment to be overlaid in an organization’s operation. FOOTNOTES
Other, international, standards for cyber security exist. One 1 Overview of International Cybersecurity and Privacy Standards Development.
example of this is the series of standards under IEC 62443. This 2
Elaine Newton, PhD. Oracle Corporation
http://standards.ieee.org/innovate/iot/stds.html
particular standard morphed from an ANSI standard (ANSI/ 3 http://www.verizonenterprise.com/verizon-insights-lab/state-of-the-market-inter-
ISA-99 or ISA99) to a standard under the International Elec- net-of-things/2017/
trotechnical Committee (IEC). It was originally developed by 4 https://www.nist.gov/cyberframework

the International Society for Automation (ISA) and includes

four general components or categories, namely: General, Pol-

IEEE Internet of Things Magazine • September 2018 7

 PRIVACY AND SECURITY
This column delves into privacy risks of the IoT using risk concepts that are more native to the security domain in order to conceptu-
ally bridge our collective understanding, articulation, and management of privacy concerns in the IoT which otherwise might not be
sufficiently considered or foreseen by existing legal and technical controls.

Scratching Below the Technical Underpinning

The technical canvas of the IoT provides the underpinning and

Surface: IoT Privacy Risk precursor for privacy threats, vulnerabilities and harms.
The architecture of the IoT reduces friction in the collection,
processing, disclosure and actuation of data. The effect is a
by Erin Kenneally
blurring of temporal, spatial and organizational boundaries that
Department of Homeland Security (USA) and have heretofore separated our physical, digital, biological, and
International Computer Science Institute (USA) social spheres. IoT sensors act as vectors for digitizing anything
that can be sensed, resulting in the reduction of communica-
It has become almost cliché that the so called Internet tions, visual, auditory, physical and biological systems to pure
of Things (IoT) means different things to different people. information that can then be managed, interconnected and
When it comes to privacy and security risks, what is implicit controlled.2
across the myriad conceptualizations of IoT lies the key to According to Boston Consulting Group, there will be 18–50
unearthing why IoT risk heralds a difference with a distinc- billion connected devices by 2020, compared with 6–14 billion
tion compared to traditional offline and online contexts. connected devices in 2014.3 Those numbers are aligned with
Critics of an exceptionalist view of IoT risk might contend Gartner’s research that forecasts 20.5 billion connected devices
that the IoT is really just distributed computing on ste- by 2020, up from 8.4 billion in 2017.4 The scope of digitiza-
roids, i.e., the IoT is merely a relabeling and repackaging tion and level of precision of IoT data capture is transforming
of technologies past like client-server, web services, SoA, everything into data, and advances beyond a capture of fuzzy
mobile, virtualization, and distributed computing, which snapshots of activities and attributes toward advanced, high
means that risk management is merely an exercise in graft- resolution data and inferences. It is only logical then to expect
ing the decades-long understanding of privacy and security that the scale and opacity of data collection and flows will influ-
from those familiar contexts onto the IoT. Without getting ence the relationship between individuals and organizations
into a religious debate, it is incontrovertible that there are regarding the collection, use and disclosure of information in
advances in the quality and quantity of data collection from ways that have critical privacy implications. The technical drive
IoT technologies as opposed to previous generations of to optimize and reduce friction in information flows results in
technology, and these are driven by real and prospective increased privacy friction for individuals attempting to maintain
socioeconomic value propositions. privacy through control of their information.
The ubiquitous and pervasive connections between
IoT components (devices, hardware, software, communi- IoT Privacy Threats
cations, and applications) are creating and mediating new Privacy threats in the IoT are characterized by access, col-
cyber-physical, psychological, and economic seams where lection, use (analysis, actuation) and disclosure of sensitive
data is produced, aggregated, fused, consumed, dissemi- information (e.g., personally identifying) information in contra-
nated and acted upon. These features quantitatively and vention of individuals’ expectations. This is no different than
qualitatively impact the relationships between individuals privacy threats in the online ecosystem we experience on a
and organizations implicated by this data in ways heretofore daily basis. However, the scope of threats associated with the
unseen or understated. More specifically, we are likely to IoT is arguably more expansive, raising the likelihood of privacy
see increasing tensions between equity rights and interests in harm. Unsurprisingly, the threat landscape at play with IoT pri-
privacy, security, innovation, free expression and governance vacy should consider the omnipresent attack vector presented
playing out, where your right to privacy may conflict with my by malicious actors, where the myriad of sensors and actuators
right to security, the government’s interest in infrastructure offer an opportunity to weaponize IoT to collect, use and/or
protection and public safety, industry’s right to commercial disclose data in ways that negatively impact privacy. What’s
free speech, and other citizens’ interests in open and trans- more interesting, however, is how the IoT drives equity conflicts
parent data. between legitimate, non-malicious stakeholders — industry, the
Accepting the above framing of our challenge, and government, and fellow citizens — as a function of competing
anchoring off the notion that privacy is the rights and inter- rights and interests introduced by IoT capabilities. Take a smart
ests between individuals and organizations regarding the cities scenario where sensors collect, analyze and share data
collection, use and disclosure of information, we necessar- from light pole sensors that monitor vehicle and pedestrian traf-
ily have to think about IoT privacy in socio-technical terms. fic, parking and local transportation. My expectation of privacy
This introduces a range of dimensions such as governance, (to not be monitored or targeted), may conflict with the gov-
economics, management and risk. This inaugural column ernment’s interest in enhanced public services, which may clash
delves into privacy risk of the IoT. Risk is typically defined with my fellow citizen’s expectation of safety, which may collide
as the severity and likelihood of harm to something of value with industry’s claim to commercial free speech (travel logistics
(an asset) when a threat exploits a vulnerability. As such, IoT products and services).
privacy risk is the confluence of threats and vulnerabilities
that negatively impact privacy rights and interests, the assets Power Imbalance
in our risk formula.1s These equity tensions represent another threat posed by the IoT
that involves power imbalance. If power is the ability to collect,
Editor’s Note: This editorial is an abridged version of a section in a larger co-authored publication, process and actuate data to exert control over individuals in
Internet of Things Privacy Forum, “Clearly Opaque: Privacy Risks of the Internet of Things,” (May 2018). ways that negatively impact their self-determination, the IoT

8 IEEE Internet of Things Magazine • September 2018

 PRIVACY AND SECURITY

threatens to exacerbate disproportionate control of personal products and services are increasingly subject to commercial
information and perpetuate a lack of transparency essential pressures to generate revenue. Facebook stands as an obvious
to consumers’ exercise of appropriate control. Hence, more reminder of the reality that its social networking platform comes
equitable power relations are a prerequisite to and an indicator at the cost of selling access to users’ social activities. While the
of meaningful consent, and serve as a social and democratic revenue model for the IoT is still emerging, the expectation of
check and balance. Power inequity, on the other hand, is a freely using an IoT service without a transfer of personal data,
barrier to meaningful negotiations, competition, and bargaining and for that matter expecting that ownership of a device will
over competing rights and interests. In the IoT, power inequi- guarantee privacy, is tenuous.
ty will be a threat to privacy to the extent that data control is Consumers are being asked to provide and link more infor-
unchecked and consolidated by owners of platforms and ser- mation (e.g., cameras and voice to detect mood, emotion, ges-
vices that consumers depend upon, the IoT equivalents of Goo- tures, activities) to avail themselves of IoT functionality, yet so
gle, Amazon, Facebook, Microsoft, etc. If these IoT platforms far have been given limited tools to control that personalization.
are fueled by data from users, the users’ lack of control over Even when an individual is not the direct target of sensing, inci-
that data will threaten their self-determination and ultimately dental data captured by other people’s devices and the inter-
create a self-perpetuating power imbalance. If past is prologue connectedness of large volumes of data are another gap in the
and IoT privacy threats resemble what has occurred in a pre- IoT that render privacy vulnerabilities.
IOT ecosystem, the now infamous Facebook and Cambridge In addition to impacting the degree of vulnerability in the
Analytica controversy serves as a crystal ball. previous examples, the IoT introduces a relative difference in
the kind of vulnerability that can enhance privacy risk. The sens-
Insufficient Understanding of Threat ing and digitization of sentiments and emotions yields a new
The scope of the IoT threat landscape is further complicated path to measuring intimate parts of people in ways not seen
by a lack of general understanding of how threats will manifest. heretofore. As well, if the current imbalanced control over data
This precludes knowing the precautions and mitigation mea- migrates to the IoT, techniques to mitigate IoT privacy vulner-
sures to put in place to avoid or minimize adverse impacts. The abilities bode equally unlikely. Examples of imbalances include
Mirai botnet in the Fall of 2016 that commandeered hundreds a lack of opt-out mechanisms, uninformed consent such as
of thousands of unwitting IoT devices to impose untold millions incomprehensible privacy policies, and the lack of collection
of dollars in damage from business interruption, fraud, and and purpose constraints.
loss of data and customer loyalty revealed what some believe Similar to how a lack of understanding exacerbates threats
to be a tip of the iceberg with regard to the harm potential of because users will not apply appropriate protections to thwart
leveraging IoT devices to wreak financial, psychological, and them, the scope of privacy vulnerability is enhanced by a lack
physical harm.5 Also, Mirai is widely believed to be collateral of transparency in the collection, use and disclosure of data.
damage from a spat between MOOC players, which leaves Opacity of data flows, dissemination and aggregation (pack-
open an entirely other dimension of our deficient understand- aging and redistribution of data) can enhance privacy vulner-
ing of the adversarial capacities and incentives that drive inten- ability exposure. When exposure of data occurs incrementally
tional attacks in the IoT. over time and across locations, the privacy harm may be more
serious than acute data breaches because mitigation is never
IoT Privacy Vulnerabilities undertaken. Even if knowledge of collection exists, the vulner-
Another element in the privacy risk equation involves under- ability may persist due to lack of awareness of the novel ways
standing the weaknesses or gaps in protections that can be that data can be combined and augmented to transform seem-
exploited by the aforementioned threats to cause privacy harm. ingly benign data into an exposure, including that which is the
First of all, the scale and volume of data available for collection basis for the triad of authentication: something you have (e.g.,
and use expands the range of opportunities to exploit data that smart card), know (e.g., password), or are (e.g., biometric).
implicates privacy and therefore increases the probability of
realized harm. As discussed above in the technical underpin- IoT Privacy Harms
ning, the digitization of anything that leaves a trace or is subject Understanding the negative impacts on privacy rights and inter-
to sensing — biometrics, emotions, behaviors — introduces a ests (the assets) comprises the third consideration in the fram-
privacy exposure point. ing of IoT privacy risk. Examples of the potential differences
Second, the traditional boundaries by which society has con- with distinction, whether in degree or kind, regarding privacy
structed privacy expectations are blurring. It is hard for individu- harms in the context of IoT include the following.
als to know if the physical features that have assured a sense of
solitude, permitted people to act anonymously, and supported Personal Information Breaches and Identity Fraud
control over identities are becoming ineffective. This is the case Conspicuous risks of harm in the IoT are those that ensue from
when data flows in an opaque, unobtrusive, automatic, regular- inadequate security, such as breaches of personal information
ized manner, all promised features of the IoT. and identity theft.6 If an IoT company loses data about users’
Third, even when users are aware of data flows, privacy vul- personal behaviors gathered in their homes or in activities in
nerability can stem from inadequate security of IoT devices. IoT public and their identity is linked, this could cause measurable
devices are often cheap and developed by companies without harm to consumers. Breaches of personal data may become
significant security budgets or incentives. IoT devices are often the next iteration of data breaches with the risk of such breach-
low-power and lack large capacities for computation or storage, es likely to be quantitatively higher. The resultant harms may be
further complicating efforts to provide the security needed to the most easily quantifiable relative to other harms discussed
protect data. below.
As well, even when deficient security is not the cause of pri-
vacy vulnerabilities, context-shifting and blurring between data Violation of Expectations of Privacy
collection for commercial and social settings creates another Law sets formal expectations of privacy rights. When there is
type of privacy vulnerability. Personal and social transactions incongruity between what society believes privacy harm to be
and activities that are mediated by commercial information and what can be remediated via the law, we have fractured

IEEE Internet of Things Magazine • September 2018 9

PRIVACY AND SECURITY

expectations of privacy. An abridged waltz down Precedent Diminishment of Private Spaces

Lane signals this incongruity with regard to defining privacy All of the privacy-challenging IoT characteristics mentioned
harms that emerge as a result of changes wrought by technolo- — proximity, scale, increased monitoring, boundary crossing,
gy. The general trend in data breach case law anchors on finan- reduced ability to opt-out of collection — add up to a likely
cial or physical harm that has provably already occurred, with diminishment of private spaces. This harms people’s abili-
little recognition of future risk of harm or negative impacts that ty to achieve solitude and reserve, both from others and in
are cumulative and collective.7 thoughts. This includes a reduction in the availability of spaces
Signals of fractured expectations of privacy are also manifest for individuals to be able to retreat to and not be observed,
from a regulatory standpoint, although at least as far as enforce- where one can control who can be present, who is listening,
ment actions by the leading U.S. consumer protection agency and who is watching; places of seclusion. Diminished private
go, the gap between principles and implementation is smaller. spaces translates to reduced ability to withhold data from third
The Federal Trade Commission has negotiated consent decrees parties, such as lifestyle preferences, family dynamics, and
based on nonmonetary, abstract, autonomy and dignitary-based hobbies.
harms.8 The purpose of this deep dive into IoT privacy risk is not
In addition, the qualitative and quantitative capture, access to advocate for a retreat from the promise of the IoT. Rather,
to and control over sentiment and emotion data can lead to it is meant to highlight and pre-empt some of the risks that
behavioral and psychological interactions with industry in ways may impede achieving the full promise of the IoT. As such,
that contravene current expectations of privacy, such as interfer- this column series is meant to motivate privacy and security
ence, manipulation, and encroachment on emotional privacy. as enablers of businesses rather than as a cost centers. Forth-
coming viewpoints in this series on IoT privacy and security
Impaired Autonomy are intended to engender this mindset by addressing gov-
Autonomy harm is that which impedes self-determination while ernance, economics and incentives, and risk management
directly or indirectly engaging with information systems. Auton- dimensions.
omy underlies many of the signals that the law uses to proxy
privacy harm: psychological (embarrassment, stigmatization, Erin Kenneally ([email protected]) is a currently a pro-
gram manager in the Cyber Security Division within the U.S.
loss of trust, chilling effects on ordinary behavior, discrimi- Dept of Homeland Security Science & Technology Director-
nation, intrusion on seclusion), economic (discrimination in ate. Her portfolio comprises cyber risk economics, data priva-
employment, credit, education, and insurance), and physical cy, trusted data sharing and research infrastructure, and ethics
(4th Amendment prohibition on unlawful search and seizure in in information and communications technology. She is found-
er and CEO of Elchemy, Inc., and served as technology law
the U.S.). specialist at the International Computer Science Institute and
Collective autonomy harms can have far-reaching disparate the University of California San Diego Supercomputer Center. She is a licensed
impact on the economic, physical, and psychological well-being attorney specializing in strategy, research and development, and execution of
of individuals and groups. Control of the personal data that challenging and emergent IT legal risk solutions.
informs the models produced by machine learning algorithms
FOOTNOTES
can result in public health and civil services disparities. Similarly, 1 See, e.g., NIST SP 800-30 Risk Management Guide for Information Technology
environmental sensor data can fail to aid vulnerable populations Systems, available at http://csrc.nist.gov/publications/nistpubs/800-30/sp800-
based on race or socioeconomic conditions. Collective auton- 30.pdf (Risk is a function of the likelihood of a given threat-source’s exercising a
omy harm can manifest as unequal access to and control of particular potential vulnerability, and the resulting impact of that adverse event on
data. This risks engendering mistrust between individuals and the organization.); The Committee on National Security Systems of United States
of America, CNSS Instruction No. 4009 (Apr. 2010) available at http://www.cnss.
institutions, resulting in impediments to or disengagement from gov/Assets/pdf/cnssi_4009.pdf.
social, political and economic activities that define individual 2 J. Ruiz-Rosero et al., Internet of Things: A Scientometric Review, Symmetry 2017,
and collective identities. Collective autonomy harms in the IoT 9, 301.
warrant attention because if left unabated, these power imbal- 3 Boston Consulting Group, “Winning in IoT: It’s All About the Business Processes,”
(Jan. 2017), available at https://www.bcg.com/publications/2017/hardware-soft-
ances get technologically embedded and institutionalized. They ware-energy-environment-winning-in-iot-all-about-winning-processes.aspx.
become hard to repeal and impact the entire fabric of social 4 Gartner IoT Report, Feb. 2017, available at https://www.gartner.com/newsroom/
relationships within which privacy interests reside id/3598917.
5 See, e.g., Cogeco “The Cost of DDoS Attacks and Building the Business Case for Protec-

