iot ecosystem study 2014
DESCRIPTION
IoT Ecosystem Study 2014TRANSCRIPT
IEEE Standards Association (IEEE-SA) Internet of Things (IoT) Ecosystem Study
i
Copyright © 2015 IEEE. All rights reserved.
IEEE-‐SA Internet of Things (IoT) Ecosystem Study
ii
Copyright © 2015 IEEE. All rights reserved.
Trademarks and Disclaimers IEEE believes the information in this publication is accurate as of its publication date; such information is subject to change without notice. IEEE is not responsible for any inadvertent errors.
The Institute of Electrical and Electronics Engineers, Inc. 3 Park Avenue, New York, NY 10016-‐5997, USA Copyright © 2015 by The Institute of Electrical and Electronics Engineers, Inc. All rights reserved. Published January 2015. Printed in the United States of America. IEEE is a registered trademark in the U. S. Patent & Trademark Office, owned by The Institute of Electrical and Electronics Engineers, Incorporated. PDF: ISBN 978-‐0-‐7381-‐9501-‐8 STDVA20089
IEEE prohibits discrimination, harassment, and bullying. For more information, visit http://www.ieee.org/web/aboutus/whatis/policies/p9-‐26.html. No part of this publication may be reproduced in any form, in an electronic retrieval system, or otherwise, without the prior written permission of the publisher. To order IEEE Press Publications, call 1-‐800-‐678-‐IEEE. Find IEEE standards and standards-‐related product listings at: http://standards.ieee.org
iii
Copyright © 2015 IEEE. All rights reserved.
Notice and Disclaimer of Liability Concerning the Use of IEEE-‐SA Documents
This IEEE Standards Association (“IEEE-‐SA”) publication (“Work”) is not a consensus standard document. Specifically, this document is NOT AN IEEE STANDARD. Information contained in this Work has been created by, or obtained from, sources believed to be reliable, and reviewed by members of the IEEE-‐SA Internet of Things Steering Committee. IEEE expressly disclaims all warranties (express, implied, and statutory) related to this Work, including, but not limited to, the warranties of: merchantability; fitness for a particular purpose; non-‐infringement; quality, accuracy, effectiveness, currency, or completeness of the Work or content within the Work. In addition, IEEE disclaims any and all conditions relating to: results; and workmanlike effort. This document is supplied “AS IS” and “WITH ALL FAULTS.” Although IEEE believes that the information and guidance given in this Work serve as an enhancement to users, all persons must rely upon their own skill and judgment when making use of it. IN NO EVENT SHALL IEEE BE LIABLE FOR ANY ERRORS OR OMISSIONS OR DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL DAMAGES (INCLUDING, BUT NOT LIMITED TO: PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT OF THE USE OF THIS WORK, EVEN IF ADVISED OF THE POSSIBILITY OF SUCH DAMAGE AND REGARDLESS OF WHETHER SUCH DAMAGE WAS FORESEEABLE. Further, information contained in this Work may be protected by intellectual property rights held by third parties or organizations, and the use of this information may require the user to negotiate with any such rights holders in order to legally acquire the rights to do so. IEEE makes no assurances that the use of the material contained in this work is free from patent infringement. Essential Patent Claims may exist for which no assurances have been made to the IEEE, whether by participants in this activity or entities outside the activity. The IEEE is not responsible for identifying essential patent claims for which a license may be required, for conducting inquiries into the legal validity or scope of patents claims, or determining whether any licensing terms or conditions, if any, or any licensing agreements are reasonable or non-‐discriminatory. Users are expressly advised that determination of the validity of any patent rights, and the risk of infringement of such rights, is entirely their own responsibility. No commitment to grant licenses under patent rights on a reasonable or non-‐discriminatory basis has been sought or received from any rights holder. This Work is published with the understanding that IEEE is supplying information through this Work, not attempting to render engineering or other professional services. If such services are required, the assistance of an appropriate professional should be sought. IEEE is not responsible for the statements and opinions advanced in this Work.
iv
Copyright © 2015 IEEE. All rights reserved.
Contents
Table of Contents
INTRODUCTION .................................................................................................................................... 1
WHAT IS IOT? ....................................................................................................................................... 3
THE MARKET FOR IOT ........................................................................................................................... 4
PLAYERS POSITIONED TO SHAPE THE IOT MARKET ............................................................................................. 4 COMMERCIAL PLAYERS ...................................................................................................................................... 4 RESEARCH AND ACADEMIA ................................................................................................................................. 4 GOVERNMENTS AND UTILITIES ............................................................................................................................ 5 OTHER PLAYERS ................................................................................................................................................ 5 MARKET SEGMENTS AND VERTICALS POISED TO DRIVE THE GROWTH OF IOT ........................................................... 5 CONSUMER GOODS ........................................................................................................................................... 5 EHEALTH ......................................................................................................................................................... 6 SMART TRANSPORTATION .................................................................................................................................. 6 ENERGY DISTRIBUTION (SMART GRID) ................................................................................................................... 6 SMART CITY ..................................................................................................................................................... 6 FIGURE 1: POSSIBLE SMART CITY/SMART GRID FRAMEWORK ................................................................................... 7 DISTRIBUTION AND LOGISTICS ............................................................................................................................. 7 PUBLIC SAFETY ................................................................................................................................................. 7 INDUSTRIAL AND MANUFACTURING ...................................................................................................................... 8 AGRICULTURE AND NATURAL-‐RESOURCE MANAGEMENT .......................................................................................... 8 BIG-‐DATA ANALYTICS ......................................................................................................................................... 8 NEW SEGMENTS ............................................................................................................................................... 8 MISSING FROM THE BUSINESS-‐MODEL POINT OF VIEW ....................................................................................... 9 QUADRUPLE TRUST ........................................................................................................................................... 9 USABILITY ...................................................................................................................................................... 10 SILOS ............................................................................................................................................................ 10 INTEROPERABILITY AND STANDARDIZATION .......................................................................................................... 10 MONETIZATION .............................................................................................................................................. 11 EDUCATION ................................................................................................................................................... 11 SCALABILITY ................................................................................................................................................... 11
IOT TECHNOLOGIES ............................................................................................................................ 12
TECHNOLOGIES ENABLING THE GROWTH OF IOT TODAY ................................................................................... 12 SENSORS, ACTUATORS AND SMART DEVICES ......................................................................................................... 12 NETWORKS AND COMMUNICATIONS .................................................................................................................. 12 COMPUTING AND STORAGE .............................................................................................................................. 13 BIG-‐DATA ANALYTICS ....................................................................................................................................... 13 MISSING FROM THE TECHNOLOGY POINT OF VIEW .......................................................................................... 13
v
Copyright © 2015 IEEE. All rights reserved.
QUADRUPLE TRUST: PROTECTION, SECURITY, PRIVACY, SAFETY ................................................................................ 13 SENSOR, ACTUATOR, AND DEVICE IMPROVEMENTS ............................................................................................... 14 NETWORKS AND COMMUNICATIONS .................................................................................................................. 15 INTEROPERABILITY .......................................................................................................................................... 15 SEMANTICS AND INTELLIGENCE .......................................................................................................................... 16 BIG DATA ...................................................................................................................................................... 16 SCALABILITY ................................................................................................................................................... 16 FUNCTIONAL SAFETY ........................................................................................................................................ 17
IOT STANDARDIZATION ...................................................................................................................... 18
STANDARDS BODIES DOING IMPORTANT WORK TO ENABLE IOT ......................................................................... 18 SPECIFIC STANDARDS ACTIVITIES RELATED TO IOT? ......................................................................................... 18 MISSING FROM THE STANDARDIZATION POINT OF VIEW ................................................................................... 19 COMMON DEFINITION OF IOT ........................................................................................................................... 19 GLOBAL REACH AND COORDINATION .................................................................................................................. 19 ARCHITECTURE AND REFERENCE MODELS ............................................................................................................ 20 QUADRUPLE TRUST: PROTECTION, SECURITY, PRIVACY, SAFETY ................................................................................ 20 SCALABILITY ................................................................................................................................................... 20 APPLICATION STANDARDS ................................................................................................................................ 21 INTEROPERABILITY .......................................................................................................................................... 21 OTHER COMMENTS ......................................................................................................................................... 21 GLOBAL STANDARDIZATION ....................................................................................................................... 22
ROLE OF ACADEMIA AND RESEARCH (AR) ........................................................................................... 23
USER ACCEPTANCE IS KEY ................................................................................................................... 24
CONCLUSIONS .................................................................................................................................... 25
ANNEX A ............................................................................................................................................ 27
IOT ROUNDTABLES SPONSORED BY IEEE-‐SA ........................................................................................ 27
IEEE-‐SA IOT ROUNDTABLE – TAIPEI ........................................................................................................... 27 IEEE-‐SA IOT ROUNDTABLE – WASHINGTON DC ........................................................................................... 29 IEEE-‐SA IOT ROUNDTABLE – SANTA CLARA, CA ........................................................................................... 30
ANNEX B OTHER MEETINGS THAT PROVIDED INPUT TO THIS ECOSYSTEM STUDY ............................... 32
IEEE COMSOC IOT RAPID REACTION STANDARDIZATION ACTIVITY WORKING MEETING ........................................ 32
ANNEX C IOT STANDARDS AND ACTIVITIES ......................................................................................... 34
IEEE-‐SA IoT Ecosystem Study
1
Copyright © 2015 IEEE. All rights reserved.
