The Case for IPv6 as an Enabler of the Internet of Things
Many discussions of the Internet of Things (IoT) appear to assume that IP address space is an unlimited resource that will scale as the IoT scales to previously unimagined proportions. But the IP address space is not unlimited. In fact, the IPv4 address space has been depleted since February 2011. And that could have been the single best reason to consider IPv6 – Internet Protocol version 6 – for the future of IoT. Research has demonstrated that many other reasons exist as well.
IPv6 represents the next iteration of IPv4, which has provided just 232 Internet addresses or about 4.3 billion addresses, which were largely exhausted by the Internet Assigned Numbers Authority (IANA) by Feb. 1, 2011. In contrast, IPv6 enables 2128 IP addresses equivalent to 3.4 x 1038 addresses or 340 trillion trillion trillion addresses – enough, arguably, for generations to come.
Scalability is but one factor in IPv6’s favor. IPv6’s manageability, end-to-end connectivity, as well as its current rate of uptake are also encouraging. In terms of manageability, IPv6 can be compressed into a couple of bytes by using 6LoWPAN translation, an advantage over IPv4, and this enables its use in low-power sensors that may proliferate in IoT deployment scenarios. In terms of connectivity, two end points anywhere in the world can communicate via IPv6 with the properties of a virtual private network (VPN).
In terms of uptake, leading wireless telecom operators in the United States have been the first to adopt IPv6, based on the need to serve a vast and increasing number of mobile devices. And major web services and online applications, such as Google, YouTube, Netflix and LinkedIn have adopted IPv6 to handle massive global traffic between IP addresses. Internet Service Providers’ chokehold on the "last mile," however, remains an impediment to a highly functional IoT. Figure 1 shows the adoption curve for global IPv6 traffic on Google servers describes the proverbial "hockey stick" – IPv6 adoption is doubling every nine months. Countries such as Belgium already achieved over 32% IPv6 adoption rate and 52 Million US end-users today use IPv6 to reach Google without even knowing it.
Figure 1: Google IPv6 statistics
Traditionally, the proprietary protocol approach was built into many if not most enterprises’ business case, so governments, vertical industries, engineers and application developers, among others, had to be persuaded of IPv6’s superior qualities as an enabler of an unfettered IoT and, presumably, resulting market growth. When analyzing the emerging IoT and M2M-related standards, they all tend to converge toward the adoption of IPv6 and its 6LoWPAN compressed version as the reference network protocol. This is the case for 6TiSCH (IETF), LWM2M (OMA) and OneM2M. The adoption of IPv6 down to the last miles of IoT deployments is a growing reality.
The following account of IPv6’s origins, its advantages to IoT, use cases and costs/benefits is based on research by the IoT6 (www.iot6.eu), a three-year FP7 European research project that has explored the issue in depth.
A brief history of IPv6
IPv6 was designed by the IETF IPng (Next Generation) Working Group in the 1990s and it has been promoted by the IPv6 Forum since 1999. Several IETF working groups worked on expanding the IPv4 protocol suite to arrive at the next iteration, which is IPv6. These groups sought a larger address space, defined new capabilities and explored deployment scenarios with transition models that recognize the need for IPv6 to interact with legacy IPv4 infrastructure and services. These working groups have also enhanced a combination of features that were not tightly designed or scalable in IPv4, such as IP mobility, to enable lowest-cost deployment of large-scale sensor networks wherever networking adds value.
Benefits of IPv6 to IoT
The next logical step from networks of mobile devices to networks of communicating "Things" is IoT. That next step will mirror the sequence of events experienced by mobile networks. Proprietary protocols came first, because an individual company’s profits often come before consideration of the common good. But the use of IP and transparency (i.e., open source protocols) is fundamental to IoT development, just as the ease of use and the invisibility of the technology is important to end users. Our view, based on our research, is that the value of transparency and ease-of-use, and even more importantly the need for interoperability, will favor IPv6 adoption by the IoT market.