Chilling Effects tion,” available at https://www.cogecopeer1.com/wp-content/uploads/2017/03/

Counting-the-Costs-of-DDoS-Attacks-DDoS-Services-Whitepaper.pdf; Altman
Lack of control can result in chilling psychological and behav- Vilandrie & Company, Are your company’s IoT devices secure? Internet of Things
ioral effects contrary to consumer-users’ intentions. These may Breaches are Common, Costly for U.S Firms (June 2017), http://www.altvil.com/
manifest as reluctance to engage or trepidation when encoun- wp-content/uploads/2017/06/AVCo.-IoT-Security-White-Paper-June-2017.pdf.
6 For example, in 2018 around 150M users’ personal details collected by the Under-
tering smart IoT devices. It can further manifest as psychologi- Armour/MyFitnessPal app, including usernames, email addresses and passwords,
cal insecurity about unwanted interference and manipulation, were leaked in a data breach.
where feelings of malaise, resignation, or helplessness are sub- 7 D. J. Solove, “A Taxonomy of Privacy,” University of Pennsylvania Law Review, vol.

jectively real but where people are unable to articulate legally 154, no. 3, p. 477, Jan. 2006; GWU Law School Public Law Research Paper No.
129. Available at SSRN: http://ssrn.com/abstract=667622.
cognizable harm due to a lack of information about whether or 8 See, e.g., In the matter of DesignerWare, LLC, No. 112-3151 (Apr. 15, 2013).
how data is actually being used.

10 IEEE Internet of Things Magazine • September 2018

 Technology insight
on demand on IEEE.tv
Internet television gets a mobile makeover
A mobile version of IEEE.tv is now available for convenient viewing.
Plus a new app for IEEE.tv can also be found in your app store. Bring
an entire network of technology insight with you:

t Convenient access to generations of industry leaders.

t
See the inner-workings of the newest innovations.
t
Find the trends that are shaping the future.

IEEE Members receive exclusive access to award-winning programs that

bring them face-to-face with the what, who, and how of technology today.

Tune in to where technology lives www.ieee.tv

IENYCM3584.indd 1 19/04/17 11:02 PM

 AROUND THE WORLD OF IOT
In this column we plan to take a tour around different physical locations in the world with the objective of highlighting the peculiarities
of the trendiest IoT-related applications in selected regions. Thus, the “IoT World” will certainly be physical, but traveling around it shall
also expose to the readers how different application domains have been addressed, with particular attention to business sustainability.

AN EXPLORATION JOURNEY
IN THE WORLD OF IOT WHERE
THE BOTTOM-LINE MATTERS
Big corporations Long tail
Consumers
by Raffaele Giaffreda Figure 2. The World of IoT.
Chief IoT Scientist, FBK CREATE-NET
We live in a technology-oriented world where the business accept for using such services? Currently for many, not high enough
bottom-line numbers matter more than ever before. The IoT to justify investment; thus, the quest for the consumer market killer
world is in a transition period, often illustrated on a Gartner Hype IoT applications is still on.
Cycle1 as the “Trough of Disillusionment,” where a technology Further exploration, beyond these mainstream applications
has to convince the world that it can deliver on the promises that and markets for IoT, also needs to reach far-away lands, looking
generated the “Peak of Inflated Expectations.” for applications to be positioned in the so called longtail of poten-
This article provides a perspective and seeks contributions that will tial IoT services. In this part of the IoT world we will find peculiar
inspire the IoT Magazine readers, reporting on challenges met and solu- products and solutions that came into existence to address a very
tions created within big corporations, consumer markets, but also longtail precise need, present in a specific region and nowhere else due
applications that are all characterized by solid “bottom-line” numbers. to the surrounding geo-cultural context. This is where unexpected
The costs of hardware procurement, like the sizes of components innovation might spark new and somewhat fresh ideas. Worth
needed to implement many parts of digitization processes across mentioning, as a source of inspiration, are some of the publica-
different industrial sectors, have been steadily going down. What is tions of the IEEE IoT Newsletter, which explore what IoT can do to
emerging as frequent and unquestionable Return on Investments support UN Sustainable Development Goals,2 or how Developing
(ROI) for IoT, is tied to savings on utilities and maintenance bills. Yet, Countries3 will leverage on IoT, narrowing down the scope with
gains in these sectors, characterized by a huge savings potential, only the needs of specific regions such as Latin America4 (mining and
become possible and more substantial with widespread availability agriculture industries), down to specific IoT projects for Canada5
of data, which in turn requires more distributed sensing capabilities (smart cities enabling IoT platforms), Japan6 (IoT in buildings for
and therefore more upfront investments. Making the numbers work earthquake-aware solutions), Finland7 (use of IoT in saunas), etc.
for a low margin business case requires substantial investment capac- Certainly, many gaps will need to be filled to enhance and extend
ity, hence this approach represents a “big corporations only” party. this initial map I have sketched here. This is what will keep this journey
Beyond this well-accepted by now role of IoT in targeting “around the world of IoT” exciting and interesting. The call for articles
savings on current operations, there is also an increased percep- therefore goes out to all explorers willing to inspire the IoT Magazine
tion, from the shareholders of various companies, that value and readers with challenges met and solutions created that fit within these
long-term strategies cannot depart from the compulsory path of three broad IoT world “continents” (big corporations, consumer mar-
digitization. This generates a subtle but compelling push, driven kets, and long-tail applications) and most importantly of all, are sup-
by peer pressure: there is no doubt that organizations in many ported by solid “bottom-line” numbers that matter so much.
sectors will extract value from a structural digitization strategy.
Raffaele Giaffreda ([email protected]) is a chief IoT scientist at FBK
The problem is whether or not such an approach is affordable
CREATE-NET, Italy. He has worked in the telecom R&D environ-
given shorter-term pressures afflicting the bottom-line of these ment since the beginning of his career, focusing in the last decade
businesses. So much for big corporations only parties. on IoT and related technology transfer activities. In his role, he is
What’s looming in the world of IoT outside this cyclopean happy now responsible for setting research and innovation directions,
island is much more complex and fragmented. High-margin business acquisition of funding, and the execution of a number of collab-
orative projects in the IoT domain. He worked in Italy and in the
opportunities for small to medium sized companies have yet to United Kingdom (10 years), acquiring experience in both corpo-
materialize. On the consumer front IoT is not yet considered as cool rate telco environments (R&D of BT and Telecom Italia) as well as in a small research
as the portable sleek gadgets we are accustomed to. Consider for organization (CREATE-NET before its merger with FBK), where the ability to acquire
example our smartphones. The ratio between the value (“too low”) funding was key to ensuring continuity of operations. He is a recognized expert with
you get for your investment and the costs (“too high”) is still minimal a substantial record of IEEE publications and conference presentations, a patent, and
various book chapters and tutorials on IoT. He is an experienced speaker and chair of
and the advantages IoT brings are not tangible enough to ensure IoT related events, serves as an EU reviewer, has served on the TPCs of a number of
the emergence of other sustainable business happy islands. Take the international conferences, and he is the Editor-in-Chief of the IEEE IoT Newsletter.
example of smart-
Peak of Inflated Expectations GARTNER HYPE CYCLE homes: IoT devices FOOTNOTES
provide support for 1 The Gartner Hype Cycle: https://www.gartner.com/technology/research/meth-
automation, helping 2
odologies/hype-cycle.jsp
https://iot.ieee.org/newsletter/september-2016/the-iot4sdg-challenge.html
to save time and 3 https://iot.ieee.org/newsletter/july-2015/iot-for-development-iot4d.html
perhaps money in 4 https://iot.ieee.org/newsletter/july-2017/ipv6-and-internet-of-things-prospects-
managing applianc- for-latin-america
es and routine jobs 5 https://iot.ieee.org/newsletter/january-2018/smart-city-application-enable-

Trough of Disillusionment from a distance. But ment-platform

6 https://iot.ieee.org/newsletter/january-2017/internet-of-things-for-buildings-that-
what is the price make-life-safe-and-secure.html
tag consumers 7 https://iot.ieee.org/newsletter/march-2015/connected-versus-intelligent-devic-

Figure 1. IoT in the Gartner Hype Cycle. would be willing to es-in-the-iot-and-in-saunas.html

12 IEEE Internet of Things Magazine • September 2018

 1988-2018 - Celebrating 30 years of Innovation and Service

EMPOWER YOURSELF!
With Global Vision, Knowledge and Networking
as a Member of the IEEE Power Electronics Society.

TECHNICAL COMMUNITIES

PUBLICATIONS

CONFERENCES

Powering the Past,

STANDARDS
Energizing the Future

EDUCATION

STUDENT ACTIVITIES & MENTORING

Join now at ieee-pels.org

Be connected to the ultimate source of power in your industry!

IENYIOT0004.indd 1 25/08/18 12:03 AM

 Why Industry Needs to
Accelerate IoT Standards
Maciej Kranz, Cisco Systems

Abstract
We review the fundamental problems of standardization for data transfer and security. With those basics established, we move
on to formulate key factors for the continued growth of IoT. Then we review a general approach for implementing IoT in a business,
and describe four recognized Fast Paths for IoT payback. We focus on the Remote Operations Fast Path, and specifically on its
implementation in the mining industry by Goldcorp, a major mining company. We present the business case for IoT at Goldcorp,
and then describe the benefits reaped by its implementation.

Introduction organizations are working on improving IoT security. IEEE, for

example, has been at the forefront of teaching security guide-
The general public may view the Internet of Things (IoT) as lines and best practices for IoT [5] and has numerous standards
nothing more than a set of gadgets that can let you turn on the related to IoT security [6] IEEE’s P2413 “Standard for an Archi-
furnace in your house or track how far you jogged, but such tectural Framework for the Internet of Things (IoT)” Project has
uses obscure the significant financial benefits that IoT can offer a sub-working group focused on Quadruple Trust: protection,
a business. Indeed, Bain predicts that business-to-business IoT security, privacy, and safety [6].
will generate $300 billion annually by 2020, more than twice In terms of the standardization of device connection and
their prediction for consumer applications over the same time data transfer, because IoT systems must carry information
frame [1]. Rockwell Automation provides one example of the from a very specific application, such as from a tire on an
benefits IoT can offer a business. Within a handful of years of earthmover in a mine all the way to the cloud and then on
Rockwell’s implementation of IoT in 20 plants worldwide, it to a control system, IoT systems are inherently vertical. More-
reduced the number of rejected parts by 50 percent, cut its over, each application will have its own requirements for items
inventory cycle by one-third, and raised its on-time delivery rate such as data collected, data privacy, and security. As a result,
from 82 percent to 98 percent [2]. Similarly, Harley-Davidson’s IoT is a very fragmented market with a plethora of specialized
use of IoT in one factory led to a decrease in build-to-order proprietary and semi-proprietary systems and technologies.
cycle time from 18 months to two weeks and reduced oper- The vertical nature of IoT leads to an array of different stan-
ating costs by $200 million [3]. General Motors used IoT and dards, which means that each step in the data transfer may
predictive maintenance on its assembly line robots to prevent require the additional expense of a translation, somewhat like
approximately 100 line stoppages over a three-year period. trying to pass a message across Switzerland and having to
With each line stoppage costing GM approximately $2 million, pay a translator every time you enter a region that speaks a
the financial benefits of IoT quickly become apparent. I could different language. The challenges and costs created by the
cite a long list of such success stories, and they would all share lack of standardization point to both the need for a strong
many of the same attributes, such as a focus on solving an exist- business case before a company can successfully implement
ing business problem, the formation of a partner ecosystem, IoT as well as the need for the IoT ecosystem to develop uni-
and the existence of a secure data processing infrastructure. versal standards to connect with each other. IEEE has several
And while each of these IoT implementations no doubt faced IoT standardization efforts making progress, including their
technical problems for their specific business problem, they also P2413 Project, TSN (Time-Sensitive Networking) Task Group,
had to deal with two fundamental challenges that continue to and PAR 2510 (Standard for Establishing Quality of Data Sen-
limit broad implementation of IoT: data security and a lack of sor Parameters in the Internet of Things Environment)). IEEE’s
standards, especially related to data tagging and transfer. P2413 Project is noteworthy in that along with defining the
To understand both the benefits and the fundamental chal- relationships among various IoT verticals and common archi-
lenges of IoT, it is necessary to view an IoT system at its most tecture elements, it also identifies planned or ongoing projects
basic level. All IoT uses start with sensors, actuators, or other with a similar or overlapping scope [7].
end-devices that can connect to remote processors. Depending In this paper we will review the fundamental problems
on the needs of the operation, such devices could be connect- of standardization for data transfer and security. With those
ed directly to the cloud or, more locally, to a fog computing basics established, we will move on to formulate key factors
system [4], depending on the amount of data that needs to be for the continued growth of IoT. Then we will review a general
transferred and processed and its time sensitivity. approach for implementing IoT in a business, and four recog-
Whether the system uses fog computing or cloud comput- nized Fast Paths for IoT payback. We will focus on the Remote
ing, devices have to connect to deliver data, which makes the Operations Fast Path, and specifically on its implementation in
IoT system vulnerable both to security breaches and to a lack the mining industry by Goldcorp, a major mining company. We
of standardization among the numerous vertical standards for will present the business case for IoT at Goldcorp, and then
device connection and for data transfer that currently exist. On describe the benefits reaped by its implementation.
the security side, the old notion of “security by obscurity,” in
which a company’s network is secure because it is not connect-
ed to the internet, does not work in an IoT system, which is by
Standardization of
definition interconnected and based on open standards. Many Security and Data Transfer
IoT security is in many ways much more complicated than tra-
Editor’s Note: Text appearing in bold indicates a live link in the online version. ditional IT security environments. For example, an IoT system is
Digital Object Identifier: 10.1109/IOTM.2018.1700011 often more distributed, more heterogeneous, and more dynam-