IEEE-‐SA Internet of Things (IoT) Ecosystem Study
Introduction Broadly speaking, the Internet of Things (IoT) is a system consisting of networks of sensors, actuators, and smart objects whose purpose is to interconnect “all” things, including everyday and industrial objects, in such a way as to make them intelligent, programmable, and more capable of interacting with humans and each other. See “What is IoT” to learn how “IoT” is used in this study. The IoT ecosystem is hard to define. It is complex, and it is difficult to capture due to the vastness of possibility and the rapidity with which it is expanding. However, there is no doubt that IoT is changing the world. It is shaping the evolution of the Internet. IoT is creating numerous challenges and opportunities for engineering and science. In response to these challenges and opportunities, the IEEE created an IoT Initiative to coordinate its IoT efforts in publications, conferences, education, and standardization. The success of IoT depends strongly on standardization, which provides interoperability, compatibility, reliability, and effective operations on a global scale.
Recognizing the value of IoT to industry and the benefits this technological innovation brings to the public, the IEEE Standards Association (IEEE-‐SA) has a number of standards, projects, and events (see 0) that are directly related to creating the environment needed for a vibrant IoT.
As a part of the IEEE IoT Initiative, the IEEE-‐SA developed this IoT ecosystem study. IEEE-‐SA engaged stakeholders in key regions of the world to obtain the input that shaped this study. This engagement was in the form of a series of roundtable discussions held by IEEE-‐SA. Invitations to the roundtables included the 10 questions listed below. The questions were intended to seed the discussion; they did not constrain the roundtables, which were spirited discussions of IoT and associated issues.
1. Who are the players that are positioned to shape the IoT market?
2. What market segments and verticals are poised to drive the growth of the IoT?
3. What are the technologies enabling the growth of the IoT today?
4. What is missing from the business-‐model point of view?
5. What is missing from the technology point of view?
6. What standardization bodies are doing important work to enable the IoT?
7. What specific standards activities do you think of when you think of the IoT?
8. What is missing from the standardization point of view?
9. Do you see IoT activities as more suitable for regional or global standardization?
10. What can research, industry and academic institutions contribute?
The structure of this study is based on these questions and divides the discussion into the three principal areas of Market, Technology, and Standards. After the discussion of these three areas, the role of academia and research as a contributor to the three areas is discussed. Finally, the importance of user acceptance is addressed.
IEEE-‐SA IoT Ecosystem Study
2
Copyright © 2015 IEEE. All rights reserved.
It should be noted that this structural division is somewhat arbitrary:
¾ Discussions in the roundtables frequently addressed multiple areas simultaneously because the areas are so interdependent, and
¾ Information on topics that were discussed in multiple areas has been partially reorganized for consistency.
However, some topics are not discussed entirely in one place when a division made sense. For example, under the issue of “missing scalability,” IP addressability is discussed in the Marketing section because this is primarily an economic issue (IPv6 exists but has not been widely deployed), while unique hardware identification is discussed in the Standards section because there is still a need for consensus on how to move forward from MAC addresses.
Additional input to this study was provided by the IEEE Communications Society (IEEE ComSoc), which held an IoT-‐related workshop.
Members of the IEEE-‐SA Corporate Advisory Group (CAG) and members of the IEEE IoT Initiative have reviewed drafts of this study and provided additional content during the review process. Additional supporting material was garnered from IoT-‐related workshops held by IEEE-‐SA.
The information provided by these sources—the IEEE-‐SA roundtables, the IEEE ComSoc and IEEE-‐SA workshops, and the reviewers—composes the body of this whitepaper.
The annexes contain the following supporting material:
¾ Annex A contains details regarding the IEEE-‐SA roundtable discussions, including lists of the participants for each discussion
¾ Annex B contains details regarding the IEEE ComSoc workshop, including a list of the participants
¾ Annex C contains a list of IoT-‐related standards and activities (standards under development, IoT evangelism, etc.)
IEEE-‐SA IoT Ecosystem Study
3
Copyright © 2015 IEEE. All rights reserved.
What is IoT? The primary differentiator between the traditional (legacy) Internet and the Internet of Things (IoT) is the proliferation of uniquely identifiable devices with embedded sensors and actuators.
There is no “official” definition of IoT. Many of the organizations working in IoT have created definitions; these definitions differ widely. These current definitions vary so much that they are reminiscent of the tale of the blind men and the elephant. Without a common definition of IoT, it will be difficult to put together solutions or even define problem areas.
Some people think of IoT as a new add-‐on to the “traditional” Internet; others think of IoT as the ongoing evolution of the Internet. Under both definitions, the end result will be the interconnection of “all” things. The distinction between the two classes of definitions is that the “evolutionary” definition of IoT includes traditional computing and networking devices, as well as smart phones, tablets, point-‐of-‐sale (POS) terminals and the myriad new sensors and smart objects that are beginning to appear. While there is merit to both conceptualizations, the evolutionary interpretation is adopted for this study because it is inclusive and avoids the need to identify individual devices and services as being either “traditional” or IoT. Thus, for the purpose of this study: IoT refers to any systems of interconnected people, physical objects, and IT platforms, as well as any technology to better build, operate, and manage the physical world via pervasive data collection, smart networking, predictive analytics, and deep optimization.
Note that there are many IoTs. There is the global IoT (evolving from the global Internet) as well as local and private IoTs. The term “IoT” encompasses all of these.
Some people also distinguish between Internet of Things, Internet of Vehicles, Smart Grid, eHealth, etc. While this distinction may make sense in some contexts, this study takes the position that all of these are part of IoT. They will likely share communications technologies and all need the quadruple trust and interoperability described herein, so it makes sense to consider all of them during the development of IoT and to distinguish between them only when needed.
IoT products include devices, apps, and services (e.g., smart phones, tablets, intelligent networks, big-‐data analytics, and cloud storage). A key aspect of IoT is the intelligent connectivity of these products.
For the most part, IoT devices will be self-‐configuring and adaptive to reduce the need for human interaction (e.g., network discovery, self-‐description of devices, auto-‐configuration of device and network parameters). However, it is likely that there will be situations where devices will be more rigidly constrained to satisfy safety, legal, and regulatory obligations.
IoT is the subject of a great deal of hype and many bold predictions about where it will eventually take us. However, there is no question that IoT is changing the world. In addition to connecting people, anytime and everywhere, it is connecting IoT products to humans and other IoT products, and it is putting these products at the service of humanity. This transformation has already begun; it will only continue to accelerate.
IEEE-‐SA IoT Ecosystem Study
4
Copyright © 2015 IEEE. All rights reserved.
The Market for IoT The IoT market is burgeoning but fragmented. Early players are active and currently creating products for which they see a market. In order to get products to market, these players are implementing proprietary solutions, some of which may evolve into de facto standards. Currently, IoT is trending toward vertical applications.
IoT development and deployment are motivated by the desire to provide and use existing goods and services more efficiently (cheaper, faster, better) and by the desire to create new goods and services that will drive new revenue streams. Connecting things and allowing data to move will open new markets, just as the Internet did.
New products and business models will disrupt traditional business models; some of these new products and models will be created by unintended consequences of technologies being deployed. Who could have foreseen the myriad ways that smart phones have been extended since their introduction?
The market for IoT is truly global. Much of the growth for IoT may be in emerging economies because they do not have to deal with as many existing infrastructure issues; instead, they can build new structures to address the needs of IoT.
Who are the players that are positioned to shape the IoT market? What market segments and verticals are poised to drive the growth of the IoT? What is missing from the business-‐model point of view?
Players positioned to shape the IoT market There are several different classes of players that will play major roles in the growth of IoT. Players in each of these classes are currently active to varying degrees.
¾ Commercial players in the “off-‐line” world
¾ Commercial players in the “on-‐line” world
¾ Research and academia
¾ Governments and utilities
¾ Other players
Commercial players Commercial players fall into two categories, “off-‐line” and “on-‐line.” Many players participate in both categories; they are classified herein based on their principal participation.
¾ Players in the “off-‐line” world are the thing manufacturers. They are producing smart appliances, home automation devices, personal appliances (smart phones, wearables, etc.), smart automobiles, etc.
¾ Players in the “on-‐line” world generally provide IoT-‐enabling services although many also provide things. Players include Amazon, Apple, Google, and Microsoft. On-‐line players are pioneering in many services and applications, e.g., Microsoft® Azure™, Amazon Web Services, Google Glass (a thing), smart phones, etc.
Research and academia Research and academia are busy creating the theories, new products, and materials that will fuel the growth of IoT. Academia is also educating the new generation of technologists and business people who will expand IoT. For further discussion, see “Role of Academia and Research.”
IEEE-‐SA IoT Ecosystem Study
5
Copyright © 2015 IEEE. All rights reserved.
Governments and utilities Governments and utilities are creating smart cities, the smart grid, etc. Specific examples of IoT devices being deployed include traffic cameras, security cameras, smart meters, adaptive traffic controllers, etc. Governments also play a key role in shaping technology through funding (defense, energy, transportation, etc.) and setting regulatory frameworks and policies (security, interoperability, etc.). Governments are already investing in IoT domains.