Specific aspects of IPv6 that will propel its use for IoT include, among others:
- Growing global adoption and product availability
- Overcoming Network Address Translation (NAT) limitations
- IP security enablement
- Mobility support
- Stateless Address auto-configuration
- Redesign of many features that existed in IPv4 but are greatly improved in IPv6: e.g. multicast, multiple addresses
We have discussed the first two features. The other features are summarized here:
Network Address Translation (NAT) enables several users and devices to share the same public IP address by associating with it several distinct private ones. This solution has been used mainly to cope with internet growth. It has serious disadvantages for IoT, however, by limiting the accessibility of those private addresses. A device can be accessed only if it has first contacted the application and maintained an open channel. Moreover, NAT breaks end-to-end connectivity and may weaken authentication processes. Use of NATs makes it more difficult to share a sensor infrastructure with several different IoT application providers. In contrast, IPv6 enables a highly scalable and NAT-free network deployment.
In terms of IP security enablement, IPv6 can provide end-to-end connectivity with a more distributed routing mechanism, thus, network intermediaries such as gateways are eliminated and the network is focused on routing and switching units, at the same time security vulnerabilities are reduced, since security is not broken by intermediary entities (e.g., NAT routers). Moreover, IPv6 is supported by a large community of users and researchers – including IPSec – investigating support and on-going improvement of its security features. (Internet Protocol Security (IPsec) refers to a protocol suite for securing IP communications through the authentication and encryption of each packet in a session.) The additional use of identifiers and multiple IPv6 addresses for the same end point enables different security and access properties for different IoT Applications Providers accessing the same IoT sensor/actuator infrastructure.
IPv6 provides features and solutions to support the mobility of end-nodes, as well as the mobility of the routing nodes of the network. Some of these features also work with IPv4, but are inefficient in how they were defined for that protocol.
IoT deployment will be massive, thus it’s particularly important that as many activities as possible occur in an automated fashion. One enabler of automation is IPv6’s stateless address auto-configuration (SLAAC), which allows a device new to the IoT network to obtain its IPv6 address autonomously without human intervention and related costs.
Multicast allows an operation to be performed on multiple devices; this was also supported in IPv4 but was unusable in its IPv4 version because it could disturb normal operation if it was misconfigured. It is very valuable in IoT operations that can be done at the network level. Multiple IP addresses for a single interface were also possible theoretically in IPv4; address shortage made its usage impractical. It is particularly valuable in IoT applications, where multiple stakeholders are using the same IoT physical deployments. Each stakeholder may use its own address space for the same end device.
At least another half-dozen features of IPv6 could be described, but would require us to digress into too much technology detail for this high-level paper.
The biggest hurdle cited by many observers to IPv6 adoption for IoT is the apparent lack of interoperability between IPv4 and IPv6. The perception of this hurdle was based on the notion that the transition from IPv4 to IPv6 would take a long time. This led to efforts to keep IPv4 working, mainly by artificially expanding its address space, with the idea that at some point the entire world would switch over to IPv6.
But a secure, practical transition from IPv4 to IPv6 would place the two protocols side-by-side in what we call a "dual-stack" approach. When an IPv4 device calls a dual-stack device, for instance, the latter would receive that call using IPv4.When an IPv6 device calls, the receiver would use IPv6. Can a dual-stack approach be maintained, based on cost? Mobile operators have resolved that question by making all new mobile devices with IPv6 and converting IPv4 to IPv6 in the central network.
But there are as many solutions to this potential hurdle as there are vertical industries. Thus, the dual-stack approach may be necessary until a market tipping point occurs in the deployment of IPv6-enabled devices and sensors – the "Things" in IoT. At that point the manufacturers of sensors and devices and "things" may have to agree to go all IPv6.
Another area that will require major advances in research and development is improving IPv6’s security and privacy capabilities. These qualities must be implemented in the design phase, or "baked in," not as an afterthought, or "bolted on." Obviously, this will influence user and market confidence, because with the anticipated proliferation of "Things" in IoT – everyone is talking about 50 billion connected devices not too far down the road – comes an exponential increase in potential attack surfaces. This issue must be addressed head-on.
A third major obstacle to the adoption of IPv6 is the current state of address translation limitations. Currently, most devices in an IoT context do not support IP or IPv6. They have their own proprietary addressing schemes, such as RFID (radio-frequency identification) numbers. That’s a small, closed universe. To track a product across the world, for instance – say, a cow from Brazil, shipped to a slaughterhouse in the U.S. and then distributed as cuts of meat to a thousand places – one could link the RFID number to an IPv6 address, at least the Host ID part of it. Thus address translation is critical to global adoption of IPv6, because most industries currently are using proprietary, closed IoT systems of their own.