14 2576-3180/18/$25.00 © 2018 IEEE IEEE Internet of Things Magazine • September 2018

ic than a network of computers. Each vertical IoT market has
specific IoT security requirements and regulations. In the utilities
industry, for example, the U.S. government recently mandated
the adoption of version 5 of the North American Electric Reli-
ability Corporation Critical Infrastructure Protection (NERC CIP)
as the cybersecurity standard, whereas the healthcare industry
follows HIPPA requirements for securing data and privacy.
Another reason that IoT security is more challenging than
traditional IT security is that it typically incorporates both IT and
operational technology (OT) functions within a business. The
use of IoT leads businesses to merge IT and OT to drive new
use cases, enable an open flow of data between networks and
applications, support better business decisions, lower costs, and
reduce overall complexity. However, the merging of IT and OT Figure 1. Examples of IoT in increasing business efficiency.
often leaves gaps in security strategy that make such a com-
bined system an attractive target for cybercriminals focused on DeviceNet Vendor Association (ODVA), the Open Platform
accessing critical equipment and infrastructure. Communications Foundation (OPC), and the International
Although there is no foolproof method to guarantee 100 Society of Automation (ISA) are working to align with the
percent security of any IoT system, there are a number of key International Electrotechnical Commission’s (IEC) 62443 stan-
principles that companies can employ to manage their IoT secu- dards for network and system security for industrial automa-
rity risk. First, when implementing an IoT system, the company tion and control systems. These standards combine vertical,
should have, or create, a unified, policy-based security archi- industry-specific best practices with horizontal approaches
tecture across the system, and design security into everything to common elements like industrial security. Groups such
from the beginning. This includes assimilating all teams and as the IETF and the Industrial Internet Consortium’s security
departments, especially those that have their own tools, devic- working group have all been active in developing IoT security
es, or networks. As much as possible, the system should employ standards and methodologies to ensure cybersecurity across
industry-supported standards. One-off or proprietary standards interconnected IoT systems.
will hamstring security efforts, and lead to significant costs in Along with security concerns, a lack of clear horizontal stan-
the future. The security system should also include real-time dards combined with a plethora of vertical standards for IoT is
monitoring and leak path detection. Understanding how the IoT probably the biggest barrier to widespread IoT adoption today.
system interacts with its components is necessary, so that a red Indeed, Ahmed Banafa states in an IEEE IoT newsletter that
flag can be raised if one IoT device is scanning another device the lack of standardization will be the primary reason for slow
it does not usually scan, or an otherwise predictable traffic pat- adoption of IoT in 2018 [10]. IoT is different from the first gen-
tern changes. For instance, manufacturers can now add an extra eration of the internet in that IoT is highly verticalized, going
layer of security to their devices by using the Internet Engineer- from sensor to network to fog or to cloud (for processing) and
ing Task Force’s Manufacturer Usage Description (IETF MUD) then back to the network to the control system to take appro-
standard to “tell” the network what access the device needs. priate action. Unfortunately, there are too many vertical stan-
This allows the network to deny any anomalous requests from dards, which as in the case of industrial automation, are often
that device. The security architecture should include a strategy overlapping and conflicting. Combined with the lack of hori-
not just for preventing attacks, but also for the response during zontal standards, the confusing landscape forces a great deal
an attack, and follow-up and improvement after an attack. Part of effort to be spent on translating custom knowledge for each
of that strategy should include the use of well-established prac- IoT implementation, rather than using one universal system. As
tices such as device and traffic segmentation and of a multi- a result, a significant portion of the cost of an IoT implementa-
tenant network infrastructure to isolate problems. Finally, just tion involves integration into legacy systems and customization.
like any computer network, it is necessary to keep systems up With so many different systems, it is no surprise that most of the
to date. An automated device discovery system is necessary current ~700 IoT platforms focus on integration of proprietary
to know the operating systems of all of the devices on the IoT end-points and systems.
network, and quickly install patches to fix vulnerabilities as they While it may be tempting for an IoT supplier to create its
are discovered. own unique platform, the reality is that no one company can
While managing the security of the electronic devices and develop a complete solution by itself. There are too many possi-
associated software in an IoT system, it is also necessary for the ble places (machinery parts, buildings, engines, etc.) for sensors,
security architecture to include people. As many as 60 percent and too many platforms for processing the information from
of IoT security attacks to date originated inside the organiza- those sensors (fog, cloud, laptop, tablet, smartphone, etc.) for
tion, so taking the human element into account is critical [8]. one company to control the entire IoT system. The implemen-
Everyone in an organization’s IoT ecosystem, including employ- tation of IoT it is not merely a problem of translating among
ees, partners, and vendors, will need to be educated about different proprietary systems and protocols. In addition, each
security practices and policies. Those security policies should IoT system also has to incorporate important physical factors,
include network segmentation and role-based access controls, such as safety and motion requirements, which will vary from
such that only authorized people, machines, and/or processes implementation to implementation. Without standards, there
can access certain classes of devices or data flows. Finally, it cannot be interoperability among the different enterprises in the
is necessary to realize that IoT security is a journey, and not a IoT ecosystems. Without interoperability, benefits for businesses
one-and-done, fix-and-forget action. The organization must con- will be much harder to find, and the total IoT ecosystem for IoT
tinue to evolve its technologies and security practices as threats, suppliers will be correspondingly smaller.
and technical solutions, also evolve. Establishing a clear set of standards will benefit all. Stan-
Fortunately, there is great momentum to improve data dardization and interoperability are the gateway to IoT value.
security. In addition to the IEEE’s efforts, in February 2018 In contrast, limited standards lead to limited access to IoT
Siemens announced the formation of a consortium of large data, which limits the value of an IoT deployment. Having set
companies, including IBM, Airbus, Deutsche Telekom, and standards would enable IoT teams to focus their efforts on the
the Daimler Group, to build greater digital security [9]. Addi- most important part of IoT for a company: solving a business
tionally, manufacturing standards bodies such as the Open problem.

IEEE Internet of Things Magazine • September 2018 15

 Figure 3. Infrastructure failure at Fortune 1000 company.

journey, and not a one-time event. Along the way the company
will encounter both unexpected challenges and unforeseen
opportunities. While it may be easy to imagine an IoT utopia
at the company in the distant future, it is necessary to identify
a small area of the company as a starting point, and then focus
on solving real problems in that small area. Consider that first
IoT area to be a beachhead, from which other benefits of IoT
can take hold and spread into the rest of the company. Once a
business problem is identified, gather stats and benchmarks for
the problem, and build a business case for using IoT to solve the
problem. The business case will help the IoT team at the com-
pany to get a C-suite champion of IoT. That internal champion
can help drive any cultural or structural changes that may be
needed for a successful implementation. The implementation of
IoT will also force the company to build a new team, including
attracting and training new and existing talent. The team will
have to be a partner eco-system that will learn and co-develop
with the company as it implements IoT. Finally, security has to
be everybody’s top priority in the IoT implementation, from the
very start of the process.

Four Fast Paths to a

Positive Return on an Investment in IoT
Most of the first IoT projects fall into one of the following four
Figure 2. Four FastPaths to the Internet of Things. general categories: [2]
• Connected Operations — connect physical devices, process-
es, and workflows.
Despite myriad, complex challenges, there are many pos- • Predictive Analytics — collect and process information, and
itive signs that improved standards are evolving in horizon- immediately take best actions.
tal and vertical standards bodies and in consortia, such as • Predictive Maintenance — increase uptime and productive
in the IEEE examples mentioned above. Organizations, such hours.
as 3GPP, IPSO Alliance, ATIS, CCSA, OMA, and NIST, are • Remote Operations — monitor and control equipment at dif-
focused on sensing and communications standards [11]. Sep- ferent locations.
arately, in the automation industry in the last year, a group of The Fast Path of Connected Operations leverages con-
prominent networking and industrial companies announced nectable devices that a company already has. By connecting
progress in developing an open, unified, standards-based com- existing devices to an existing IP network, business processes
munication protocol between sensors, actuators, controllers, can be adapted to take advantage of the information that the
and cloud. IEEE’s TSN is a great example of evolving the Eth- devices provide. Rockwell Automation and Harley-Davidson,
ernet standard to include factors not usually addressed in mentioned above, are examples of Connected Operations.
computer networks, such as manufacturing motion and safety PepsiCo provides another example of Connected Operations.
requirements. Although all of this work is good, it still consists Like most established businesses, PepsiCo’s IT and OT devel-
of a number of standardization efforts working in parallel. oped somewhat independently in each of its plants, resulting
Optimally, these efforts would coalesce into one universally in a combination of limited IT and OT resources within its
accepted system. plants and an excess of server capacity, with each server han-
dling only a small portion of the production load. In essence,
General Steps Required to Implement IoT PepsiCo was spending a lot of money on an ever-expanding,
Based on the successes and challenges of IoT implementations underutilized system of servers, networking devices, and oper-
to date, we can formulate two key factors for the continued ating systems. PepsiCo suffered from a one-two punch of poor
growth of IoT. use of capital, and a lack of useful data from that capital. Pep-
1. For the IoT ecosystem as a whole: standardization, to improve siCo’s solution was to adopt a standard network infrastructure
data security, increase value, and reduce complexity. connected to virtual industrial servers and 24  7 centralized,
2. Within each company: a focus on solving existing business expert support. These changes improved the reliability of their
problems. manufacturing systems and led to a decrease in both down-
While an individual company cannot create a universal IoT time and support costs. PepsiCo’s perceived value of IoT can
standard, as called for above, it can target specific business be seen in the ~ $500 million contract PepsiCo signed with
cases, and use the following general steps when implementing HCL Technologies in 2014 to provide infrastructure manage-
IoT. First, it is important to realize that IoT implementation is a ment services for seven years [12].

16 IEEE Internet of Things Magazine • September 2018

 The Fast Path of Predictive Analytics takes remaining tag battery life on a digital display.
advantage of previous knowledge and real- The system tracks the location and movement
time data to predict trends with statistical pre- “It is essential to of both equipment and personnel in the mine,
cision. Such predictions enable plant managers while also assessing the mine’s air quality. That
and operators to take corrective actions or
make adjustments to maintain high production
resolve the data information enables Goldcorp to control ven-
tilation based on the number of people and
yields and decrease downtime. Cisco demon-
strated the benefits of Predictive Analytics in
security and privacy equipment in the mine. Dubbed “ventilation
on demand,” the IoT system sends air only to
one of its contract manufacturing plants in
Malaysia. Cisco focused on energy consump- issues. Blockchain the areas of the mine where it is needed.
Beyond ventilation on demand, Goldcorp’s
tion and set up a network of 1500 sensors IoT system has tremendous safety benefits. In
across the factory. Cisco captured the resulting is one approach the underground control room, visual maps
energy consumption data, which gave insights allow for real-time visibility of both personnel
into energy inefficiencies of machines and to secure data and vehicles at all levels underground. These
production processes. By replacing inefficient maps are key safety tools used to ensure that
equipment and adjusting their operations,
the plant reduced its energy consumption
and regulate data nobody is left underground while blasting. In
fact, Éléonore’s safety record has improved 20
by 15–20 percent. On a more general level,
it has been estimated that the average cost
access.” percent every year since the project started
four years ago. Additionally, with an estimated
of an infrastructure failure at a Fortune 1000 90 percent Wi-Fi coverage underground, the
company is $100k per hour. That cost rises to miners and their families have an extra level of
$500,000 to $1 million per hour if the failure is comfort knowing that they have been account-
in a critical application [13]. ed for, and employees in the mine can communicate with each
The Fast Path of Predictive Maintenance is similar to Pre- other, as well as with employees up on the surface.
dictive Analytics in terms of limiting unplanned downtime, but Goldcorp’s IoT system at its Éléonore mine is built on one
is focused specifically on machines and equipment that are multiservice, secure IPV6 network for wireless connectivity [2].
expensive to repair and that are in challenging environments for The network consists of 160 of Cisco’s Aironet 1500 and 3600
repair. The mining industry has benefitted significantly from Pre- series access points, as well as its 5500 series wireless local
dictive Maintenance with IoT. For example, mining equipment area network controller. Those link to Cisco’s Catalyst switches,
is typically gigantic, so that a lot of material can be transported UBR7225VXR broadband routers, and 2900 series integrated
or processed at one time. With gigantic equipment comes cor- service routers. Linking into this communications backbone
respondingly gigantic costs; each earthmover truck tire by itself are Voice-over-Internet-Protocol phones, used to communicate
costs $69,000 [14]. There are also costs in terms of the mine with underground workers, as well as battery-powered RFID
not being able to process ore while the truck is out service. The tags and telemetry units integrated into vehicles operated in the
cost of one truck having unplanned downtime is approximately mine. An RTLS tag on each of the ~600 underground workers,
$2 million per day. At Rio Tinto’s Pilbara mine in Australia, they ~200 machines, and ~200 other important mobile items trans-
automated 900 giant dump trucks with 192 sensors in every mits a unique ID number via a Wi-Fi connection to the Cisco
truck, on everything from the engine to the transmission to access points. The telemetry units are manufactured by ISSAC
those expensive tires. The IoT sensors monitor temperature in Instruments and communicate data over the Wi-Fi network
and around critical components, enabling Rio Tinto to antici- as well [15]. Goldcorp personnel can access software to gain
pate failures with 80 percent accuracy up to three months in mobility views from computers, located either inside an under-
advance [2]. ground control room or as far away as Goldcorp’s Vancouver
The Remote Operations Fast Path has numerous examples, headquarters, or via tablets that workers carry with them under-
starting with something as simple as a Nest thermostat in a ground. They can view a map of the entire mine and see the
home. By being able to control remotely whether or not a locations of employees and assets, or drill down to search for
home furnace is running, the homeowner can reduce ener- specific individuals or things.
gy costs by turning off the furnace when nobody is at home. Air quality for the automated ventilation system is moni-
Something as simple as a Nest thermostat can be expanded tored with Drager Polytron 8000 sensors wired to program-
and extrapolated to the business environment to provide a mable logic controllers. The controllers are connected to cable
rapid payback on a real business problem. In the next section modems and linked into the mine’s IT system via Cisco 7200
we will discuss in detail an example of Remote Operations in cable modem termination systems. The Drager sensors deter-
mining provided by Goldcorp, which in some ways is analogous mine the levels of ambient carbon monoxide, propane, and
to a Nest thermostat in a home writ large. nitrogen dioxide in each part of the mine. Based on this air
quality data and the location of people and the operation of
Example of vehicles, as determined by the RTLS tags, the ventilation system
determines where fresh air is required at all times.
Remote Operations Implementation of IoT In addition to their Remote Operations IoT work for their
Goldcorp is a multinational gold producer with more than ventilation on demand system, Goldcorp also leveraged their
19,000 employees. At their Éléonore gold mine 500 miles north IoT system to implement Predictive Maintenance in the Éléon-
of Montreal, miners and equipment operate in tunnels as much ore mine. It is worth zeroing in further on how Goldcorp broad-
as 4000 feet below the surface. The miners and equipment ened its focus to predictive maintenance because it is a classic
need clean air in the tunnels, but providing that air continuously illustration of discovery and scale during an IoT journey. The
is costly. The business problem that Goldcorp focused on was tags on their equipment monitor 30 different functions and
the ventilation costs in its mine. systems within each vehicle’s engine. A tag will send an alert to
Goldcorp turned to IoT, and with the help of Cisco and its the control room if any system shows an unexpected change in
partner ecosystem deployed an IoT tracking system connect- temperature or pressure. The operator in the control room can
ed to an underground Wi-Fi network. Every person entering then alert the driver to bring the truck to maintenance before
the mine has a real-time location system (RTLS) tag affixed to it breaks down. The predictive maintenance software can also
their hardhat. That tag sends their information over a Wi-Fi net- auto-generate maintenance schedules within the mine’s enter-
work to a system that displays their name, picture, location, and prise resource planning system, based on actual rather than