Other players In addition to established players, entrepreneurs will also play a role, creating new markets and disrupting existing markets with novel new products. In areas of high unemployment, one path that talented and ambitious individuals are taking as an alternative to being unemployed is to become entrepreneurs.
Arguably, consumers themselves are players who will shape IoT by their purchases. Examples of these choices include wearables (Google Glass, smart watches, etc.) and crowdfunding support for entrepreneurs. For further discussion, see “User Acceptance is Key.”
Finally, regulatory agencies cannot be overlooked. Nearly every nation has one or more regulatory organizations that will govern some aspect of IoT (e.g., privacy, health, wireless technology), and their impact will be felt.
Market segments and verticals poised to drive the growth of IoT There are many market segments and vertical poised to drive IoT growth:
¾ Consumer goods: smart phone, smart home, smart car, appliances, etc.
¾ eHealth: fitness, bioelectronics, and healthcare
¾ Smart transportation
¾ Energy distribution (smart grid)
¾ Smart city
¾ Distribution and logistics
¾ Public safety
¾ Industrial and manufacturing
¾ Agriculture and natural-‐resource management
¾ Big-‐data analytics
¾ New segments
Consumer goods and eHealth are thought to be the two primary market segments (application domains) that will initially account for explosive growth of IoT sensors and devices.
Consumer goods In the consumer-‐goods segment, there are already more than a billion smart phones in use worldwide. Newer smart phones include more sensors, and new apps are adding to the power of smart phones to participate in even more application domains. Arguably, smart phones now function as sensor devices
IEEE-‐SA IoT Ecosystem Study
6
Copyright © 2015 IEEE. All rights reserved.
and application platforms in many different market segments (consumer goods, eHealth, smart city, etc.) but are included in the consumer-‐goods segment since that is where they got their start.
Consumer goods also include appliances, electronics, clothing/accessories (embedded technology), automobiles, etc., all with sensors and embedded devices to provide new functionality.
Smart homes include smart appliances, lighting and curtain controls, multimedia entertainment, security systems, etc. Hubs are being introduced to consolidate control of many different devices that use proprietary protocols and to extend that control to the Internet. Convenience and energy saving are major motivators.
Smart cars will park themselves, respond to voice commands, connect to your smart phone, connect to wearable sensors, etc. These features are beginning to appear in high-‐end models.
eHealth eHealth is a broad segment that includes such things as telemedicine, virtual healthcare teams (providing diagnoses and surgical support), mHealth, etc. Some people argue that it also includes things like electronic health records and healthcare information systems.
mHealth addresses the mobile aspects of eHealth. mHealth leverages bioelectronics sensors to provide mobile services such as heart monitoring implants; infant and geriatric monitors, real-‐time monitoring of patient vitals, and fitness monitoring.
Smart transportation Smart transportation includes both public-‐transport and private-‐transport (cars and trucks). Public-‐transport activities include integrated mobility platforms, train control, and electronically accessible timetables and ticketing. Among the focuses of smart transportation are fuel efficiency, cleaner fuels, advanced vehicle technologies, route planning, and logistics to reduce costs and improve efficiency.
In the private-‐transport domain, a major focus is vehicle automation. This includes both vehicle control (access by IoT systems will be limited since these systems are essential for safe operation of the vehicle) and more openly accessible IoT systems such as navigation, climate control, etc. IoT applications could be car-‐to-‐car, car-‐to-‐curb, or completely internal to the car (including external devices such as smart phones or laptops that are brought into the car).
Another aspect of smart transportation is the appearance of services like Uber and Lyft that are transforming the taxi/limo industry and services like ZipCar and Community CarShare, which provide car sharing services for people who only need a car occasionally.
Energy distribution (smart grid) Management of the energy distribution segment is growing, with solutions already in place on both the supply side and the demand side. Several years of effort have already gone into the smart grid, using information and communications technology (ICT) to gather and act on information—e.g., behavior of suppliers and consumers—to automate the electrical grid and make it more efficient and more resilient.
Smart city The smart city is a conceptual model (see Figure 1) that describes the use of information and communications technology (ICT) to create networked infrastructure(s) to improve efficiency and development of all areas of urban life, including business services, social services, housing, leisure, and cultural services. The focus is on the connected city as a vehicle for growth.
IEEE-‐SA IoT Ecosystem Study
7
Copyright © 2015 IEEE. All rights reserved.
Figure 1: Possible Smart City/Smart Grid Framework
An interesting example is Boston’s new pothole tracking system. According to a press release from the Mayor’s Office of Boston a new app— STREET BUMP®—uses the technology already built into smart phones (including accelerometers and GPS systems) to identify bumps and automatically report them to the city. The city then uses big-‐data techniques to eliminate reports of (intentional) speed bumps and builds a map ranking potholes by size and number of reports. In terms of the smart city, this IoT application will be used in concert with driver hotlines and road crew inspection to help Boston fill more potholes faster. This example will be referenced throughout this study because it is exemplary of many of the items discussed herein.
Another example is the City of Philadelphia’s use of IoT to manage its public garbage cans; compacting the garbage and tracking the cans and the amount of garbage they contain. This information has enabled the city to reduce collections substantially, saving approximately US$900,000 per year.
Distribution and logistics Distribution and logistics is leveraging IoT devices such as radio frequency identification (RFID) tags and barcode readers to save time and money in the areas of inventory management, order fulfillment, shipping management, warehousing, etc.
Public safety Public safety covers disaster prevention and relief. It also includes systems that overlap with segments such as smart cities and smart transportation. Activities range from food-‐safety to road-‐traffic safety,
IEEE-‐SA IoT Ecosystem Study
8
Copyright © 2015 IEEE. All rights reserved.
police and fire services, emergency medical services, rescue squads, etc. IoT is changing all of these areas with mobile communication, traffic and security cameras, and novel apps such as Boston’s pothole tracking system.
Industrial and manufacturing Activities in this market segment include such diverse activities as industrial automation (e.g., automated assembly lines) and material flow traceability systems.
Automation of production already plays an important role in manufacturing. IoT is increasing automation and providing more flexibility and efficiency to the production process. IoT allows localization of individual products in the production lines and the automatic selection of the required production steps. Furthermore, remote and predictive maintenance are improving the efficient operation of entire factories.
Agriculture and natural-‐resource management Agriculture and natural-‐resource management activities include GPS mapping technology and sensors to analyze crop yields during harvesting, identify water and nutrient needs, monitor crop growth, and use this information to apply nutrients and water to specific segments of the fields. (This is sometimes called underground IoT.) Other activities include the deployment of biochip transponders in animals (domestic, farm and wild).
IoT systems are starting to monitor the food-‐safety supply chain from the field to point of sale.
Big-‐data analytics Big-‐data analytics will both contribute to and benefit from the growth of IoT. The amounts of real-‐time data that are being generated will grow rapidly and present new business opportunities. On the raw data level, both owners and producers will be able to sell the information captured by IoT devices to companies to feed their business intelligence and to convert into knowledge and wisdom. On the processed data (knowledge) level, companies capable of big-‐data analytics will be able to sell the knowledge they have derived to other entities (service providers and manufacturers) to augment existing or create new services and products. On the scenario planning (wisdom) level, companies can make use of the information they have acquired for long-‐range planning.
Boston’s pothole tracking system (described in “Smart city”) is an example of big-‐data analytics taking IoT-‐generated data and using it to provide a solution to a real world problem (road maintenance).
New segments Looking to the future, as sensor and device technology improves and IoT grows, there are countless opportunities for novel new applications and devices in all market segments. The convergence of multiple technologies will create new market segments as yet unthought-‐of.
The following list presents some of the potential new business models mentioned during the roundtables. Arguably, rudimentary examples of some of these are already in use.
¾ The lending of devices or device capacity when they are not in use. IoT technologies allow easy traceability of the devices and billing for their use.
¾ Reduction of data acquisition costs will make possible sustainable business models based on the collection of previously unavailable data.
IEEE-‐SA IoT Ecosystem Study
9
Copyright © 2015 IEEE. All rights reserved.
– Mobile crowd-‐sensing: individuals with sensing and computing devices can collectively share data and extract information to measure and map phenomena of common interest.
– This “cheap data” also makes possible scientific research that was not economically feasible in the past (e.g., a small and medium lake research project in Taiwan to determine whether shallow lakes could sustain their ecologies).
– Real-‐time data could open new opportunities in decision making and policy making.
¾ Cloud storage and computing could augment existing business models (pay-‐per-‐use, micro-‐payments, etc.).
¾ The IoT could foster new business models for communications (pay per use, real-‐time analysis, etc.).
¾ Semantics and security might be combined to provide new services such as mobile real-‐time warning, security for one-‐way or two-‐way devices, system alerts, etc.
Missing from the business-‐model point of view There are many different business models already in use. There are sure to be new business models and new markets as IoT develops. IoT will be used to make existing devices and services more economical and/or powerful; it will enable the development of hitherto undreamed of new devices and services that will generate new revenue streams.
There are still many business issues that need to be addressed:
¾ Quadruple trust: protection, security, privacy, safety
¾ Usability: implementation and system integration
¾ Silos
¾ Interoperability and standardization
¾ Monetization
¾ Education
¾ Scalability
Note that many of these topics also appear in “Missing from the technology point of view” and/or “Missing from the standardization point of view,” depending upon the nature of what is missing. For example, in this Market section, scalability is included because the need for greater addressability is primarily an IPv6 deployment issue.
Quadruple trust The quadruple trust—protection, security, privacy, safety—will be discussed in greater detail in the IoT technology discussion, but these items are important in the IoT market discussion since these attributes can play a deciding role in the success of IoT-‐related devices and services and dramatically affect the reputation and success of the companies providing the products and services. Open questions include who provides the various elements of the quadruple trust and how the difficult trade-‐offs between quadruple trust and usability will be resolved.
IEEE-‐SA IoT Ecosystem Study
10
Copyright © 2015 IEEE. All rights reserved.
Some people use “system integrity” in place of quadruple trust. In this study use of the term system integrity is avoided in this context because it can also apply to software and hardware issues not related to the quadruple trust.
Usability IoT systems need to be optimized for usability. Ease of use will be a vital factor for the growth of all IoT devices and services; it is particularly important for the consumer-‐goods segment.
Many of today’s IoT-‐based systems are complex. It is imperative to reduce this complexity and to make IoT easier to use. One should not need to be an “expert” or to hire an “expert” to put a device into service, configure a system, or use a system once it is operational. IoT should be usable by a consumer without extensive training.
At the same time the system is easy to use, it must also satisfy the quadruple trust. There are difficult trade-‐offs between ease of use and the quadruple trust. This will be a fundamental challenge affecting both the usability and quadruple trust efforts.
Design, implementation, and system integration are vital components of the usability process. Subject-‐matter experts in many fields do not have technological expertise or knowledge of how to embrace technology (e.g., clothing manufacturers who want to enter the field of wearable computing). Cross-‐boundary collaboration of usability experts, subject-‐matter experts of the relevant application areas, and technology experts is needed to optimize the usability.
Silos Many products start as application silos, but IoT market growth will be slowed by silos. Note that there is a subtle distinction between verticals (vertical apps) and silos (siloed apps): silos are verticals with strong barriers preventing other apps from using the information. While there may be occasions where silos are necessary (e.g., health-‐related data that must be protected for privacy reasons), silos that are primarily intended to thwart competition will delay or prevent market growth.
Interoperability and standardization When openness of a system becomes a deciding factor for purchase, it drives the need for standards. For example, near-‐field communications (NFC) in a smart phone communicating with a point-‐of-‐sale (POS) device requires interoperability among a number of device from different manufacturers; Bluetooth® communication between smart phones and automobiles requires interoperability between different smart-‐phone providers and automobile manufacturers. Both of these examples require interoperability among devices from different manufacturers and thus require that standards exist and be accepted.
An excellent example of what happens when standards are missing or not widely accepted1 is the home automation market. This market segment has been struggling for decades; there has been very little progress and the market is fragmented. Products from different manufacturers do not interoperate and thus have (relatively) low market penetration and high prices.
1 Home automation standards do exist. For example, KNX and LonMark®are international standards (ISO/IEC) for home automation. However, they have a limited market acceptance and a lot of proprietary systems exist.
IEEE-‐SA IoT Ecosystem Study
11
Copyright © 2015 IEEE. All rights reserved.
Monetization What is the motivation of the organization providing IoT-‐based service? Is it increased sales, better profit margins, a better relationship with the customer, or some other motivation?
Different companies will have different business models depending upon whether they are providing equipment or a service.
For equipment, is the equipment sold outright? Sold with a service plan? Leased or rented? What is the term of the lease/rental? Yearly? Monthly? Hourly (e.g., extremely short term car rentals like ZipCar or Community CarShare)?
For services, how is the service monetized? Is it the end user who pays? Do they pay by subscription or by event usage? How is the revenue collected? How will services be billed? Is the service a credit or debit based service?
New revenue models that have not been seen before are likely to arise.
Education Education will be an ongoing challenge at all levels.
Users of IoT systems will need to be trained to use the system. Training will be particularly challenging in the case of consumers, most of whom abhor steep learning curves. Training will also be needed for users of job-‐related IoT systems; this will be somewhat easier because learning to use the IoT system will also be part of their job. However, if training costs are too high, adoption of IoT systems may be delayed (another argument for complexity reduction).
The technologists who are developing IoT systems will also need education. Many technologists who are highly skilled in their particular area will—perhaps because of business schedules and funding or because of hubris—venture outside their field of expertise and attempt to implement something they know very little about. Examples abound of botched security implementations containing well-‐known flaws that a security expert would have known better than to implement.
Scalability The market players are all anticipating huge numbers of things (e.g., numbers like 50 billion things are being forecast within the next 10 years). The sheer numbers of things that will need to be attached to the Internet require widespread adoption of the IPv6 protocol. The IPv4 address space is exhausted; while some IoT products can be implemented behind gateways employing workarounds such as network address translation (NAT) technology, these are basically stopgap solutions. This absence of IPv6 technology with its enormous address space and other benefits is primarily a market issue. The technology exists but has not achieved widespread deployment because there has not been sufficient economic demand. The IoT will provide that demand.
IEEE-‐SA IoT Ecosystem Study
12
Copyright © 2015 IEEE. All rights reserved.
IoT Technologies Technological advances are fueling the growth of IoT. Improved communications and network technologies, new sensors of various kinds, and improved—cheaper, denser, and more reliable and power efficient—storage both in the cloud and locally are converging to enable new types of products that were not possible a few years ago.
What are the technologies enabling the growth of the IoT today? What is missing from the technology point of view?
Technologies enabling the growth of IoT today The technologies enabling the growth of IoT can be categorized as
¾ Sensors, actuators, and smart devices
¾ Networks and communications
¾ Computing and storage
¾ Big-‐data analytics
Sensors, actuators and smart devices As sensors, actuators and smart devices become smaller, more versatile, lower cost, and more power efficient, they are being deployed in greater numbers, either as special-‐purpose devices or embedded into other products.
Large numbers (more than a billion) of smart phones are already deployed; they provide a collection of things (sensors, actuators, displays, and modems) to support the implementation of IoT systems.
The unification and convergence of the platform (smart phones), the vast number of platforms already deployed, the accessibility (APIs and interfaces) of the platform to app developers, and a (hard-‐to-‐quantify) social component have combined to make smart phones a key enabler of current IoT systems.
These APIs and interfaces have enabled programmers to create huge new opportunities in many market segments. An example is the Boston pothole tracking system mentioned earlier.
In addition to smart phones, many other types of devices are already being deployed, both independently (e.g., smart watches and other wearables) and as part of other devices (e.g., appliances and consumer electronics). The evolution of this technology category will enable many new applications as IoT grows.
Networks and communications Cellular and Wi-‐Fi® are ubiquitous; they are evolving to support higher bandwidths and lower cost. Bluetooth is also becoming lower cost. New communication technologies like Bluetooth® low energy (Bluetooth LE) and NFC are opening new possibilities for IoT.
Fog networks extend cloud computing and services to the edge of the network, positioning data processing power and storage closer to the things that are generating the data. Think of fog networks as “IoT LANs” or “local clouds” on the edge of the larger IoT. These configurations are necessary when the things generate large amounts of raw data that would be overwhelming—in terms of bandwidth, latency, etc.—to communicate in the cloud or when fast reaction-‐time requirements do not allow for long communication delays. The fog network can send summarized data (knowledge) to the cloud as needed.
IEEE-‐SA IoT Ecosystem Study
13
Copyright © 2015 IEEE. All rights reserved.
This configuration can also mitigate a number of other problems like unique addresses for things, quadruple trust issues, etc. It is likely that many of these local clouds will use a simplified communication technology (rather than IP) and rely on a gateway to provide an adaption layer to access the larger IP-‐based IoT.
Mesh networks and mobile ad hoc networks (MANETs) are other technologies proving useful in IoT applications. In these peer-‐to-‐peer networks, each node can be a router (i.e., forward traffic unrelated to its own use). This networking style is particularly useful for low-‐powered devices operating in close proximity.
Computing and storage The cost of both processors and storage has dropped dramatically while compute power and storage capacity have substantially increased.
This has made cloud computing and cloud storage rapidly growing sectors of the traditional Internet as well as playing a prominent role in the growth of IoT.
The availability of denser, lower cost local storage and local processors that are cheap, powerful, power efficient, and small enough to be integrated with I/O devices and sensors are arguably even more important to the growth of IoT.
Big-‐data analytics Big-‐data analytics can examine large volumes of data of a variety of types to find hidden patterns, unknown correlations and other useful information. Data sources include transaction data, server logs, Internet clickstream data, social media activity reports, mobile-‐phone call detail records, and sensor data.
Results can facilitate better business decisions, competitive advantages, more effective marketing, and can provide a new revenue source.
Missing from the technology point of view Today’s technology is enabling the deployment of IoT products. However, there are many areas where technological improvements are needed. As these improvements come on line they will improve existing products and open the door for new products that are not currently feasible. Improvements that are needed include
¾ Quadruple trust: protection, security, privacy, safety
¾ Sensor, actuator, and device improvements
¾ Networks and communications
¾ Interoperability
¾ Semantics and intelligence
¾ Big data
¾ Scalability
¾ Functional safety
Quadruple trust: protection, security, privacy, safety The quadruple trust is about the integrity of data. Users must have confidence that their data:
IEEE-‐SA IoT Ecosystem Study
14
Copyright © 2015 IEEE. All rights reserved.