IEEE Internet of Things Magazine • September 2018 17

estimated usage, by logging each machine’s hours of operation. connected vertical and horizontal standards for security and
The outcome of Goldcorp’s implementation is that opera- data transfer, coupled with new technologies such as artificial
tions at its Éléonore gold mine are improving worker safety and intelligence, fog computing, and blockchain, it is easy to imag-
reducing energy costs. Its use of ventilation on demand has cut ine IoT evolving from sensor-driven projects with significant yet
its ventilation requirements in half, thereby decreasing its ener- incremental gains to even smarter outcomes with transforma-
gy costs by $1.5–$2.5 million per year. The improved tracking tional business value.
system of personnel can locate employees in the event of an
emergency 45 to 50 minutes faster than before. The benefits of References
IoT exceeded expectations to the point that Goldcorp is now [1] P. Bowen et al., “Choosing the Right Platform for the Industrial IoT,” 2017,
http://www.bain.com/publications/articles/choosing-the-right-platform-for-
working on spreading its use of IoT to other mining operations, the-industrial-iot.aspx.
and deepening its use at the Éléonore mine. Indeed, Éléonore’s [2] M. Kranz, Building the Internet of Things, 2017, Hoboken, New Jersey: John
manager predicts that “all mines of the future” will follow suit Wiley & Sons, Inc.
[16]. [3] M. Kranz, “Success with the Internet of Things Requires More Than Chasing
the Cool Factor,” 2017, https://hbr.org/2017/08/success-with-the-internet-of-
Conclusion things-requires-more-than-chasing-the-cool-factor.
[4] H. Antunes, “The Future of IoT is Clearly in the Fog,” 2017, http://www.macie-
Goldcorp’s example points both to key factors for the success- jkranz.com/future-iot-clearly-fog/.
ful implementation of IoT within a business, as well as elements [5] G. Corser et al., Internet of Things (IoT) Security Best Practices, 2017, https://
internetinitiative.ieee.org/images/files/resources/white_papers/internet_of_
of IoT that are a fundamental challenge for all businesses. Gold- things_feb2017.pdf.
corp made a good start with IoT by identifying a well-defined [6] C. Tom, “Security and IoT in IEEE Standards,” IEEE Standards University, 2016,
business problem that could be solved with IoT: reducing ven- 6, https://www.standardsuniversity.org/e-magazine/march-2016/security-and-
tilation costs. The well-defined nature of the business problem iot-in-ieee-standards/.
[7] Standard for an Architectural Framework for the Internet of Things (IoT), 2016;
made it possible for them to gather stats and benchmarks for http://grouper.ieee.org/groups/2413/Intro-to-IEEE-P2413.pdf.
the problem, both to justify the IoT project initially and verify its [8] IBM 2016 X-Force Research Cyber Security Intelligence Index, 2016. http://
performance after implementation. ibm.biz/2016CyberIndex.
At the same time, the Goldcorp example demonstrates [9] D. Takahashi, “Siemens Builds Consortium of Powerhouses for A Community
of Cyber Trust,” Venture Beat, 2018. https://venturebeat.com/2018/02/16/
that IoT is highly verticalized, and therefore different from siemens-builds-consortium-of-powerhouses-for-a-community-of-cyber-trust/.
previous IT efforts like installing a computer network. Verti- [10] A. Banafa, “Eight Trends of the Internet of Things in 2018,” IEEE Internet
calization can be seen in the case of Goldcorp’s Predictive of Things Newsletter, 2018, https://iot.ieee.org/newsletter/january-2018/
Maintenance effort, where different types of knowledge are eight-trends-of-the-internet-of-things-in-2018.html?highlight=WyJzdGFuZGFyZ-
CIsInN0YW5kYXJkcyIsInN0YW5kYXJkaXphdGlvbiIsInN0YW5kYXJkaXplZCIs-
needed at every node in the chain, from knowing which parts InN0YW5kYXJkaXplIl0=.
of their equipment need to be monitored, to how to transfer [11] E. Borgia, “Computer Communications,” 2014, vol. 54, pp. 1–31.
that information to the fog or the cloud, to how to assess [12] C. T. N. Desk, “HCL Wins $500 mn PepsiCo Deal over Rivals,” 2014, http://
that information and respond. No one company can efficient- www.cxotoday.com/story/hcl-wins-500-mn-pepsico-deal-over-rivals/.
[13] A. Shimel, “The Real Cost of Downtime,” DevOps.com, 2015. http://
ly sustain an expertise in all of those nodes simultaneously, devops.com/2015/02/11/real-cost-downtime/.
especially across several different applications. Goldcorp’s [14] J. Scholze, “Mining Industry Strategically Protects Assets with IoT,” 2017,
comprehensive partner ecosystem working together efficient- https://blogs.sap.com/2017/08/25/mining-industry-strategically-protects-as-
ly enabled operators to address all of the vertical levels of sets-with-iot/.
[15] M. C. O’Connor, “At Goldcorp’s Éléonore Mine, the IoT Is Worth Gold,” IOT
an IoT system. However, because of the lack of standardiza- Journal, 2015, http://www.iotjournal.com/articles/view?12790.
tion, Goldcorp was forced to use a proprietary system for [16] J. Perkins, “Internet of Things Improving Goldcorp Operations,” Mining
their RTLS tags. A proprietary system inherently entails more Innovation News, 2015, https://republicofmining.com/2015/03/13/inter-
costs in terms of setup, data translation, and training than a net-of-things-improving-goldcorp-operations-by-james-perkins-mining-innova-
tion-news-march-12-2015/.
standardized system. Therefore, there is a tremendous need
for standardization for each step of the way: standards for Biography
sensors, for data flow, for enterprise infrastructure, and for Maciej Kranz brings 30 years of networking industry experi-
fog and cloud computing, and overarching all of that, for data ence to his position as Vice President, Strategic Innovation
security. Without standardization, companies trying to use IoT Group at Cisco. He leads the team focused on incubating
to improve their business will continue to pay a significant, new businesses in security, blockchain and IoT, accelerating
internal innovation, and driving co-innovation with customers
additional amount for translating between different systems and partners through a network of Cisco Innovation Centers.
and protocols. Standardization will progress the IoT ecosys- Prior to this role, Kranz was General Manager of the Connect-
tem into a secure, universal system for things communicating ed Industries Group at Cisco, a business unit focused on the
with things. Internet of Things (IoT). He built a $250M business from the
ground up in 18 months and relentlessly evangelized the IoT opportunity across
The Goldcorp example also points to the evolving nature Cisco and the market, making IoT one of Cisco’s major priorities. Previously, Kranz
of IoT. IoT is indeed a journey. The initial use case improved drove business and product strategy for the wireless and mobility business and
operations, employee safety, and financial performance, but led product management for the stackable Ethernet switching business through
that case quickly led to other uses of IoT, such as predictive its expansion from $400M to $6B in revenues. In his New York Times Best Selling
book, Building the Internet of Things, and its companion Project Workbook, Kranz
maintenance. These different business cases for the use of IoT offers practical advice to business decision makers on how and why to implement
are the stepping stones to disruptive business models and value IoT today. He formed a thought-leadership forum and newsletter to exchange
propositions. IoT will continue to grow in terms of the number industry insights on IoT developments. Kranz is also a faculty member of Singulari-
of applications, and how it is used. With the formation of inter- ty University, focusing on IoT and corporate innovations.

18 IEEE Internet of Things Magazine • September 2018

 BECAUSE THE
FUTURE
IS NOW

OVER
200
INTERNATIONAL
CHAPTERS

OVER
Join the IEEE Robotics & Automation Society (RAS) today. 40
TECHNICAL
COMMUNITIES

Exchange and engage with a world community.

RAS lets you collaborate with peers across the globe, grow in your career, and
OVER
be on the cutting edge of technological advances in this fascinating field.

Now you can:

13,000
MEMBERS
WORLDWIDE
· Gain access to leading industry publications and resources like IEEE Xplore®
· Enjoy discounts on publications and conferences
· Participate in student and Young Professionals activities to connect
with peers and learn new skills
· Engage in conferences and workshops, including expert
panels, tutorial sessions, and more

To become a member…
Go to www.ieee-ras.org and click on the JOIN RAS button
in the upper-right-hand corner. Or simply scan our QR code now.

IENYIOT0005.indd 1 25/08/18 12:05 AM

 Drilling Deep into Digital Industrial
Transformation Will Determine Who
Survives and Thrives
William Ruh, GE Digital

M ore than one-third of businesses today will fail in the

next 10 years. John Chambers, longtime CEO of
Cisco, made that prediction just before his retirement
last year. He further stated [1], “The ones that survive will turn
their companies into digital, techie versions of themselves, and
many will fail trying.”
same year Gartner predicts that 80 percent [3] of today’s busi-
ness products or processes will be reinvented, eliminated or
digitized.
Clearly, in the minds of many, we are steamrolling toward
a critical inflection point that may determine the future of busi-
ness. That is especially true in manufacturing, power generation,
Whether or not John’s prediction comes true, it is grounded oil and gas, healthcare and other critical industries, where sys-
in one indisputable truth: today’s most successful companies tem failures and lengthy downtime can result in high-risk, high-
are unlocking the mysteries of data to unveil valuable insights cost, or safety situations. Take Gerdau, for example, the largest
that spawn powerful business outcomes. producer of long steel in the Americas. When the company
At GE, where we coined the term “Industrial Internet” in experiences an asset failure, they lose an average of $12,000
2012, we estimate the market for the Industrial Internet of per hour in downtime [4].
Things (IIoT) could be $225 billion by 2020 [2]. That is the Based on the results of a study [5] by Kimberlite, a market
research and analytics firm serving the oil and gas industry,
Editor’s Note: Text appearing in bold indicates a live link in the online version. offshore operators incur an average financial impact of $49
Digital Object Identifier: 10.1109/IOTM.2018.1700031 million because of unplanned downtime. However, for those

20 2576-3180/18/$25.00 © 2018 IEEE IEEE Internet of Things Magazine • September 2018

operators who employ analytics-driven predictive
maintenance techniques, unplanned downtime is,
on average, 36 percent less, a savings of about $17
million annually.
In the case [6] of one of the world’s largest
exploration and production companies, the sav-
ings was $20 million. Using one of the industry’s
premier predictive analytics solutions, engineers
detected signs of machine stress on vital equipment
12 months before scheduled maintenance. Clos-
er inspection revealed the cause of the problem,
which was repaired during planned downtime. The
company estimates the early detection, and result-
ing proactive maintenance, enabled it to avoid cost-
ly lost production time.
Nevertheless, for most companies, the road to Noble Globetrotter I drillship. Credit: Noble Corporation.
industrial digital transformation remains long. While
80 percent of leaders GE surveyed [7] believe that
the IIoT will or could be transformational to business, only about digital twin, along with advanced analytics to provide a holistic
8 percent [8] say their organizations have actually embarked on view of a vessel’s entire health state and real-time equipment
a digital transformation journey. performance.
We should not be surprised by this disparity. Transformation- The result? The detection of multiple anomalies on major
al change can be arduous, especially for industrial companies marine and drilling equipment that could have led to potential
facing numerous challenges: cost volatility and price pressure, failures, up to two months before they would occur, prompting
shrinking profit margins, growing environmental scrutiny and restorative maintenance before a problem strikes. Examples of
compressed cycle times, among others. these anomaly warning systems include the ability to automat-
All of these challenges are important. And most, if not all, ically track drilling processes to minimize process deviations
can be addressed — even solved — by reimagining how to use and non-productive time, thereby enhancing drilling efficiency
data and advanced analytics to truly reimagine plant operations, and predictability. Furthermore, early automatic warnings on
product manufacturing and customer experiences. equipment degradation leads to faster corrective actions and
Still, envisioning and then strategically deploying an IIoT reduced unplanned downtime. The reduction in unplanned
framework requires a significant mindset change for organi- downtime drives higher revenues and saves up to 20 percent in
zations. Cultural shifts do not happen overnight, and massive operational expenditures on targeted equipment.
companies are not always accustomed to agility and flexibility, Equally important, offshore personnel can focus their atten-
in thinking or in systems. tion on strategic activities, rather than time-consuming and
But industrial companies must move boldly, because they resource-consuming “fire drills.” This is particularly important
need digital transformations to survive and thrive. And they in an age where the total cost of manufacturing and industrial
don’t need to wait. The Industrial Internet is off and running. skills shortages has been reported to be as much as 11 percent
of net earnings [10].
D D ,M
rilling igitally iles and iles rom horeM F S In announcing the groundbreaking news, Noble SVP of
From the frigid waters of the North Sea to the stifling heat in Operations Bernie Wolford stated [11] that “digital technol-
the Gulf of Mexico, Noble Corporation drilling rigs are central ogy is facilitating a new era of drilling and asset performance
to offshore oil and gas exploration all over the world. With com- improvements that are unprecedented. The potential of digi-
plex machinery plumbing the sea beds from sky-high platforms, talization will go beyond a single vessel, opening the door to
miles from land, drilling is a challenging business encumbered transforming our entire fleet.”
by high costs and worker safety risks. The Industrial Internet is
making it better. II T: A D
o B B
igital lueprint for usiness uccess S
In 2017, Noble launched the world’s first digital drilling Bernie’s vision of the future state is not hyperbolic. As numer-
vessel [9], an important step toward a future of autonomous ous use cases are confirming, the Industrial Internet provides a
drilling and machine operations in one of the world’s most chal- digital blueprint for business success. It gives authority to data
lenging industries. While teams of highly-trained workers are for informing short-cycle feedback loops. A successful build-
the backbone of any drilling operation, harsh conditions and out of industrial applications requires careful examination of a
technical complexities heighten the possibility of human error. company’s entire system of assets, from the edge to the cloud
Powered by the world’s first operating system for the Indus- and everything in between, to create truly intelligent systems for
trial Internet, GE’s Predix platform, the digital rig has converted success.
the Noble Globetrotter I drilling vessel into a model of IIoT Consider the case [12] of a global elevator manufacturer,
efficiency. With all systems integrated, data from targeted con- which enabled its service organization to evolve from a typical
trol systems, including the drilling control network, the power cost center to a revenue-generating and revenue-protecting
management system and the dynamic positioning system, asset.
is collected through individual sensors and control systems, By implementing a holistic predictive maintenance initiative,
then harmonized and centralized on the vessel before being which enabled a greater percentage of operational problems
transmitted in near-real time to GE’s Industrial Performance & to be avoided before they impacted elevator performance, the
Reliability Center for predictive analytics. Supported by these company improved its first-time fix (FTF) rate from 65 to more
intelligence systems, personnel on the vessel or onshore can than 90 percent, in the process avoiding countless customer
gain a holistic view of the entire vessel’s health state and real- calls and service tickets for repeat calls. The key was to couple
time performance of each piece of equipment onboard. This its IIoT deployment with an asset performance management
shift to data-driven decisions will have a significant effect on solution that delivered targeted inspection and maintenance
drilling efficiencies. strategies for the equipment, including performance, mainte-
A key feature of the Predix platform is the ability to combine nance criticality, timing and suggested service components.
data models from a digital replica of physical assets, known as a Delivering proactive, cost-effective service can be done.