¾ Is protected from loss or corruption
¾ Is secure from malicious attacks
¾ Is private from all but the intended users
¾ Is safe from unintentional loss, corruption, or disclosure
Failures of these trusts could be life threatening, could result in legal violations, and would certainly result in loss of user confidence with the concomitant consequences: retarded growth of the IoT ecosystem and potentially costly legal actions.
As with earlier developments of computer and networking technology, the quadruple trust has been an afterthought in IoT deployments. There is much work needed on quadruple trust in all areas of traditional computing and networking. The solutions that currently exist are porous. Since IoT systems are—by their very nature—decentralized, they present even greater challenges to establish and maintain the quadruple trust. There are currently no effective established methodologies in any of the quadruple-‐trust areas for IoT or for the sub-‐field of sensor networks.
Implementation of quadruple trust should not be approached from a use-‐case perspective. There should be an overarching framework of strong trust; the strength of pertinent properties can then be relaxed as appropriate for a given use case. This overarching framework is essential for interoperability.
Sensor, actuator, and device improvements Great strides have been made in sensors, actuators and devices; in order to enhance the growth potential of IoT, further improvements are needed.
Power consumption is one of the most important factors that will influence the growth of IoT. Sensors, actuators and devices must continue to reduce their power consumption. Considering the number of IoT products expected to be deployed, their power requirements—albeit small for each device—could, in aggregate, put a significant burden on the power grids (either directly while in use or indirectly when their batteries are charging). Improvements in battery capacity and advances in energy harvesting that make small sensors independent from batteries will both help to reduce power demand.
Sensors, actuators, and devices must also become dramatically less expensive in terms of both initial deployment and life-‐cycle costs. Manufacturers should build universal chips/modules that integrate many different sensors on a single chip. Multiple sensors on a single chip would not only reduce costs but could also enable features that would not be practical if multiple chips were required (e.g., the selected chip might include sensors not required for the application and these “optional” sensors might inspire designers to create new features). Having many different sensors on a single chip will also enable new applications on devices already deployed, as has been the case with smart phones.
Mobile devices may need to provide localization information quickly, precisely, and at low cost. Since GPS localization is not always available (e.g., indoors) new localization techniques are needed.
There are also concerns regarding how IoT sensor and actuator modules that have relatively short lifespans can be integrated with long-‐lived products like automobiles and appliances. Lifespan assessments are appropriate to determine how long sensor and actuator modules need to live in different applications and whether updates or upgrades or reconfiguration of the modules will be needed. Lifespan-‐cost reductions will create additional value.
IEEE-‐SA IoT Ecosystem Study
15
Copyright © 2015 IEEE. All rights reserved.
Networks and communications IoT products will present different possible data communication scenarios. Some may involve sensors that send small data packets infrequently (e.g., once a day) and do not prioritize timely delivery. Others may have storage in order to sustain periods when the communication link is down. Others may need high bandwidth but be able to accept high latency. Others may need high quality, high bandwidth, and low latency.
Depending upon the scenario, using large numbers of sensor nodes can require higher quality and higher reliability in the sensor network. The growing number of connected things can cause the amount of data that must be moved to explode. This could be mitigated by processing the raw data on the devices themselves or in a module close to the device (e.g., a gateway) to condense and/or aggregate data for more efficient transport to the ultimate destination. With many conversations among IoT devices and gateways, the communication network itself needs to be robust, high capacity, and always available with minimal congestion.
Large amounts of traffic with relatively short packet sizes will require sophisticated traffic management. More efficient protocols can help reduce overhead but may present challenges to system integrity, usability, scalability, etc.
Interface standardization is desirable so that IoT devices and gateways can communicate quickly and efficiently.
Devices will need a way to quickly and easily discover each other and learn their neighbor’s capabilities.
As mentioned earlier, adoption of IPv6 has lagged; global adoption will be necessary for the IoT to proliferate. IPv6 provides the following benefits to IoT configurations:
¾ IPv6 auto-‐configuration
¾ Scalable address space (sufficiently large for the enormous numbers of IoT devices envisioned)
¾ Redefined headers that enable header compression formats
¾ Easy control of the system of things
¾ Open/Standard communication
¾ IPv6 to IPv4 transition methods
¾ IPv6 over constrained node networks (6LO, 6LoWPAN)
Interoperability Interoperability is critical to foster competition and drive down costs. It is especially important for large IoT applications (e.g., smart city and smart grid). Interoperability is of paramount importance for building systems of IoT systems.
It is desirable that common IoT building blocks are usable in different verticals (i.e., service standardization). IoT building blocks should be able to interoperate horizontally (i.e., across verticals) so that individual IoT building blocks can simultaneously support multiple applications. These building blocks will be common architectural components to enable capabilities such as routing, security, prioritization, cross-‐domain API’s, etc.
Interface standardization is also needed, particularly for sensors, actuators, and smart devices.
IEEE-‐SA IoT Ecosystem Study
16
Copyright © 2015 IEEE. All rights reserved.
The goal should be for devices to be interoperable and interchangeable. There needs to be a clear means for devices to communicate and share information about their capabilities. A uniform electronic device description language would benefit this need.
An example of such cross-‐vertical interoperability is a medical device that can interface with service providers (to provide performance data) and manufacturers (to provide predictive maintenance data) while fulfilling their primary function of providing medical data to a doctor or clinic. The building blocks will need to know which data to expose through which socket (to which vertical) to maintain patient privacy, etc.
Semantics and intelligence Data from IoT devices will need to be provided in standard formats and with standard semantics that can be understood by other IoT devices and gateways. While some applications (e.g., medical) may have special needs, those special needs should be provided in a standardized way to support interoperability.
There will need to be rules for the abstraction of data, both to make the data pathways interoperable and to be certain that the quadruple trust is adequately maintained.
There are many instances where IoT devices could enhance a user’s experience by inferring the user’s intentions and providing services based on those inferences with minimal user interaction. For example, when a user sits on the sofa what are his/her intentions? If the user is alone, perhaps he/she wants to watch TV or a movie; if there are multiple people in the room perhaps they want to chat with soft music in the background. A smart entertainment system could use the current environment state, the user’s past behavior and possibly some earlier user configuration to show a TV program (based on the TV schedule and learned user preferences) or play music (based on known user preferences and possibly on emotion sensing).
As the sensing and semantic capabilities of IoT grow, more of these types of scenarios become possible. The fundamental rule is that the user requires simplicity; the scenarios presented to the user will need to be easy and seamless.
Big data As the individual morsels of sensor/device data move up the application they can be aggregated into larger quantities of related data. What can be done with this big data? Where is it stored? How easy/difficult is it to retrieve a piece of information from this big-‐data store? How is this big data massaged before it is provided to the customer?
How can/should the information be prioritized? Is there a way to detect if the incoming data is incorrect? What should the response be? Is there a way to automate some activities based on the information coming up from the lower layers?
These are only some of the technical and ethical questions that arise from the big-‐data aspect of IoT. And, of course, there are the questions of how this big data can be monetized.
Scalability As the number of IoT devices grows, different methods of managing them must be devised; the current methods simply do not scale. In addition to demonstrating the scalability problem, the examples listed here show how much overlap there is among the various market and technology needs.
In the smart home, consumers will inevitably need to replace a residential gateway, either because it is worn out or in order to upgrade for other benefits. Even in legacy environments where just a few
IEEE-‐SA IoT Ecosystem Study
17
Copyright © 2015 IEEE. All rights reserved.
computers are connected to the gateway, many consumers have difficulty when switching gateways. It is unreasonable to expect them to reconfigure each of the IoT devices—appliances, light bulbs, curtain controllers, etc.—when they replace their residential gateway.
Wireless technology is often cellular; cellular technologies utilize SIM cards to authenticated devices. Switching operators for cellular providers requires swapping out the SIM cards, presenting the problem of how to swap out SIM cards2 in multitudes of devices. Managing cellular networks will be problematic when there are hundreds of thousands of SIM cards needing to be replaced. More generally, managing devices will be a challenge when there are billions of connected devices.
A key aspect of this will be improvements to device life-‐cycle management. There must be ways to manage and update devices as they age. There will need to be better methods for managing device credentials.
Another scalability challenge is the development of meta-‐modeling tools for systems of IoT systems. The Smart-‐Grid Architecture Model is an example for such a tool.
Functional safety Functional safety is necessary to minimize the risk of physical injury or damage to the health of people directly or indirectly. With IoT, there will be more and more autonomous systems that will react to sensor data without human control or intervention (e.g., autonomous driving cars, robots). Hardware and software mechanisms must be designed to ensure the safe operation of things. This includes inherently safe algorithms, default states in case of failures, and exhaustive simulation and testing.
2 This may be a short-‐term issue. As the deployment of cellular devices with embedded SIMs progresses, this problem will diminish.
IEEE-‐SA IoT Ecosystem Study
18
Copyright © 2015 IEEE. All rights reserved.
IoT Standardization What standardization bodies are doing important work to enable the IoT? What specific standards activities do you think of when you think of the IoT? What is missing from the standardization point of view? Do you see IoT activities as more suitable for regional or global standardization?
Standards bodies doing important work to enable IoT There are many regional and international standards organizations actively pursuing IoT-‐related activities. IoT is growing rapidly and since standards organizations move relatively slowly, many alliances have arisen to fill the void, working to develop specifications, promoting industry collaboration, etc.
It will be important for these bodies to coordinate their work; it is not desirable to duplicate work that has already been done elsewhere.
The major standards bodies active in IoT include the following:
¾ Institute of Electrical and Electronics Engineers (IEEE) www.ieee.org
¾ International Electrotechnical Commission (IEC): www.iec.ch
¾ International Organization of Standardization (ISO): www.ios.org
¾ International Society of Automation (ISA): www.isa.org
¾ International Telecommunication Union (ITU): www.itu.int
¾ Internet Engineering Task Force (IETF): www.ietf.org
¾ World Wide Web Consortium (W3C): http://www.w3.org/
It must be understood that this is only a partial list of the standards bodies that are active in IoT development and standardization. Annex C contains a list of the organizations that were mentioned during the roundtables and subsequent review.
Specific standards activities related to IoT? In general, current IoT-‐specific standardization activities are confined to very specific verticals—e.g., energy management, health, wearables—and represent islands of disjointed and often redundant development. There are also a number of standards in communications and networking that are either directly applicable to IoT or are currently being extended to support IoT.
Within IEEE there is a new project, IEEE P2413™ Draft Standard for an Architectural Framework for the Internet of Things (IoT), that will define an IoT architectural framework, identify commonalities and relationships among various IoT verticals, define abstractions, provide a reference model, define architectural building blocks and provide mechanisms to develop multi-‐tiered systems from these building blocks. The goal is to provide the ability for verticals to interact with each other while still retaining the isolation required within each vertical (e.g., a medical device could provide medical information to the doctor, performance information to the service provider, and maintenance information to the manufacturer while still providing the necessary security and privacy within each vertical).
In the IEEE, there are more than 350 IEEE standards that are applicable to IoT, 40 of which are being revised to better support IoT. Furthermore, there are more than 110 new IoT-‐related IEEE standards in various stages of development. The IEEE is also sponsoring 10 or more different IoT advocacy and support groups. Links to these activities are provided in Annex C.
IEEE-‐SA IoT Ecosystem Study
19
Copyright © 2015 IEEE. All rights reserved.
IETF has also been looking into IoT issues for several years. These efforts include providing IPv6 on small devices and addressing several other issues that are protocol agnostic. See Annex C for specifics.
IoT-‐related standards work is also being done at the other major standards bodies identified in the previous section; specific activities were not enumerated in the roundtables.
Missing from the standardization point of view Among the items missing for IoT from the standardization point of view are
¾ Common definition of IoT
¾ Global reach and coordination
¾ Architecture and reference models
¾ Quadruple trust: protection, security, privacy, safety
¾ Scalability
¾ Application standards
¾ Interoperability
– Device interoperability within verticals
– Cross-‐vertical interoperability
¾ Other comments
Common definition of IoT While not a solution in and of itself, a common definition of IoT for the entire standards community will certainly simplify the coordination effort that was identified as a basic need.
Global reach and coordination Fragmentation of any sort (industry, vendor, regional, global) is highly detrimental to the IoT market. The challenge is to overcome the fragmentation that already exists and prevent additional fragmentation. Collaboration and coordination are key. It will be a huge challenge to resolve the fragmentation issue; everyone will need to keep in mind that “giving a little gets a lot.”
IoT standards need to have a global reach; some standards bodies do not have global reach. Thus, standards bodies may not just work in isolation anymore; they need to have more collaboration and agreement. They must coordinate, align, and interwork to avoid replication of functionality and to ensure that whatever functionality is defined by one of these bodies works with the functionality defined by the other relevant bodies. This need for collaboration and coordination exists between both vertical standardization efforts and horizontal standardization efforts.
Regional programs in smart manufacturing are giving way to efforts of broader scope through efforts like the IEC activity on Industry 4.0/Smart Manufacturing that moves the European Industry 4.0 and US Smart Manufacturing activities to a common global level.
Another collaborative effort along these lines is oneM2M (oneM2M.org), which was founded by seven major standards bodies “to ensure that Machine-‐to-‐Machine Communications can effectively operate
IEEE-‐SA IoT Ecosystem Study
20
Copyright © 2015 IEEE. All rights reserved.
on a worldwide scale.”3 oneM2M is working to consolidate a lot of regional work that has already been done.
The International Society of Automation (ISA) is very closely watching what IEEE and IETF do to see how to apply IoT to industrial automation.
Many IoT systems are being put together without waiting for standards bodies (which move too slowly to serve the rapid growth of IoT). Academia, research and commercial players are all developing IoT systems, sometimes independently and sometimes in collaboration with other players.
Open-‐source bodies are important in IoT. They are not standards bodies, but they may be the next best thing: providing non-‐proprietary platforms that can be used to build the applications that bring IoT to life. The AllSeen alliance is an example of an open-‐source alliance; it has assumed management of AllJoyn, an open-‐source project that lets compatible smart things recognize each other and share resources and information across brands, networks, and operating systems.
Another example is the OpenIoT project in Europe. OpenIoT is building middleware to get information from sensor clouds without having to worry about what sensors are being used.
Architecture and reference models As already mentioned, most current standards activity is insular. There is a need for a “standard of standards,” an architectural framework and reference model that accommodates all of the standards. This reference model needs to focus on creating the glue between all of the existing standards and providing the flexibility and scalability to accommodate future standards and business models.
Quadruple trust: protection, security, privacy, safety The quadruple trust is described in “Missing from the technology point of view.” technical aspects of the quadruple trust will be standardized separately, but the quadruple trust—protection, security, privacy and safety—must be considered and accommodated as part of every standard activity.
Scalability Mechanisms that have worked in the past simply do not scale to the magnitude anticipated by the explosive growth IoT brings. One issue is the need to have unique hardware identification for billions of things.
Unique hardware identification in the past has been provided by MAC addresses. The MAC address space (48-‐bit addresses) is supposed to last for 100 years; it is difficult to expand. Unfortunately, it will not last for 100 years if MAC addresses are given to things like light bulbs. The IoT requires that things be uniquely identified in a bigger address space (e.g., a 64 bit EUI-‐64 or similar address space) and then given a temporary layer-‐2 address (e.g., a MAC address) from the network. One consideration that needs to be addressed is privacy: MAC addresses can be used to track people as they move around. There is a need to preserve privacy (and the other pillars of the quadruple trust).
Many of the issues discussed elsewhere in this study (e.g., networking and communication between large numbers of devices) are also, at their heart, scalability issues.
3 Quoted material is from http://portal.etsi.org/Services/CentreforTestingInteroperability/Activities/M2M/oneM2M.aspx.
IEEE-‐SA IoT Ecosystem Study
21
Copyright © 2015 IEEE. All rights reserved.
Application standards Most standards to date address pieces of the IoT framework (e.g., communications and networking). There is a need for application standards that will enable interoperability between products in the application space.
Interoperability Interoperability is needed to break through the logjam of proprietary solutions to reduce industry fragmentation and build a successful IoT ecosystem.
Standards are needed to promote the interoperability of devices both within verticals and within networking and communications environments.
Standards are also needed for cross-‐vertical interoperability (e.g., the exchange of information and services between vertical applications).
Other comments In addition to the more general needs identified above, several specific needs were identified. The following is not a complete list of specific needs, just some of the many specific needs identified in the roundtables and reviews:
¾ There are no standards yet at L3/L2 gateway between 6lo and IPv6
¾ Need standards for
– Mobility across sensor networks
– Device discovery
– Security
– Naming
– Connecting non-‐enabled devices to devices that are enabled
– Predictability and guarantees for systems of IoT systems
– Linkage of value creation and business models with thing attributes and the information life cycle
– Layer-‐2 automation buses over layer-‐3 connections (CANopen, PROFIBUS, etc.)
The following lists identifies “dos and don’ts” for standards bodies that were identified during the roundtables. While most of these are “obvious” to people working in standards, it is valuable to note that industry recognizes the need for the right process. It is also important to note that IoT is moving forward and the standards development process needs to evolve to remain in sync.
¾ Standards:
– Need a purpose
– Should be visionary
– Need interoperability
– Need safety (something can be safe in one person’s hands, but unsafe in another’s)
¾ Start with a kernel, do not try to standardize the big vision. Document it, but do not standardize it. Do not “bake-‐in” limitations in the standard.
IEEE-‐SA IoT Ecosystem Study
22
Copyright © 2015 IEEE. All rights reserved.
¾ Which elements of the IoT should be standardized? Need to identify what is intrinsic regardless of where someone or something is (application and/or industry-‐wise and/or geographically).
¾ Look at successes and how they were achieved. Bring in past (i.e., existing) use cases; there is a valuable history of use cases to consider. Look to the past successes instead of the future. For example, in the early 1900’s light vehicles were invented. There was chaos in transportation in the early days (not even agreement on whether to drive on the right or left side of road); this situation is analogous to the current state of the IoT.
– Reflection is good, but models should not be carried forward when they no longer fit (e.g., people used the early auto as if it were a horse).
– Collect many use cases from many verticals. Find core features, as the kernel, to standardize. However, use cases may not be known yet.
– Need to address enough of the use cases to satisfy most players.
¾ It may be more difficult to create a standard when there are too many members. However, standards are about satisfying most of the people most of the time.
– There needs to be a balance between too many and too few voices
– There needs to be a level playing field for standards participation
¾ Need to have interoperability and security in all standards activity.