IEEE Internet of Things Magazine • September 2018 21

The key is enabling central collection and anal- [2] GE Digital, Everything You Need to Know About the
Industrial Internet of Things, website, https://www.ge.com/
ysis of data constantly flowing from sensors
installed on an asset, and then using it to drive “As these examples digital/blog/everything-you-need-know-about-industrial-in-
ternet-things.
intelligent, predictive decisions. As the Noble [3] Forbes, “Gartner Predicts Three Big Data Trends
example illustrates, having the capability to run
analytics on the equipment’s real-time data —
demonstrate, it is for Business Intelligence,” Feb. 12, 2015, https://
www.forbes.com/sites/gartnergroup/2015/02/12/gart-
ner-predicts-three-big-data-trends-for-business-intelli-
on the edge, where it is generated — enables
personnel to localize performance improve-
one thing to have gence/#3c77492b6de4.
[4] GE Digital, Gerdau Saves Millions Annually with GE Dig-
ital, website, https://www.ge.com/digital/sites/default/files/
ments and asset intelligence, further saving
time and money. Doing so requires smart soft- data. It is quite gerdau-saves-millions-annually-with-ge-digital.pdf.
[5] GE Digital, Study: Digital Helps Lower Unplanned Down-
ware to filter and analyze the massive volume
of data being ingested into the system and rec- another to use it time in Oil & Gas, website, https://www.ge.com/digital/
blog/study-digital-helps-lower-unplanned-downtime-oil-gas.
[6] GE Digital, Global Oil & Gas Company Saves Millions
ommending the most optimal actions. For oil
and gas, this applies to such critical factors as effectively in the of Dollars by Replacing Turbine 12 Months Ahead of
Schedule, website, https://www.ge.com/digital/stories/
well flow rates and the useful life of wells. global-oil-gas-company-saves-millions-dollars-replacing-tur-
As these examples demonstrate, it is one
thing to have data. It is quite another to use
quest for continu- bine-12-months-ahead-schedule.
[7] GE Digital Industrial Evolution Index, Executive Summa-
ry, Oct. 2017, https://www.ge.com/digital/sites/default/
it effectively in the quest for continuous oper-
ations. By leveraging advanced analytics to
ous operations.” files/GE-Digital-Industrial-Evolution-Index-Executive-Summa-
ry.pdf.
provide real-time operational intelligence, the [8] Ibid.
engineers responsible for making the best [9] GE and Noble Link up the World’s First Digital Rig, Mov-
ing towards Digitizing Marine Operations Excellence and
decisions have, at last, all the information they Analytics Based on Predictive Models, Feb. 22, 2018, press release, https://
need at their fingertips. Noble’s initial integration has opened www.genewsroom.com/press-releases/ge-and-noble-link-world%E2%80%99s-
the door to transform three additional drilling rigs in the coming first-digital-rig-moving-towards-digitizing-marine.
months, and eventually all of Noble’s fleet, paving a pathway to [10] Manufacturing Institute, 2014 Manufacturing Skills & Training Study, website,
http://www.themanufacturinginstitute.org/Research/Skills-and-Training-Study/
autonomous drilling and more first-of-its-kind asset performance Skills-and-Training-Study.aspx.
improvements. [11] Op. cit. GE and Noble.
Industrial digital transformation marks the beginning of an [12] ServiceMax, The Future of Service is Predictive and Equipment-Centric. Are
ongoing revolution away from business-as-usual. Its continuous You Ready?, May 31, 2018, blog post, http://fsd.servicemax.com/blog/the-
future-of-service-is-predictive-and-equipment-centric-are-you-ready/.
improvement, agility, automation and transparency is making
organizations stronger, with better products and more efficient Biography
operations. And it can ensure your spot in the two-thirds of
William Ruh is Chief Digital Officer of GE and Chief Executive
businesses that will survive and thrive 10 years from now, and Officer of GE Digital, which is the leading software company
beyond. for the Industrial Internet, reimagining industry’s infrastruc-
ture by connecting software, apps and analytics to industrial
References businesses.
[1] Business Insider, “Retiring Cisco CEO Delivers Dire Prediction: 40% of Com-
panies Will Be Dead in 10 Years,” June 8, 2015, http://www.businessinsider.
com/chambers-40-of-companies-are-dying-2015-6.

22 IEEE Internet of Things Magazine • September 2018

 The Internet of Things Meets
The Internet of Space
20-21 January 2019 // Orlando, Florida

2ND IEEE INTERNET OF THINGS

VERTICAL AND TOPICAL SUMMIT
AT RADIO WIRELESS WEEK (RWW2019)
CALL FOR PARTICIPATION

The IEEE Internet of Things Vertical and Topical Summit will be part of
RWW2019 in Orlando, Florida on January 20-21, 2019. The theme for
the Summit is “The Internet of Things (IoT) Meets the Internet of Space
(IoS)” and participants will explore the essential role that Space-Based
Communications and Sensor Systems play in making IoT a reality.

Don’t miss this unique opportunity to gather with

colleagues to share experiences and knowledge
about IoT and IoS.

Visit http://rww2019.iot.ieee.org/
for more information and to register.

IENYIOT0007.indd 1 04/10/18 4:22 PM

 Connecting the Dots on
IoT for the Industrial World
Karine Lavoie-Tremblay, Pratt & Whitney

Abstract
Today, businesses in every industry are under increased pressure to deliver outcomes. Whether you sell hamburgers or elevators,
it is not enough to simply ship a great product. The new onus is to ship a great experience. Owing to the rapid commodification of
sensors and the advent of cutting-edge data analytics, heavy equipment manufacturers have arrived at the dawn of a new industrial
revolution. Whether this tidal shift spells boom or bust for organizations like UTC is entirely up to us.

I t was October 1948 when the sibling proprietors of a small

California food stand noticed an anomaly in their books.
Though their menu had swelled over time to include every-
thing from hot dogs to orange juice, a single offering seemed
to account for the vast majority of the restaurant’s profits: the
hamburger. In short order, the brothers pared the menu down
The vast datasets captured by IIoT systems have applica-
tions well beyond proactive asset maintenance and automated
field service management. For industrial OEMs themselves, IIoT
affords an unprecedented, real-time window into how equip-
ment operates, not in the carefully controlled environment of
a testing facility, but on the ground, within the customer’s four
to a handful of items before turning their eye to operations. walls. This data will have a profound impact on the product
They redesigned the kitchen in the image of Henry Ford’s development lifecycle. OEMs will be able to step up design and
assembly line, vastly expediting burger production, while also, manufacturing, creating better machines, quicker and at lesser
crucially, standardizing it. cost. Asset data may even sow the seeds of brand new product
As a result, when their customers lined up, they were guar- lines and services.
anteed not only a short wait, but a meal that exactly met their All this will not be achieved overnight. It doesn’t need to
expectations. Through data and technology, the brothers had be. These are economies of massive scale: even as progress is
ensured speed, quality and consistency. They were no longer a iterative, gains will be exponential. Moving the needle forward
simple fast food stand. They were now in the business of deliv- just one percent can represent millions of dollars in added pro-
ering outcomes. The brothers settled on a new name that year: duction or mitigated maintenance costs. But by the same token,
McDonald’s. enterprises must begin investing in IIoT today. Our collective
Today, businesses in every industry are under increased pres- evolution will no longer be measured in months and years, but
sure to deliver outcomes. Whether you sell hamburgers or ele- in days and weeks. The divide between tomorrow’s innovators
vators, it is not enough to simply ship a great product. The new and yesterday’s forgotten brands will not be a stone’s throw,
onus is to ship a great experience. At United Technologies, we but a gaping chasm.
build machines that power the world’s most critical industries.
From jet engines to climate control systems, our customers
need to know that the product they buy today will continue
EngineWise:
to perform over the course of decades. They want reliability, Rethinking Airline Fleet Management
but without the inefficiency and waste of time-based mainte- It is important to note that this IIoT renaissance is not taking
nance. They want maximized productivity, but not at the cost place in a vacuum. To the contrary, this technology is in many
of unscheduled downtime. They want answers before they ask ways a direct response to the very real market challenges facing
questions. This requires tremendous speed and agility. organizations across virtually every industry and sector. In the
Fortunately, the tools at our disposal are significantly more airline business, for instance, operators are constantly looking
powerful than the McDonald brothers’ analog accounting for new ways to achieve more with less, balancing surging fuel
records. Owing to the rapid commodification of sensors and costs, rising customer expectations, the emergence of budget
the advent of cutting-edge data analytics, heavy equipment carriers and a bevy of other factors. At the same time, service
manufacturers have arrived at the dawn of a new industrial disruptions present ever greater challenges for airlines. In the
revolution. Whether this tidal shift spells boom or bust for orga- age of social media, even minor delays can quickly snowball
nizations like UTC is entirely up to us. into a Twitterstorm of reputational liability.
Make no mistake: we have entered a new era of asset intelli- To address these and other challenges, UTC’s Pratt & Whit-
gence, powered by what can only be characterized as an Indus- ney subsidiary introduced their EngineWise™ services brand in
trial Internet of Things (IIoT). This new technology is connecting 2017. The advantage of EngineWise comes in Pratt & Whitney’s
equipment in a way never before imagined. Where the last commitment to advance and integrate their engine expertise
generation gathered asset telemetry by hand onto paper clip- and fleet intelligence into service offerings that allow custom-
boards, Generation Y will have it delivered in real-time, stream- ers to optimize engine performance and keep their business-
ing by the petabyte. Where yesterday’s technicians were sent es running smoothly and efficiently. Through the EngineWise
out by human dispatchers to repair broken-down machines, portfolio, we are investing in data acquisition and predictive
today’s maintenance workers are dispatched by automated analytics tools that enable both operators and Pratt & Whitney
systems, arriving at the job site to avert shutdowns before they to make recommendations that reduce operational disruptions
ever occur. Where warranties were once a vague promise, and increase engine availability.
today’s guarantee could not be clearer: 99.9 percent reliability. Over the next five years, UTC will invest another $300 mil-
lion across its subsidiary companies and leverage the power
Contributors to this article include: Sudeep Gautam, Vice President & Chief Digital Officer, Pratt of information technology and software analytics to differenti-
& Whitney; Gabe Levine, Software Engineer, United Technologies Digital ate our products and services to run the enterprise more effi-
Digital Object Identifier: 10.1109/IOTM.2018.1700041 ciently. This will further build out IIoT offerings like EngineWise,

24 2576-3180/18/$25.00 © 2018 IEEE IEEE Internet of Things Magazine • September 2018

strengthening the value proposition of our ser- their operations. In the near future, we aim to
vice brands, while enabling the implementa- deploy eFAST across additional platforms.
tion of digital solutions and the optimization “The key to doing For carriers, this degree of configuration
of business processes for our customers. While management flexibility enables them to find
the technology behind these solutions (which
we will explore in greater depth below) rep-
this well is building answers to very specific questions that can
often yield operational insights and new effi-
resents some of the most compelling work
currently underway in the industrial world, it
robust feedback ciencies. For engineers, it allows them to inter-
act with and get to know their fleet in a way
is important to recognize the human element
that will play a pivotal role in driving IIoT adop- loops that gather previously not possible, while cutting down
on requests for manual downloads from cus-
tion in the years to come. tomers.
Our engine health management (EHM) the right informa- While this is exciting, getting access to
services manage a tremendous quantity of more data is only a part of the puzzle. In order
data that cannot always be leveraged through tion, in the right to extract maximum value both for our own
intuition alone. Customers must be trained to business as well as our customers’, we must
translate it into insight, and apply it to inform
smarter business decisions. It is our belief that
form, in the right connect the insights derived through data col-
lection across the entire product lifecycle, from
this burden rests squarely on the shoulders of
OEMs. In 2016 alone, we provided more than
context.” design and fabrication to flight and mainte-
nance and, finally, to end-of-life.
140,000 hours of training across three state- In an era of intense competition, tightening
of-the-art facilities in the U.S., India and China. margins and rapidly evolving customer expec-
We are currently developing immersive, vir- tations, this technology is perhaps the surest
tual-reality tools to offer effective training for users in remote way to carve out a competitive advantage in an increasingly
locations. crowded market.
One of the major cornerstones of the EngineWise portfolio
is the eFAST™ (enhanced flight-data acquisition storage and Translating Data into Insight
transmission) eco-system. Let’s take a look at how it works. While the ability to cost-effectively capture and transmit full
flight data is an exciting advance for aviation, equally exciting
eFAST: Full Flight Data Ecosystem is what happens on the backend. That’s where physics-based
It is important to note that UTC and its subsidiaries have long modeling and domain expertise combine with machine learning
understood the importance of data collection and are well and other cutting-edge techniques to yield valuable insights that
equipped with the proper tools to capture this critical infor- improve engine reliability and optimize operations.
mation. In fact, the subsidiary that makes up the other half of Since the 1970s, airlines and OEMs primarily performed ana-
UTC’s aerospace portfolio, UTC Aerospace Systems, is one of lytics through a process of normalization. We looked at the way
the industry leaders in terms of sensor hardware and has more an engine performed in the real world against what was expect-
than 1,000 sensors on the Boeing 787 alone. Generally speak- ed, and then accounted for factors like ambient conditions and
ing, a modern aircraft comprises hundreds of distinct systems, operating environment to generate comparable datasets. This
each with dedicated sensors feeding telemetry to a data bus. is how telemetry from a flight to Alaska can be compared to
Pratt & Whitney recognized the opportunity to leverage new that of a flight to Arizona, despite operating in radically different
digital technology to enhance our EHM capabilities and invest- conditions. By taking variables like air temperature out of the
ed significantly over the last decade in expanding our expertise equation, we’re better able to get to the root cause of engine
in data acquisition, transmission and storage. The end result is a performance issues.
system called eFAST. Over the last 10 years, as techniques like machine learning
The eFAST ecosystem is designed for large commercial have matured, OEMs have been able to complement this tra-
engines and enables Pratt & Whitney’s next generation EHM ditional physics-based approach with mathematical represen-
capabilities. It encompasses a highly secured acquisition, stor- tations. This allows us to bring in a much broader spectrum of
age and transmission infrastructure capable of accessing and data, such as air quality information from NASA and externally
recording aircraft and engine full-flight data. It generates reports sourced route data. Combined with full-flight data supplied
based on recorded data and offloads them to a remote ground through eFAST, our system can produce much more accurate
station upon landing. One of the key benefits of this system recommendations.
is the flexibility and control the operator has in sharing data The key to doing this well is building robust feedback loops
access to other parties. The system also offers sophisticated that gather the right information, in the right form, in the right
configuration management capabilities and enhanced visualiza- context. For instance, we want to look at a plane’s component
tion tools, tailored to the customer’s operational needs. change history so that if Pratt & Whitney recommends swap-
The hardware component of eFAST includes a small avi- ping out a part prior to its normal interval in order to better
onic device that can reside anywhere on the aircraft, but is manage engine service intervals; the algorithm can learn wheth-
commonly found in the avionics bay. The system has access er the recommendation achieves the desired maintenance ser-
to 100,000 engine and aircraft parameters and is capable of vice intervals or not. Over time, as the algorithm learns which
recording 6,000 parameters on a continuous basis. Significantly, recommendations provide the best benefit for the operator’s
eFAST provides customers advanced configuration manage- fleet, the overall accuracy of the predictions will improve dra-
ment options without requiring physical access to the aircraft. matically.
Data can be sent via Wi-Fi, cellular or ACARS, allowing custom- Taking this a step further, we are looking to speed up how
ers to better manage the cost of data transmission. quickly our prediction engine is able to explore new avenues of
The ability to access full flight data has also enabled the cre- correlation. For example, if an operator today wants to account
ation of our most advanced set of data analytics, vastly improv- for all the dust or snow an aircraft has ever flown through to
ing our ability to detect and prevent maintenance events. We see whether it is an indicator of an engine problem, they would
are now able to provide critical services in a matter of hours need to add that in manually. While the actual data analysis can
and days, rather than months and years. Our customers have be completed in a matter of weeks, gathering the data itself can
already benefited from this technology and are continuously take months. This is a process we hope to automate within the
exploring new ways to leverage this ecosystem to streamline near future.

IEEE Internet of Things Magazine • September 2018 25

 Much of this capability is available through our ADEM prod- and things, understanding a little more each day. What might
uct, one of the pillars of the EngineWise portfolio. It employs a we learn about ourselves, about our relationship to our sur-
suite of web-enabled software tools to provide expert analysis roundings? How can we apply that data to urban planning,
of engine health data for more than 8,000 engines in service. market research, social services? As IIoT systems continue
For Geared Turbofan™ (GTF) fleets, the GTF engine incorpo- to intersect in new and unexpected ways, the possibilities
rates 40 percent more sensors than the V2500, and can gener- become endless.
ate approximately four million data points per engine per flight, Let’s not be afraid to dream big and swing for the fences.
enabling significant improvements in addressing unplanned
maintenance. This additional data enables us to better identify Biographies
any adverse events affecting performance. Karine Lavoie-Tremblay is the Associate Director for Engine
Health Management, Commercial Aftermarket, Pratt & Whit-
The Shape of (IO)Things to Come ney. She has over 20 years of experience in engineering,
industrial and program management. In her current role, she is
For anyone who has been around heavy industry long enough, responsible for the PW Engine Health Management services,
it is tempting to look back just five or 10 years and marvel at including product and collaboration strategies for the next
how far we have come. Sensors and software have not only generation of services. She received her Bachelor of Science
Degree in mechanical engineering from Universite Laval in
redefined the way we do business, but the very nature of who Quebec, Canada, and her Master Degree in mechanical engi-
we are and what we offer our customers. Still, context is illumi- neering from the Rensselaer Polytechnic Institute in Hartford, Connecticut. She
nating. As much as we have achieved, it is important to step also received an MBA from the Rensselaer Polytechnic Institute in Hartford.
back and take stock of how nascent all of this really is, and how
Sudeep Gautam is responsible for the enterprise-wide digital
much work remains to be done. technology team and strategy at Pratt & Whitney, a world
At Pratt & Whitney, we have connected 8,000 engines so far leader in the design, manufacture and service of aircraft
and have already learned more about the business and science engines and auxiliary power units. He leads the implementa-
of aviation than we could have possibly imagined. What hap- tion of digital strategies worldwide to leverage technology as
a differentiating strength, providing enhanced performance
pens when we connect the next 8,000? What happens when within our operations and value to our customers. Prior to
the whole of the industrial world enters this new digital age of joining Pratt & Whitney, he served as vice president and gen-
asset intelligence? For an industrially led, digitally accelerated eral manager of the Global Digital Transformation business
company like UTC, it is a truly exciting proposition. at Hewlett Packard Enterprise, Enterprise Services (HPE). In this role, he drove
thought leadership and customer engagements worldwide to help clients along
Pratt & Whitney is only one piece of a much larger IIoT their transformation journey to becoming a digital enterprise. He also held exec-
puzzle. For example, take UTC’s Otis Elevator Company and utive positions at Cognizant and Infosys. He received a bachelor’s degree in
its service transformation work. The brand recently launched Computer Science & Engineering from Bangalore University and holds a Master
Otis ONE™, a personalized service platform that is trans- of Business Administration from the University of Texas. He is also a graduate of
the Executive Management Program from the Wharton School of Business.
forming the passenger experience and building change on
the foundations of the company’s historic data, leveraging Gabriel Levine is a Software Engineer in Brooklyn, NY, work-
the huge potential of the Internet of Things. Looking fur- ing on Systems for Aerospace IoT within the United Technol-
ther across our portfolio of subsidiaries to brands like UTC ogies Digital Accelerator. His interests are in Data Engineering
and Machine Learning. He has given talks on using Generative
Climate, Controls & Security, we see more than a loose Models and Algorithms for creating new sounds and music.
amalgam of intelligent buildings and HVAC systems. We see
synergy. We see the foundation for a new kind of building,
one that learns and adapts to the needs of its occupants.
Now imagine that at scale: an entire city of connected places

26 IEEE Internet of Things Magazine • September 2018

 Join our
Community!
Networking • Conference Discounts • Technical Publications • Volunteer

Member Benefits and Discounts

Valuable Discounts on IEEE ComSoc IEEE WCET Certification Program
Conferences Grow your career and gain valuable knowledge by
ComSoc members save on average $200 on Completing this certification program. ComSoc members
ComSoc-sponsored conferences. save $100.

Free subscriptions to Highly Ranked IEEE ComSoc Training Courses

Publications* Learn from industry experts and earn IEEE Continuing
Receive digital access to IEEE Communications Education Units (CEUs) / Professional Development Hours
Magazine, IEEE Communications Surveys and Tutorials, (PDHs). ComSoc members can save over $80.
IEEE Journal of Lightwave Technology, IEEE/OSA Journal
of Optical Communications and Networking, and IEEE Exclusive Events in Emerging Technologies
RFID Virtual Journal. Attend events held around the world on 5G, IoT, Fog
Computing, SDN and more! ComSoc members can save
*2017 Journal Citation Reports (JCR) over $60.

If your technical interests are in communications, we encourage you to join

the IEEE Communications Society (IEEE ComSoc) to take advantage
of the numerous opportunities available to our members.

Join today at www.comsoc.org

IENYCM3618.indd 1 03/11/17 4:44 PM

 The Future of IoT
Joern Ploennigs, John Cohn, and Andy Stanford-Clark, IBM

Abstract
The Internet of Things (IoT) is infiltrating many businesses. It provides simple means to collect and analyze technical system data
to identify and optimize the performance of many things in our private and work lives. This technical revolution is also revealing
new challenges and issues with our current IoT technologies. New solutions like Artificial Intelligence, Blockchain or 5G promise to
overcome these challenges. Within this paper we discuss with leading experts the pros and cons of these technologies and what it
means for the future IoT business.

Introduction ure the production from digital twin models and produce more
individualized products. This is the Industry 4.0 value proposi-
IoT is a broad field with many different technologies and appli- tion and will lead to increased production efficiency.
cations reflected by the variety of questions and opinions peo- Joern: I agree that there will be big societal value in building
ple have about IoT. Within this article we want to discuss some bigger and more resilient systems with IoT.
of these questions and review the current state and future of John: Also, more customized systems. We see this in the
IoT. To echo the variety of opinions, the paper is written as maker space. Here now people can easily build things at home
a panel discussion with three experts in IoT. First, we want to that were reserved to experts in the past. The same happens
introduce our panelists: John Cohn, Andy Stanford-Clark and in industry across domains. IoT makes it easier and cheaper to
Joern Ploennigs. measure many things and allows us to instrument everything.
John is an IBM Fellow for IoT and the technical lead at Wat- The value for the customer is that he has more information
son IoT headquarters and now delegate for the IBM AI collabo- available. For the business the value is that they do not have to
ration with MIT. His background is originally in semiconductors, send people with a checklist to collect information across the
and he has been building things from small devices to large supply chain.
robots for years. He is one of the first advocates of blockchain Joern: Of course, there is value in this. But does just collect-
for IoT, which he promoted in 2014 when blockchain was ing the data create enough business value?
largely unknown in this space [2]. John: There was a recent study [3] that showed that this
Andy is one of the creators of the MQTT protocol, the most is currently changing across the IoT industry. In the last few
common communication protocol in IoT. In addition to his years, most companies only focused on instrumentation. They
extensive experience in standardization, he is also very hands- now have collected the data and look stronger into analytics.
on. He has been building IoT solutions for years and IoT-tized The value of IoT is in deriving actionable insights that allow
his own house in 2000 with MQTT. He is now the CTO for IBM us to optimize business processes. By following some design
UK and Ireland and rolling out large IoT solutions for clients. guidelines as shown in Fig. 1, it is easier to create value. Busi-
Joern is working at IBM Research on Artificial Intelligence ness value derives from combining and analyzing IoT data from
(AI) solutions for IoT and Digital Twin. He combines semantics different sources. These data have to be continuously available
with machine learning and intuitive user interfaces to automate from a secure, resilient and flexible architecture. Only then can
IoT systems and increase their usability. He is a board member value be holistically delivered.
of the IEEE IoT Initiative and active in many conferences in the Andy: Value needs to be well designed. There is a risk that
areas of IoT and AI. people are too technology centric and do things only because
it is technically feasible. They do not think about what the prob-
Where is the Value in IOT? lem is they want to solve. As a result, many developments are
Joern: Let start the discussion with a simple question. Andy, just experimental, like smart toothbrushes, and the world will
where do you see the value in the Internet of Things? decide what we need.
Andy: IoT allows monitoring and actioning at a distance. It Joern: There are many more important problems that we
is about knowing about what is happening in a place that one need to solve, like climate change and resource limitations.
is not. Some ideas for applications go back to Bill Gates’ book, Is IoT not providing us the sensors to measure energy con-
The Road Ahead [1]. Applications can be a simple thing like sumption and evaluate and optimize the performance of our
getting notifications when the bus is coming. This leads to an systems?
improved perceived quality of life for people. IoT makes our John: To address these problems, we know that collecting
lives better in subtle ways. the data is necessary and that analytics help derive value. Now
Joern: This is the idea of ambient intelligence [4]? is the time when we have to really deliver on the value and roll
Andy: It is similar. The important difference in IoT is that we this out everywhere.
measure things not only in the environment but also at a dis- Joern: In summary, IoT solutions can improve the quality of
tance. We called it remote telemetry in the past. life and productivity, and also help save energy and the planet.
Joern: I see this value, but don’t we already have this? There are recipes that we can follow to design good solutions
Andy: It is the amount of data that gets richer and the ability that help us to deploy scalable IoT solutions. We will come
to interpret multidimensional datasets from sensors at differ- back to these aspects later in the discussion.
ent places with machine learning and AI that leads to deeper
insights. For businesses this turns data into gold. Where are we on the Hype Cycle?
John: This will, in particular, have a large impact on industrial Joern: People often ask me if IoT is hyped and if they can
production. We started with automation in industry in the 80s. trust that it will grow further and lead to substantial business.
IoT is the next evolution that allows us to automatically config- What is your answer to this question?
Andy: A very good model for this is the Gartner hype cycle
Editor’s Note: Text appearing in bold indicates a live link in the online version. for IoT [7]. The Gartner hype cycle describes the hype around
Digital Object Identifier: 10.1109/IOTM.2018.1700021 a technical development. It divides it into an initial phase where

28 2576-3180/18/$25.00 © 2018 IEEE IEEE Internet of Things Magazine • September 2018

expectations are overhyped. This is followed by a phase of dis-
illusionment where people realize that not all promises come
true and then a slope of enlightenment, where productive tech-
nologies are established.
John: This is exactly the right model. In IoT we are now at
Business
Value
1 Focus on the
business problem
the top of the hype and people have started realizing that it is
not about getting the data and connecting it to the cloud. To
derive value, they need to analyze the data. Our goal is to help
them shorten the trough of disillusionment and move quickly Digital
2
Combine Consider new
into the enlightenment zone. Smart Datasets business models
Joern: What are our tips to do this? Contracts
John: Always start from the business problem. IoT is not
the answer looking for a problem. You should start with the
business problem, and when you find the way to IoT then it will
have lasting value. Not to say that we do not have miraculous
3
Security Resilience Flexible Fog Design
innovations. Digital assistants are a good example of a product & Privacy Design Architectures Flexible
group that we did not know that we needed, and now they are Systems
indispensable.
Joern: But is it a good idea to always start with a business
problem or are we are not just solving incremental problems? Figure 1. Principles to derive business value from IoT.
John: There should be a proper balance between both. You
should not close out any breakout innovations, but they are
harder to plan for. If you want to create value intentionally, then ARTIFICIAL INTELLIGENCE
Fully Mature Twins
people should start backwards from business problems. § Own decision making
§ Cooperative behavior
Joern: Is the combination of IoT and AI not one of the future
AUTOMATE
breakout innovations as it is enabling new approaches to learn Advanced Twins § Process automation
from data? IoT created new ways to collect data. Similarly, AI § Online-learning
PREDICT & OPTIMIZE
is changing how we use machine learning. Both together allow § Time series prediction
us to automatically derive insights and predictions [5] to further Simple Twins § Optimization
§ Model-based control
optimize the IoT system. This leads step by step to a self-learn- UNDERSTAND & LEARN
§ Context modelling
ing Digital Twin of the IoT system that understands its internal § Data analytics &
function, as shown in Fig. 2. This will build the foundation of INTEGRATE machine learning
Augmented Intelligence systems that assist us in operating com- § Data and service
integration
plex systems, if not having them operate autonomously.
John: Yes, AI is one of the upcoming innovations in IoT and Figure 2. Digital Twin as an enabler for AI.
will change the maturity of IoT to become one part of many in
future solutions.
Andy: As someone who grew up with IoT, this is a major Joern: You can, for example, identify the program people
change as there are not many pure IoT things anymore. IoT is watch on TV by just analyzing the TV’s energy consumption [8].
still treated as something special, but in the future, it will just John: Exactly. On the one hand, there are technical things
embed in all processes and become a part of many business that we can do like ensuring data encryption end to end start-
architectures. ing at the edge, that the right data policies are in place and
Joern: So, IoT isn’t just hype. In your experience, it adds that data is managed according to laws like GDPR [9]. On the
value to many business processes. Your advice is to see it not in other hand, we need to make people aware of what data they
isolation as a singular solution, but rather to analyze what busi- share, with whom they share, and how it is used. When we
ness value it can add to improve existing processes. can protect their data and put them in control of it, then it also
opens ways for them to use it in a positive way and trade it for
What are the Risks For IoT? services that provide them insights, like how we trade our infor-
Joern: What are the risks in IoT that could lead to a deeper mation in the Internet for search results.
trough of disillusionment? Andy: Data in IoT is a two-sided sword with privacy on one
Andy: One common thing is that people just consider the side and business value on the other. I agree that we do not pay
capability of collecting data, and they are missing business enough attention to it and a couple of bad privacy and security
questions. Second is that people think that IoT is easy. They do examples may slow down the adoption of IoT. Therefore, we all
not consider technical challenges in connectivity, battery life, need to get better control of our data and we also need to get
and analytics. Finally, there is the cost of implementation that value for sharing it. Companies are currently overvaluing the
can still be large if one wants to roll out solutions globally. We data they have and are only thinking of ways they can sell it.
have already discussed several of these things above. We need to find neutral ways of exchanging the data.
John: I think that privacy is a huge concern. Besides this, we Joern: The question is if the data that we collect in IoT is
have complexity and robustness. Finally, pushing for standard- really that valuable and privacy-invasive. Isn’t the data we share
ization is an issue. with our Google search history or our Facebook social network
Andy: Privacy and security are big issues as people only hear more dangerous as it is much easier to build personal profiles
about the security issues of IoT and forget about the benefits. from it [10] as it contain many kinds of private data?
John: These are the things that lead us into the trough of disil- Andy: The difference there is that it is free, and people for-
lusionment. The data collected by IoT is where the value of IoT is, get that the associated cost for the service needs to be earned.
and we need to protect it. We worry a lot about security and the John: I think that most people are aware that they are giving
danger of people hacking IoT systems like a car. That certainly away data, and hopefully, they also understand the potential
can happen, like the Mirai bot net attack on web cams [6]. But consequences from cases like Cambridge Analytica. With IoT
the real danger is about data privacy and that we are sacrificing it is different as it is not that obvious anymore. Take an IoT
too much of our own information for no value. Systems will find microwave for example. From the usage of the microwave the
out more about us than what we intended to reveal. manufacturer might learn when you are home, when you eat,

IEEE Internet of Things Magazine • September 2018 29

 ess
sin la
pr ck o
Certify Device Authenticate Contracts Micro Traceable
g bu ls iva f Firmware Devices between Payments data usage
in e
miss mod cy devices

Figure 4. Blockchain as universal digital ledger for various IoT

transactions.