Global standardization There is almost universal consensus that global standardization is necessary.
Motivations for global standards include the following:
¾ Most of the existing challenges are not unique to any particular region.
¾ Companies at all levels of IoT want to be able to produce products and services that can be marketed globally.
¾ Global standards are particularly important because they will lead to economies of scale, which will in turn open up new markets.
Global standards will provide challenges such as accommodating the regulatory environments of different governments. However, the benefits of global standardization outweigh the challenges. Regional standards are only welcome if regional needs cannot be adequately treated within global standards; preferably regional needs should be addressed as extensions to and/or profiles of global standards.
There are significant regional issues that must be addressed, such as the following:
¾ Wireless regulations differ from country to country even within the EU
– Spectrum issues are regional
– Whitespace is a problem that has not been solved
– Bringing in governmental bodies can also be a challenge
¾ Security and privacy regulations vary widely by region
¾ Other regulatory issues
IEEE-‐SA IoT Ecosystem Study
23
Copyright © 2015 IEEE. All rights reserved.
Role of Academia and Research (AR) What can research, industry and academic institutions contribute?
Academia is educating new generations of technologists and business people who will support the growth of IoT.
While businesses are generally focused on specific products, many support “blue sky” research organizations within their corporate structure. These research organizations frequently collaborate with academia. This collaboration is beneficial, particularly in new areas like big-‐data analytics, the cloud, and systems of IoT systems. AR is creating the new theories, materials, devices, and apps that will fuel the growth of IoT.
At the application level, AR can combine and cross-‐pollinate application verticals to provide new applications. It is already providing insight into how APIs on smart phones can be used in new ways, leading to new applications. For example, Boston’s pothole tracking system mentioned earlier is the outgrowth of research done by Fabio Carrera, a professor at Worcester Polytechnic Institute. This pothole tracking system combines IoT devices (smart phones), communications and big data in a novel new way.
AR can also provide advancements in energy efficiency, security, cost reduction, etc.
AR can make available large test beds to enable organizations to deploy, test, and validate concepts. This would be especially helpful for small organizations that might not have the capacity for large-‐scale testing.
AR can support the early uptake and solidification of standards by implementing prototypes, discovering bugs/weaknesses, and providing feedback. Every standard needs to have a maturation phase; AR can shorten the duration of this phase.
IEEE-‐SA IoT Ecosystem Study
24
Copyright © 2015 IEEE. All rights reserved.
User Acceptance is Key Users of IoT were never mentioned as a player driving its growth because they play a passive role. However, consumer acceptance was identified as a key driver of speed of growth (i.e., it would not necessarily be the most useful products but the ones that received user acceptance that would succeed).
Several of the themes that kept coming up are clearly aimed towards user acceptance:
¾ Certainty of value
¾ Usability
¾ Maintainability
¾ Quadruple trust
¾ “It just works”
The degree to which progress is made in these areas will directly affect the speed at which IoT grows.
Certainty of value is a key requirement for user acceptance. If users do not see value, they probably will not accept an IoT service/application. An example is the smart home. While a smart home can be more convenient, more energy efficient, and more fun, today’s consumers see it primarily as an expensive toy for the tech set and a solution looking for a problem. IoT developers need to identify and address user needs and expectations rather than expecting users to adopt whatever is presented to them. Additionally, governmental and civic adoption could be vital in building the consumer comfort level.
Ease of use is a major concern for users/consumers. If the item is too complex to use or does not interoperate well, then consumers will simply give up on it. However, it is clear that the younger generation’s vision will become the reality; users will expect the benefits that IoT brings.
Users are more willing to release (share) their data if they feel they are gaining benefit. Thus one of the challenges for IoT is to position things so they are perceived as being positive for consumers rather than being specters of big government/big business looming over people. An example of this is Boston’s pothole tracking system, which enlists user participation in solving a problem that directly affects them (they are willing to participate in fixing the streets they drive on rather than refusing because of worries about being tracked by the city of Boston).
User benefits may explain why the two market segments identified as being the initial drivers of IoT growth are consumer goods and eHealth.
IEEE-‐SA IoT Ecosystem Study
25
Copyright © 2015 IEEE. All rights reserved.
Conclusions This study reflects the opinions of those participating in the roundtables, workshops, and reviews. It is basically a snapshot of the current state of the IoT ecosystem as seen by a collection of experts. It should not be construed as showing a complete picture of the IoT ecosystem; it is more of a step on the IoT evolutionary ladder and a basis for further discussion.
There are hotbeds of activity in many different IoT market segments. Early deployments exist or are imminent in many of these segments. However, these early deployments are fragmented, and the fragments are segmented and specialized. Fragmentation needs to be reduced for optimal growth of IoT.
For broad adoption, users need interoperable and portable systems that allow for ease of use and interchange. Companies, researchers, and standards bodies must work for this goal in order to grow the IoT market as envisioned for the future. The issues of security, privacy, and authentication must be considered as part of this evolving technology.
The following missing elements include many that are common for the market, the technology and standards development:
¾ Quadruple trust: protection, security, privacy, safety
¾ Interoperability:
– Device interoperability within verticals
– Cross-‐vertical interoperability
¾ Scalability
¾ Usability: Implementation and system integration
Missing market-‐specific elements include
¾ Silos
¾ Monetization
¾ Education
Missing technology-‐specific elements include
¾ Sensor and device improvements
¾ Networks and communications
¾ Semantics and intelligence
¾ Big data
Missing standards-‐specific elements include
¾ Common definition of IoT
¾ Architecture and reference models
¾ Global reach and coordination
IEEE-‐SA IoT Ecosystem Study
26
Copyright © 2015 IEEE. All rights reserved.
¾ Application standards
Organizations in different areas—research, industry, government, and academic institutions—are actively involved in IoT development. What is needed from these organizations is
¾ More collaboration and cooperation
¾ More involvement in international (global) standards
¾ More attention to user needs and wants
Standards organizations must collaborate with each other and with the other IoT players. Standards must be global for optimal IoT growth.
The challenges that IoT faces to become a fully functional ecosystem are those that are faced by the many components that are part of IoT. The issues that components like the smart grid or eHealth face combine to become the problems that could halt the implementation of a vibrant IoT. But a better understanding of the ecosystem of IoT will help deduce the solutions to move the vision of IoT into reality.
IEEE-‐SA IoT Ecosystem Study
27
Copyright © 2015 IEEE. All rights reserved.
Annex A IoT roundtables sponsored by IEEE-‐SA
IEEE-‐SA IoT Roundtable – Taipei Date: 5 June 2014
Organizer: IEEE Standards Association
Co-‐Organizers: Institute for Information Industry Semi Tatung STMicroelectronics
Participants:
Name Company/Institute/Industry Title
Logvinov, Oleg (Host, Moderator)
STMicroelectronics Director, Special Assignments
Wen-‐Yen K. Lin
(Co-‐host)
Tatung President
Ming-‐Whei Feng (Co-‐host)
Institute for Information Industry Vice President & General Director
Eduardo Cerristos National Chiao Tung University Student
J. Y. Chen SPIL 矽品 Manager
Ping Chen 陳秉毅 D-‐Link Chief Technology Officer
F.C. Cheng 鄭福炯 Tatung University Professor, Dept. of Computer Science & Engineering
Rutgers Chow SPTS Technologies, Ltd. Vice President of Asia Field Operations
Weileun Fang 方維倫
National Tsing Hua University Distinguished Professor
C. Fu 傅強 GUC 創意電子/TSMC Vice President
Mavis Ho SEMI Vice President
IEEE-‐SA IoT Ecosystem Study
28
Copyright © 2015 IEEE. All rights reserved.
Name Company/Institute/Industry Title
Ken-‐Ching Huang UMC 聯華電子 Director, Corporate Marketing
C. P. Hung ASE Group 日月光半導體 Vice President
Giuseppe Izzo STMicroelectronics Vice President of GC&SA
Emery Jou Institute for Information Industry Advisory Engineer, Smart Network System Institute
Charles Kao Episil Technology 漢磊科技 Director, Equipment R&D
Chung-‐Ta King 金仲達
National Tsing Hua University Professor
Albert Lan SPIL 矽品 Senior Director
Joey Lee WNC 啟 碁 科 技 R&D Manager
Chih-‐Kung Lee 李世光
National Taiwan University Distinguished Professor
Yi-‐Bing Lin 林一平次長
Ministry of Science and Technology 科技部
Political Deputy Minister
C. P. Lin 林常平 Tatung Co. Senior General Manager, Smart Grid BU
Rung-‐Bin Lin 林榮彬 Yuan Ze University 元智大學 Professor, Dept. of Computer Science & Engineering
Phone Lin 林風 National Taiwan University Professor, Dept. of Computer Science & Infor. Eng.
Rita Liu WNC 啟 碁 科 技 R&D Manager
Ray Tain Unimicron Tech. Corp. 欣興電子 Deputy Director
Stephen Tsai SPIL 矽品 Manager
Luca Tseng Cyntec 乾坤科技/Delta Senior Director
IEEE-‐SA IoT Ecosystem Study
29
Copyright © 2015 IEEE. All rights reserved.