Internet
not be accessed without permission through a smart contract,
of Things and every access to the data is logged on the blockchain and
tion

secu g
is therefore transparent for the user. The blockchain can be

miss
menta
imple ost

further used to implement micro-transactions to pay for data

rity
in
c

usage such that a full business model can be implemented. This

gives individuals more control and gives enterprises ways to
monetize the data during the life cycle, as shown in Fig. 4.
Andy: There are already applications of Blockchain for sup-
complexity ply chain management. For example, for deep frozen goods
of analytics you have to ensure that the cold chain is never broken. IoT can
monitor this, and every transport company signs the blockchain
with IoT data as proof that it handed over the goods in good
condition. This creates transparency along the whole supply
Figure 3. Challenges in the Internet of Things.
chain [17].
Joern: I agree that this distributed nature of a Blockchain
and what you eat. He might want to sell this data to your health is very promising and technically feasible nowadays [11]. But
insurance company. The question then is: Who owns this data? IoT is machine-to-machine communication, and we explicitly
Because the manufacturer builds your microwave, does that exclude the human with this definition. Thus, the idea that we
give him the right to own the data and insights? As another have full control of our data is an illusion as we are not part of
analogy, if someone builds your house, does he keep a key? the process. Will blockchain not enable the IoT device to just
Joern: You are right, and the manufacturer should not have sell data to whoever wants to pay for it?
the right to the data. The question is generalizable beyond IoT. John: This is a matter of contract design and a device should
With the advancements in technology it becomes easier to not act autonomously. What I mean is that I as an owner can
monitor our lives from many perspectives and we lose control switch the aftermarket maintenance provider for my system and
of our data. For example, it is hard to buy a phone that allows decide what data I provide. I want to keep contractual flexibility
you only to make calls. Similarly, it will be hard to buy a micro- if a service provider goes out of business. Blockchain allows
wave that is disconnected. you to define a contract that ensures that the data remains
John: It will be a case of “buyer beware,” where people yours and is traded only with your permission.
need to make informed choices about the usage agreements Joern: I agree that blockchain and smart contracts give us
they blindly sign. The other side must be legislated, like GDPR, more flexibility than what we have today. Still, how does it
which ensures users’ rights like the right to be forgotten, that work? Is this increased flexibility not overwhelming us? Simi-
gives me the right to demand that my data is removed. larly, the many privacy settings we have now on our phones
Joern: But the issue is that we are not necessarily aware that have become very hard to use. Can we as users really manage
we provide data. this or will we end up blindly signing extensive usage agree-
John: That is the whole problem. With Google you are sign- ments?
ing conditions of use, and you are kind of aware of the data Andy: There are two points here. One is about the democ-
you provide. With IoT you are not signing anything and the IoT ratizing of the participants in a multi-way contract, and by that
device is gathering data all the time. I mean a traditional contract, written in English, and negotiated
Joern: So we face multiple challenges in IoT, as shown in by humans. The contract and its addendums and modifications
Fig. 3, to avoid the trough of disillusionment. Most important and agreements and endorsements and signings are all commit-
is that we need to make people more aware of what data they ted as immutable documents onto a blockchain.
share and we need technical solutions to put them in control of The other point is about smart contracts, which as many
who and for what use they are sharing the data. people are realizing are very hard to articulate accurately, and
once a computer gets involved with executing them and that
How Will Blockchain Change IoT? execution becomes immutable with no “compensating trans-
Joern: How will technologies like Blockchain change IoT and action,” then you are opening the way to all sorts of problems.
give us, for example, control over our data? Smart contracts are code, and code has bugs that need to be
John: Blockchain is a very flexible technology with many identified and fixed.
use cases. At its core, it creates a digital ledger in which each John: This means that blockchain in IoT needs to be sim-
transaction that is put into this ledger is encrypted, verified and ple and self-managing. When we build in all these possibilities,
signed by trusted entities. This creates the first benefit where it should not translate into complexity. The devices need to
each transaction is verified and certified with cryptographic address the skill level of the people. What we need are simple
keys. The second benefit is that all old transactions are kept protocols for transactions and also for verifying device authen-
in the ledger such that one can always trace back the history ticity and finding them in the Internet, like a search engine for
of the transaction. This creates a chain of trusted transactions devices and their services.
that nobody can modify without breaking the cryptographic Joern: You think we need something of a global registration
code. In the banking industry, it allows banks to trust electronic office for devices and services?
transactions. In IoT, it allows us to encrypt data at its source and John: We need discoverability in a universal registry that lists
protect it through its life cycle. As the data is encrypted, it can- the functionality, owner, status, and rights. Then devices can

30 IEEE Internet of Things Magazine • September 2018

search and negotiate with their own service providers and look
Mobile Ubiquitous Ambient Internet of Things
them up. This is a core element of the device democracy idea Computing Computing Intelligence
[2]. The negative thing is, this could become big brother.
Joern: We can avoid big brother with a distributed block-
chain registry, as then nobody would know everything, and it
also scales better [12]. To summarize the discussion, we know
that blockchain allows us to manage access to private and Demand for AI to filter Data Amount
User Interface Size Data
valuable data and solve several data privacy issues. It further
enables smart contracts that can realize completely new and
more flexible business models around services for devices.

Will AI Change Everything?

Joern: We already discussed that the true value in IoT lies • Large Screen • Large Screen • Small Screen • Small Screen
• Keyboard & Mouse • Touch & (Keyboard) • Touch & Speech • Speech & Gesture
within the created insights. Will AI create completely new value
propositions for IoT? Desktop PC Tablet PC Smart Phones Augmented Reality
Andy: AI will allow us to derive insights more easily and
automate these processes. Particularly for computer vision, the Figure 5. Evolution of IoT Devices vs User Interface technologies
new AI approaches around Deep Learning (DL) are enabling and the resulting demand for AI to filter the large amount of
completely new solutions. These are core technologies for data onto smaller screens.
autonomous vehicles and also for improving product quality
by detecting anomalies in production [18]. However, AI is also that allow us to write programs that we could never have writ-
adding layers of complexity. In the past we could go through ten in machine code. Our tools need to further evolve to help
lines of code to understand how it works. The increasing use practitioners design autonomous AI solutions that process the
of DL networks changes this as they are hard to interpret. data.
The more widely DL is used the more problematic this aspect Joern: When IoT systems are self-healing and autonomic,
becomes. don’t we end up with Skynet?
Joern: This is one of the greatest challenges that we face in John: I am not as worried about Skynet as I am worried
AI. Many practitioners don’t want to trust a system that they about Bedlam. Skynet is less concerning than an unmanage-
cannot interpret and thus not understand. Therefore, AI has able collection of things that stop working due to lack of
to become more transparent and capable of explaining itself robustness.
to build trust and credibility. This is one of the reasons people Joern: So no matter what we do, we end up in both cases
have doubts about autonomous vehicles. It is not only techni- with a system that we cannot control for better or worse?
cally challenging, people simply do not trust a system that they Andy: That is not going to happen. People will turn off the
do not understand. Therefore, we are working on technologies systems when they realize they have gone too far. Like when
that make DL networks more understandable and also improve they switched off the self-learning chatbots that insulted people.
their training performance. We will always be in control of the power switch.
John: One of the issues must be that devices sort out their Joern: This is a good conclusion. We should focus on under-
own problems. In the distributed systems, we do not have the standable and resilient approaches for AI. Then AI will help us
unified world of the past. This makes it harder to report issues in many ways to deal with the large amount of IoT data and to
and to build in resilience, in particular in the application of AI derive true value from it.
for differentiating normal from not normal operation. AI can
be used in security to identify what is normal, such as usage Can we Standardize IOT?
patterns that were seen before, or to detect security issues from Joern: We have had interoperability problems in communi-
abnormal patterns. The other day we were looking into the cation networks since the beginning. Technological develop-
sensors in our IoT headquarters in Munich to build a predictive ment is usually two steps ahead, and once a standard is defined
maintenance solution for Maximo Asset Health Insights. When another new one is already emerging. Within IoT we face the
we looked into the data there were periods of missing data same challenge as we need to connect legacy devices and new
across all sensors, which turned out to be periodic WiFi outag- device types into one system where we can also run analytics
es. AI could identify these patterns. across the created data. What do you think is the role of stan-
Joern: AI can certainly correlate the data of the WiFi to diag- dardization?
nose the problem. An important function of AI will be to run John: There is no single solution to the interoperability prob-
such analytic tasks and then assist the user in solving the prob- lem. It might be good to adapt some paradigms from securi-
lem. This ensures that the human remains the decision maker ty: “Any security system that believes that it is invulnerable is
and the AI is the helper. already flawed.” They have to assume they are already compro-
John: There is a theme here in designing for resilience. AI mised and need to develop robust strategies to deal with this
systems have to be understandable; they have to explain their and live with the fact that it is not perfect. The same is true with
actions, ask for what they need, and point to problems or fix interoperability. The idea that in the future we have a universal
them. Otherwise, we will never manage the complexity of 50 standard that solves these problems will not work. We need to
billion IoT devices in the coming decade. AI is just one of the design systems that can deal with interoperability issues and
tools. device diversity rather than hoping for uniformity.
Joern: It is curious, that while we are facing this growing Joern: On the protocol level we have some standards like
volume of data, our mobile user interfaces are actually getting MQTT that Andy worked on. This is one of the most commonly
smaller and can display less data, as shown in Fig. 5. Automated used protocols in IoT.
machine learning and AI are the only way to summarize this Andy: MQTT provides a very efficient and lightweight
data [5] and point to the relevant insights. Many actions need container for device messages. However, it is the con-
to be automated by the systems and the systems need to be tent of these messages that is important and their seman-
self-diagnosing, self-healing and self-sustained while just serving tics are not very well regulated. We need to resolve this
me rather than bothering me with all kinds of requests. for coexistence of devices. Also, the Internet is designed
Andy: This also needs to extend to the system design. Now for diversity of many different protocols from HTTP to
we have 3rd generation programming languages and compilers email to video. It works because all protocols coexist and

IEEE Internet of Things Magazine • September 2018 31

 Cloud-centric Systems Edge-centric Fog-centric From Cloud to
(Computing in Cloud) (Computing on Cloud) (Computing everywhere)
Edge and Fog Computing and 5G
Joern: Edge computing is discussed by many people as a
core enabler of IoT as it allows processing at the edge where
the devices are. Therefore, not all data needs to be sent to
the cloud, which reduces communication costs and improves
robustness and privacy. What are the challenges for edge com-
puting in IoT?
John: We are going to need something like Kubernetes to
IoT Device Gateway Analytics Knowledge Security
deploy our analytics at the edge. Kubernetes allows us to easily
deploy and manage containerized applications in the cloud. It
Figure 6. Comparison of cloud-, edge- and fog-computing archi- allows us to easily scale a solution by starting new containers
tectures. or to deploy a new version of the code. We need something
similar for edge systems.
Andy: Particularly when we think about putting the analyt-
ics near the data sources. This would allow us to create more
Sensors robust solutions. This way each edge system can learn its own
Number of Devices

Blockchain Acoustic Recognition customized ML and AI models from all available data and only
send out aggregated results. To do this, we need to extend
Robotics Smart Conversational analytic workflow tools that we have in the cloud to include the
Cars Interaction Visual Recognition
Semantic (Deep Learning) edge and form a fog of edge and cloud. The question is then,
Integration how we can split up our algorithms into components that can
be sent and distributed in the fog?
Telephones,
Lo

Augmented Reality Joern: Aren’t the current IoT systems designed oppositely,
w

Webpages
Re

where the devices are rather simple and send all their data
sp
on

to the cloud? Figure 6 compares the common cloud-centric

se

Virtual Reality
Tim

architecture with an edge-centric one where the processing

e

happens at the edge. The fog-centric architecture merges both

Bandwidth approaches and provides flexibility to deploy processing in the
cloud or at the edge. However, we are far from this flexible fog
Figure 7. Communication requirements of IoT applications. design.
Andy: This is just a point in time. The history of communica-
support the same communication layer TCP/IP. We need tion networks has always iterated between centralized and dis-
something similar for IoT. tributed architectures. We are rapidly moving to a stage where
John: Semantics will play a central role in this. That is why the sensors get more sophisticated and create so much data
self-describing systems like Hypercat [14] and semantic domain that we cannot stream it all into the cloud.
models like Brick [15] can be helpful. They allow us to create a John: There are many reasons why one wants to do edge
taxonomy of sensor types that are generalizable across devices analytics, ranging from latency and bandwidth cost to robust-
and services. AI can be used to map legacy systems to these ness and resilience, and to privacy and law compliance. There is
models and then the whole analytic process can be automated a great need for standardization in this space too.
afterwards [5]. On top of this, we need to create standards that Joern: Isn’t this demolishing some benefits of IoT? The cur-
control the emergent behavior and robustness of the system rent beauty of cloud-centric architectures is that they are so
and enable systems that can tolerate non-standard inputs. easy to deploy. In the past of distributed sensor networks, I had
Andy: We are moving in the right direction. Fully solving to configure and connect individual devices. Nowadays, I only
this problem is NP-complete. We just don’t have the answer turn on my device, connect it to the Internet, and I am done.
to all the questions. We have to start on the problem side and Andy: With low-power wide range networks [16] it becomes
then define what we need. For robustness, we need to have even more simple as we only need to turn it on.
defensive programming that can deal with issues. If a device John: The big game changer will be 5G. You do not need to
is missing in the field, then the system should use defaults and give up bandwidth for low power and universal reach.
keep working. Programmers today don’t necessarily consid- Andy: People dangle 5G, like blockchain, as a solution to all
er the robustness, scalability and defensiveness of the solu- problems. The sweet spot for IoT is the ubiquitous connectivity
tion. If the chaos monkey hits the system, it might lose some at low bandwidth, and low-power wide range IoT networks are
nodes, but it is still carrying on and this is intrinsic to the sys- already establishing in the market.
tem design. Joern: Both technologies will prevail as we have different
Joern: Isn’t this something that should be covered by the communication requirements in IoT, as shown in Fig. 7. 5G
programming language? As a programmer, I want to concen- will give us a wider range of bandwidth to choose from, and
trate on the high-level functionality and these defensive mech- therefore, the ability to create more flexible fog architectures
anisms are common patterns that can be generated by my that combine the benefits of edge and cloud architectures and
compiler. support a wider range of use cases.
Andy: Programming tools for IoT need to grow up to handle
this. Things like Kubernetes in the cloud already do this and Conclusion
build in redundancy in deploying the system. We need a similar Joern: To summarize the discussion: We assessed that the
solution for IoT. value of IoT is not only in collecting the data, but in building
Joern: To summarize, standardization may not resolve all business cases where the insights from analyzing data are creat-
interoperability issues in IoT. However, there are many top- ing additional value. We further discussed the hype curve of IoT
ics that need to be addressed from self-description models to and identified that we are currently at the top of the hype and
semantic domain models that are essential to automate analyt- that we need ways to quickly pass through the trough of disil-
ics in IoT. Also, more robust and flexible programming models lusionment. Therefore, it is essential to resolve the data secu-
are needed to create resilience and secure systems. rity and privacy issues. Blockchain is one approach to secure

32 IEEE Internet of Things Magazine • September 2018

data and regulate data access. It further allows us to build new [14] M. Blackstock and R. Lea, “IoT Interoperability: A Hub-Based Approach,” IEEE
Int’l. Conf. Internet of Things (IOT), Oct. 2014, pp. 79–84.
business models around smart contracts allowing more flexible [15] B. Balaji et al., “Brick: Towards A Unified Metadata Schema for Buildings,”
transactions. Further, we said that standardization might not 3rd ACM Int’l. Conf. Systems for Energy-Efficient Built Environments, 2016, pp.
fully solve the underlying problems in interoperability and that 41–50.
we also need more semantic domain models and better pro- [16] L. Krupka, L. Vojtech, and M. Neruda, “The Issue of LPWAN Technology
Coexistence in IoT Environment,” Int’l. Conf. Mechatronics-Mechatronika, Dec.
gramming tools to really resolve the data security and privacy 2016, pp. 1–8.
issues. Within all these points, we discussed pros and cons of [17] H. T. Vo, H. Kundu, and M. Mohania, “Research Directions in Blockchain
the underlying technologies. Extrapolating this to the future: Data Management and Analytics,” 21st Conf. on Extending Database Tech-
How will IoT look in 20 years? Will IoT then be a bionic implant nology, 2018.
[18] B. Bhattacharjee et al., “IBM Deep Learning Service,” IBM J. Research and
or a brain interface? Development, vol. 61, no. 4, 2017.
Andy: Maybe we will have those, but this will not be called
IoT anymore. IoT will just be part of the enabling infrastructure Biographies
in the same way a wireless communication or battery is today. Joern Ploennigs [M’04, SM’17] ([email protected].
com) leads the team on AI 4 Digital Twins at IBM Research
John: The miniaturization in IoT will continue and will lead — Ireland. He works on several aspects of enriching IoT by AI
to a huge explosion of very simple devices. This will allow us including machine learning, semantic reasoning, and natural
to integrate IoT in more systems and processes and lower the interfaces to enable autonomous, highly scalable, and accessi-
cost of their operation. It will also create new problems like the ble IoT solutions for a sustainable future. Prior to joining IBM
in 2012, he was leading a junior research group on Energy
management of so many assets that we will need to address. Design of CPS at Technische Universitaet Dresden, Germa-
Joern: So we will not be out of a job? ny, as well as the data analytics group in the Irish strategic
John: Definitively not. It is a great time to be a nerd! research cluster ITOBO as a Feodor-Lynen fellow of the Humboldt-Foundation.
He holds a master in electrical engineering for automation and control and a
References Ph.D. and a Habilitation in computer science from Technische Universitaet Dres-
[1] B. Gates et al., The Road Ahead, 1995. den. He is a program committee member of several renowned international con-
[2] P. Brody, V. Pureswaran, and J., Cohn, “Device Democracy: Saving the Future ferences and journals and board member of the IEEE IoT initiative.
of the Internet of Things,” IBM, Sept. 2014.
[3] IDC European Vertical Markets Survey, 2015, 2016, and 2017. John Cohn is an IBM Fellow and is now at the MIT/IBM Wat-
[4] C. Ramos, J. C. Augusto, and D. Shapiro, “Ambient Intelligence — the Next son AI lab. Previously he was technical lead at the Watson
Step for Artificial Intelligence,” IEEE Intelligent Systems, 23.2, 2008, pp. 15–18. IoT headquarters in Munich. There his focus is on physical
[5] J. Ploennigs, “Automating Analytics: How to learn Metadata Such That Our infrastructure for IoT, open data, Internet of Things commu-
Buildings Can Learn from Us,” SECON Wksps. — 2016 IEEE Int’l. Conf. Sensing, nications and real-time data analytics. Before joining the IoT
Communication and Networking, 2016. Division, he was an innovator in the area of design automa-
[6] E. Bertino and N. Islam, “Botnets and Internet of Things Security,” Computer, tion for both analog and digital custom integrated circuits. He
50.2, 2017, pp. 76–79. received his undergraduate degree in electrical engineering
[7] A. Velosa, W. R. Schulte, and B. J. Lheureux, Gartner Hype Cycle for IoT, 2017. at MIT and earned a Ph.D. at Carnegie Mellon University. In
[8] U. Greveler et al., “Multimedia Content Identification Through Smart Meter 2005 he was elected a Fellow of the IEEE in recognition of his contributions to the
Power Usage Profiles.,” Proc. Int’l. Conf. Information and Knowledge Engi- design automation for high performance custom circuits. He has authored more
neering (IKE), The Steering Committee of The World Congress in Computer than 30 technical papers and has contributed to four books on design automa-
Science, Computer Engineering and Applied Computing (WorldComp), 2012. tion. He has more than 100 patents in the field of design automation, methodolo-
[9] J. P. Albrecht, “How the GDPR Will Change the World,” Eur. Data Prot. L. Rev., gy, circuits and smarter systems.
2, 2016, p. 287.
[10] A. Nosko, E. Wood, and S. Molema, “All About Me: Disclosure in Online Andy Stanford-Clark is the Chief Technology Officer for IBM
Social Networking Profiles: The Case of FACEBOOK,” Computers in Human in the UK and Ireland. He is an IBM Distinguished Engineer
Behavior, 26.3, 2010, pp. 406–18. and Master Inventor with more than 80 patents. Andy is
[11] K. Christidis and M. Devetsikiotis, “Blockchains and Smart Contracts for the based at IBM’s Hursley Park laboratories in the UK and has a
Internet of Things,” IEEE Access, vol. 4, 2016, pp. 2292–303. long background in Internet of Things technologies. He has a
[12] M. Divya and N. B. Biradar, “IOTA-Next Generation Block Chain,” Int’l. J. BSc in computing and mathematics and a Ph.D. in computer
Engineering and Computer Science, 7.04, 2018, pp. 23823–26. science. He is a visiting professor at the University of New-
[13] J. Ploennigs, A. Ba, and M. Barry, “Materializing the Promises of Cognitive castle, an honorary professor at the University of East Anglia,
IoT: How Cognitive Buildings are Shaping the Way,” IEEE Internet of Things an adjunct professor at the University of Southampton, and a
J., 2017. Fellow of the British Computer Society.

IEEE Internet of Things Magazine • September 2018 33

AFTERWORD AND INTRODUCTION TO NEXT ISSUE

Introduction to Next Issue

T hanks to the authors, editors, reviewers, and publications
staff who made this issue of IEEE Internet of Things Magazine
(IoTM) possible. Launching a new publication is a challenge,
and everyone involved stepped up and made it happen.
IoT is a large and growing field that touches nearly every
industry and makes use of technologies ranging from sensing
Topic

IoT Enabling Technologies

IoT and Smart Cities

Sensors and Actuators for IoT

Submission Due Date

Oct 2018

Dec 2018

Feb 2019
Publication Date

Mar 2019

June 2019

Sept 2019
and communications to big data analytics and artificial intelli- IoT and Agriculture May 2019 Dec 2019
gence, to name just a few. IoTM will cover the field by alternat-
ing issues that focus on the “verticals” — that is, the application Communications for IoT —
Aug 2019 Mar 2020
domains — and the “horizontals” — the enabling technologies. Connectivity and Networking
The IoTM Editorial Calendar for 2019 and 2020 is summa-
rized in the accompanying table. Individuals and organizations IoT and the Environment Nov 2019 June 2020
involved in the topic areas listed are invited to submit articles. AI for IoT Feb 2020 Sept 2020
Articles not specifically addressing the topic areas will also be
accepted. The IoTM General Call for Articles can be found at IoT and Privacy May 2020 Dec 2020
https://iot.ieee.org/iot-magazine.html and is reproduced below.
The Internet of Things Magazine (IoTM) publishes high-quali- material along with their articles for review. Authors should tar-
ty articles on IoT technology and end-to-end IoT solutions. IoTM get 4,500 words or less (from introduction through conclusions,
articles are written by and for practitioners and researchers excluding figures, tables, and captions), or six (6) pages. Figures
interested in practice and applications, and selected to repre- and tables should be limited to a combined total of six. The
sent the depth and breadth of the state of the art. The technical number of archival references is recommended not to exceed
focus of IoTM is the multi-disciplinary, systems nature of IoT fifteen (15).
solutions. IoTM is a forum for practitioners to share experienc- IoTM also publishes regular columns on topics of interest to
es, develop best practices, and establish guiding principles for IoT practitioners. Topical columns update readers on issues and
technical, operational and business success. events in the world of IoT. Regular columns are published in the
The magazine is currently soliciting articles for publication. following areas:
Articles should examine one or more actual deployments of an • Around the World of IoT — Recent events or technology
IoT solution and discuss: developments in IoT.
• A high-level operational description of the IoT solution, • Bridging the Physical, the Digital, and the Social — Social-
addressing: the problem space; a summary of systems ly-aware advancements in IoT.
operation; and how the overall problems were solved. * Policy and Regulatory Affairs — Discussions and reports on
• A high-level technical description of the IoT system: What policy issues facing the world of IoT.
technical challenges were encountered? What solutions • IoT Standards — Discussions and reports on efforts in stan-
were developed? What were the technical risks encoun- dardization of IoT technology and systems.
tered in development? How were they overcome? • Privacy and Security — Discussions and reports on interac-
• A summary of the business case: What kind of benefits tion of IoT with privacy and security concerns.
did the stakeholders receive from the solution? Were they • Book Reviews.
greater than or less than expected? Were any policy or Columns should be of general interest to all members of
regulatory issues encountered? the IoT community. Columns should inform the reader about
• Lessons learned from deployment and operation: What issues and events that may affect the business and practice of
were the key lessons learned? Can this experience con- IoT; sales/marketing materials are not appropriate. Authors
tribute to defining best practices? What were the risks and are asked to strive to make their articles understandable by
rewards? the general IoT practitioner. Authors should target 1500 words
Articles should be general and present real-world experienc- or less (from introduction through conclusions, excluding fig-
es, with the intended audience being all members of the IoT ures, tables, and captions), or two (2) pages. Figures and tables
community, independent of technical or business specialty. Arti- should be limited to a combined total of two. The number of
cles are expected to add to the knowledge base or best prac- archival references is recommended not to exceed five (5).
tices of the IoT community; sales/marketing materials are not Authors should submit articles and columns to https://
appropriate. Authors are asked to strive to make their papers mc.manuscriptcentral.com/iotmag
understandable by the general IoT practitioner. Mathematical IoTM does not have a specific template and does not
material should be avoided; instead, references to papers con- require manuscripts to be submitted in any specific layout.
taining the relevant mathematics should be provided. Authors However, authors can use the template for IEEE Transactions
are encouraged to use color figures and submit multimedia to get a rough estimate of the page count: https://www.ieee.
org/publications_standards/publications/authors/author_tem-
Editor’s Note: Text appearing in bold indicates a live link in the online version. plates.html.

34 IEEE Internet of Things Magazine • September 2018

 IEEE COMSOC

1HZ2SWLFDO&RPPXQLFDWLRQ
7HFKQRORJLHVIRU*:LUHOHVV
:HGQHVGD\2FWREHUDPWRSP(DVWHUQ
2QOLQHYLD:HE([

%DFNJURXQG&RQFHSWVRI2SWLFDO
&RPPXQLFDWLRQV6\VWHPV
:HGQHVGD\2FWREHUDPWRSP(DVWHUQ
2QOLQHYLD:HE([

1HZ6RIWZDUH'HILQHG1HWZRUNLQJIRU
2SWLFDO1HWZRUNV$3UDFWLFDO,QWURGXFWLRQ
:HGQHVGD\1RYHPEHUDPWRSP(DVWHUQ
2QOLQHYLD:HE([

&2856()((6
,(((&RP6RF0HPEHU
To learn more or register visit: ,(((0HPEHU
www.comsoc.org/training. 1RQ0HPEHU

IENYCM3652.indd 1 26/09/18 8:18 PM

 w
Best Ne
urn al in
Jo 5
0 1
STM 2

Become a published
author in 4 to 6 weeks.
Get on the fast track to publication with the multidisciplinary open
access journal worthy of the IEEE.

IEEE journals are trusted, respected, and rank among the most
highly cited publications in the industry. IEEE Access is no exception;
the journal is included in Scopus, Web of Science, and has
an Impact Factor.
Published online only, IEEE Access is ideal for authors
who want to quickly announce recent developments,
methods, or new products to a global audience.

Publishing in IEEE Access allows you to:

• Submit multidisciplinary articles that do not fit
neatly in traditional journals
• Reach millions of global users through the
IEEE Xplore® digital library with free access to all
• Establish yourself as an industry pioneer by
contributing to trending, interdisciplinary topics
in one of the Special Sections
• Integrate multimedia and track usage and
citation data on each published article
• Connect with your readers through commenting
• Publish without a page limit for only $1,750
per article
17-PUB-013 3/17

Learn more at:

ieeeaccess.ieee.org

IENYCM3633.indd 1 21/02/18 5:27 PM

 Increase Your Knowledge
of Technical Standards
The foundation for today’s global high-tech success

Prepare yourself to enter the high-tech industry by learning how technical standards
develop, grow, and relate to technologies worldwide.

IEEE Standards Education is dedicated to helping engineers and students increase

their knowledge of technical standards, applications, and the impact standards have
on new product designs.

Begin your journey into the high-tech world: http://trystandards.org

Stay current with Discover learning Understand the

access to regularly opportunities role of technical
updated educational through tutorials standards in
programs and case studies the world

For information about IEEE Standards Education,

visit http://standardseducation.org

IENYCM3581.indd 1 19/04/17 10:59 PM

 Introducing IEEE Collabratec™
The premier networking and collaboration site for technology
professionals around the world.

IEEE Collabratec is a new, integrated online community where IEEE members,

Network.
researchers, authors, and technology professionals with similar fields of interest
can network and collaborate, as well as create and manage content. Collaborate.
Featuring a suite of powerful online networking and collaboration tools, Create.
IEEE Collabratec allows you to connect according to geographic location,
technical interests, or career pursuits.

You can also create and share a professional identity that showcases key
accomplishments and participate in groups focused around mutual interests,
actively learning from and contributing to knowledgeable communities.
All in one place!

Learn about IEEE Collabratec at

ieeecollabratec.org

IENYCM3577.indd 1 19/04/17 10:55 PM