Name Company/Institute/Industry Title
Webber Wang C Sun 志聖工業 President
Dustin Wu ASML Product Manager
Sog Yang Cyntec 乾坤科技/Delta Senior Manager
Susan Yao SEMI Senior Manager
Aaron Yu CyberTAN 建漢科技/Foxconn BU Head
IEEE-‐SA IoT Roundtable – Washington DC Date: 9 June 2014
Organizer: IEEE Standards Association
Participants:
Name Company/Institute/Industry Title
Mary Lynne Nielsen (Host, Moderator)
IEEE Standards Association Technology Initiatives Director
Isaac Chan US Department of Energy Supervisory General Engineer
Walton Fehr US Department of Transportation Manager, ITS Systems Engineering
Julian Goldman Medical Device Plug and Play/Mass General Hospital
Medical Director, Biomedical Engineering, Partners HealthCare Staff Anesthesiologist, Massachusetts General Hospital (MGH) Director, Medical Device Interoperability Program, MGH Chair, ISO TC 121 Chair, ASTM F29 Committee
Christian Lederer CISC Semiconductor GMBH Senior Software Engineer
Rainer Matischek Infineon Technologies Austria AG Senior Research Engineer
William Miller Maximum Control Technologies (MACT-‐USA)
President
IEEE-‐SA IoT Ecosystem Study
30
Copyright © 2015 IEEE. All rights reserved.
Name Company/Institute/Industry Title
Bakul Patel US Food and Drug Administration Senior Policy Advisor Center for Devices and Radiological Health
Michael Taborn Intel Platform Architect, IOT Solutions Group
IEEE-‐SA IoT Roundtable – Santa Clara, CA Date: 17 September 2014
Organizer: IEEE Standards Association
Participants:
Name Company/Institute/Industry Title
Mary Lynne Nielsen (Host, Moderator)
IEEE Standards Association Technology Initiatives Director
Yoshiaki Adachi Hitachi
Martin Bauer NEC Laboratories Europe Senior Researcher
Josef Blanz Qualcomm Principal Engineer
Samita Chakrabarti Ericsson Principal Software Engineer/Architect
Penny Chen Yokogawa Principal Systems Architect
Jean-‐Pierre Desbenoit Schneider Electric ICT Standardization and Industry Relations Director
Rob Gillan dZhon Pty Ltd Chief Technology Officer
Juergen Heiles Siemens AG Standardization Manager
Taizo Kinoshita Hitachi Vice President
Victor Kueh Huawei
Oleg Logvinov STMicroelectronics Director of Special Assignments
IEEE-‐SA IoT Ecosystem Study
31
Copyright © 2015 IEEE. All rights reserved.
Name Company/Institute/Industry Title
Stefan Lueder Siemens Corporate Research Senior Scientist
Yuichi Nakamura Hitachi Solutions Senior Researcher
Philippe Nappey Schneider Electric Solution Architect
Nobuyuki Ogura Hitachi Executive General Manager, Platform Development
Tobin Richardson Zigbee Alliance President and CEO
Eric Rotvold Emerson Distinguished Technologist
Francesco Russo IBD
Daniel Smolinski Renesas Electronics Marketing Manager
Padmakumar Subramani Alcatel-‐Lucent Product Manager
Pat Thaler Broadcom Senior Technical Director
Ludwig Winkel Siemens AG Fieldbus Standards Manager
IEEE-‐SA IoT Ecosystem Study
32
Copyright © 2015 IEEE. All rights reserved.
Annex B Other meetings that provided input to this ecosystem study
IEEE ComSoc IoT Rapid Reaction Standardization Activity Working Meeting Date: 30 September 2014
Location: IEE Headquarters, Piscataway, NJ USA
Organizer: Standards Activities Council of IEEE Communications Society (ComSoc)
Participants:
Name Company/Institute/Industry
Wilbert Adams Huawei, USA
Lillie Coney US House of Representatives, USA
Mahmoud Daneshmand Stevens Institute, USA
Adam Drobot OpenTechWorks, USA
Omar Elloumi Alcatel-‐Lucent, France
M. Hussain Fazil Consultant, USA
Rob Fish IEEE Communications Society
Alex Gelman IEEE Communications Society
Yacine Ghamri-‐Doudane University of La Rochelle, France
Stefano Giordano University of Pisa, Italy
T. Russell Hsing Consultant, USA
Myung Jong Lee City College of CUNY, USA
Kevin Lu Consultant, USA
R.R.Venkatesha Prasad Delft University of Technology, NL
Ming-‐Jye Sheng Nextcomm, USA
IEEE-‐SA IoT Ecosystem Study
33
Copyright © 2015 IEEE. All rights reserved.
Name Company/Institute/Industry
JaeSeung Song Sejong University, South Korea
Mehmet Ulema Manhattan College, USA
M. Can Vuran University of Nebraska-‐Lincoln, USA
Chonggang Wang InterDigital, USA
IEEE-‐SA IoT Ecosystem Study
34
Copyright © 2015 IEEE. All rights reserved.
Annex C IoT standards and activities This annex contains lists of organizations and activities related to IoT. This list comprises organizations that were identified during the roundtables or review. It does not attempt to be a complete list. It is not even clear that a complete list is possible because IoT is growing so rapidly.
The list of IEEE-‐based standards and activities is too large to be included in this study. It may be found at http://standards.ieee.org/innovate/iot/stds.html. A list of IEEE projects under development that are related to IoT can be found at http://standards.ieee.org/innovate/iot/projects.html.
An IETF presentation at the Santa Clara roundtable presented the following list of IETF activities, which include providing IPv6 on small devices and several other issues that are protocol agnostic. Building on the success of 6LoWPAN (which supports IPv6 on IEEE 802.15.4™), IETF has the following active working groups (see www.ietf.org for additional details):
¾ 6Lo—IPv6 over Networks of Resource-‐constrained Nodes (extending 6LoWPAN to additional layer 2 technologies)
¾ 6man—IPv6 Maintenance
¾ 6TiSCH—IPv6 over TSCH mode of IEEE 802.15.4e™
¾ ACE— Authentication and Authorization for Constrained Environments
¾ ROLL— Routing Over Low power and Lossy networks
¾ DICE— DTLS In Constrained Environments
¾ LWIP— Light-‐Weight (IP) Implementation Guidance
Other organizations and activities include
¾ 3rd Generation Partnership Project (3GPP): http://www.3gpp.org/
¾ Alliance for Telecommunications Industry Solutions (ATIS): http://atis.org/
¾ Allseen Alliance: https://allseenalliance.org/
¾ Bluetooth® SIG: https://www.bluetooth.org/en-‐us
¾ Broadband Forum (BBF): http://www.broadband-‐forum.org/
– TR-‐069: www.broadband-‐forum.org/technical/download/TR-‐069.pdf
¾ Consumer Electronics Association (CEA): http://www.ce.org/
¾ Digital Living Network Alliance (DLNA): http://www.dlna.org/
¾ Eclipse M2M Industry Working Group: http://eclipse.org/org/workinggroups/m2miwg_charter.php
¾ European Telecommunications Standards Institute (ETSI): http://www.etsi.org/
¾ GSM Association (GSMA): http://www.gsma.com/
¾ Health Level Seven International (HL7): www.hl7.org/
¾ Home Gateway Initiative (HGI): http://www.homegatewayinitiative.org/
IEEE-‐SA IoT Ecosystem Study
35
Copyright © 2015 IEEE. All rights reserved.
¾ Industrial Internet Consortium (IIC): www.iiconsortium.org
¾ Institute of Electrical and Electronics Engineers (IEEE) www.ieee.org
– IEEE 802.15.4: http://ieee802.org/15/pub/TG4.html
– IEEE P2413: http://standards.ieee.org/develop/project/2413.html
¾ International Electrotechnical Commission (IEC): www.iec.ch
¾ International Organization of Standardization (ISO): www.iso.org
¾ International Society of Automation (ISA): www.isa.org
¾ International Telecommunication Union (ITU): www.itu.int
¾ International Telecommunication Union -‐ Telecommunications (ITU-‐T): http://www.itu.int/en/ITU-‐T/Pages/default.aspx
– ITU-‐T Focus Group M2M: http://www.itu.int/en/ITU-‐T/focusgroups/m2m/Pages/default.aspx
¾ Internet Engineering Task Force (IETF): www.ietf.org
¾ Internet Protocol Smart Objects (IPSO) Alliance: www.ipso-‐alliance.org
¾ IoT European Research Cluster (IERC): http://www.internet-‐of-‐things-‐research.eu/
¾ oneM2M: oneM2M.org
¾ Open Interconnect Consortium (OIC): http://openinterconnect.org/
¾ Open Mobile Alliance (OMA): http://openmobilealliance.org/
¾ OpenIoT: http://openiot.eu/
¾ Organization for the Advancement of Structured Information Standards (OASIS): https://www.oasis-‐open.org/
– OASIS Message Queuing Telemetry Transport (MQTT): http://mqtt.org/
¾ Personal Connected Health Alliance (PCHA): http://www.continuaalliance.org/pchalliance
¾ SAE International (SAE): http://www.sae.org/
¾ Smart Grid Interoperability Panel (SGIP): http://www.sgip.org/
¾ Smart Manufacturing Leadership Coalition (SMLC): www.smartmanufacturingcoalition.org
¾ Thread Group: http://www.threadgroup.org/
¾ Weightless SIG: http://www.weightless.org/
¾ World Wide Web Consortium (W3C): http://www.w3.org/
¾ Zigbee Alliance: http://zigbee.org/