A path to adoption
The single most efficacious path to IPv6 adoption is to get national governments around the world on board. They have the budgets to create a business case for manufacturers of IPv6-based devices and networks. And that creates economies of scale that will motivate engineers and developers to focus on a truly global, open source IoT. Adoption by large governments will probably be necessary to persuade vertical industries to abandon proprietary communication protocols in favor of an open source protocol such as IPv6 for an open, transparent, useful, worldwide IoT. In any case, considering the evolution of the global internet infrastructure towards IPv6 and the perfect match between IPv6 and IoT requirements in terms of scalability and mobility, the question is not any more whether IPv6 will be at the core of the global IoT architecture, but how fast.
Sébastien Ziegler is the founder and Director of Mandat International, a foundation based in Geneva with special consultative status to the UN and a member of the International Telecommunication Union. He graduated in international relations at the Graduate Institute of International Studies in Geneva, followed by a Master in Environment, an MBA in international administration (HEC Geneva), and complementary executive courses at Harvard Business School, Stanford University, UC Berkeley and EPFL. Sébastien founded two foundations, as well as two ICT-related SMEs and several organizations, including ICT-related alliances. He is Vice President of the IoT Forum and Vice Chair of the IEEE ComSoc Subcommittee on the IoT. He initiated several national and international research projects in the area of ICT, with a focus on Internet of Things, IPv6, multiprotocol interoperability and crowdsourcing. He initiated and coordinates several European research projects on IoT, including IoT6 (www.iot6.eu), IoT Lab (www.iotlab.eu) on IoT and crowdsourcing, and Privacy Flag on privacy and personal data protection.
Peter Kirstein is Professor of Computer Communications Systems at University College London. He is a fellow of many professional bodies including the Royal Academy of Engineering, American Academy of Arts and Science, US National Academy of Engineering. He has received many awards including the Commander of the British Empire, SIGCOMM, Postel, Lifetime achievement of Royal Academy of Engineering and Marconi.
Peter has led many projects in computer networks, communications and applications – both National and EC – included IPv6 activities in public safety, videoconferencing, security and sensor networking. He led the Networking Work-packages in both U2010 and IoT6.
Latif Ladid holds the following positions: Founder & President, IPv6 FORUM; Founder & Chair, 5G World Alliance; Chair, ETSI IP6 ISG; Chair, IEEE ComSoc 5G MWI & IoT subTC; Emeritus Trustee, Internet Society; Board Member IPv6 Ready & Enabled Logos Program. He is a Research Fellow at the University of Luxembourg on multiple European Commission Next Generation Technologies IST Projects. He is also Board Member of 3GPP PCG (www.3gpp.org), member of UN Strategy Council, and member of the Future Internet Forum EU Member States (representing Luxembourg).
Antonio F. Skarmeta received the M.S. degree in Computer Science from the University of Granada and B.S. (Hons.) and Ph.D. degrees in Computer Science from the University of Murcia, Spain. Since 2009 he is a Full Professor at the Computer Science department of the University of Murcia. Antonio Skarmeta has worked on different research projects in the national and international area in the networking, security and IoT areas, such as Seinit, Deserec, Enable, Daidalos, SWIFT, IoT6, SMARTIE and SocIOtal. He has published over 90 international papers and is a member of several program committees. He has also participated in several standardization activities being co-author of some drafts at the IETF.
Antonio Jara has received two Master Sciences (Hons. – valedictorian) degrees: a Master in Business Administration – MBA (Hons), and PhD (Cum Laude). He is especially focused on the design and development of new protocols for security and mobility for the Internet of things, the topic of his Ph.D. He is currently working on IPv6 technologies for the Internet of Things in projects such as IoT6, and also Big Data and Knowledge Engineering for Smart Cities in collaboration with projects such as SmartSantander.
Antonio has published over 80 international papers about the Internet of Things and holds one patent in the Internet of Things area. His specialties are: Internet of Things, IPv6, Future Internet, e-health, AAL, healthcare, 6LoWPAN, RFID, Bluetooth Low Energy, IoT6, and security. Antonio is interested in developing new proposals for H2020 in the areas of IoT, Big Data, mHealth, IPv6, SDN and User Experiences.
Subscribe to the Newsletter
Join our free IoT Technical Community and receive our Newsletter.
Calendar of Events
IEEE ComSoc International Communications Quality and Reliability (CQR) Workshop
16-18 May 2017
Naples, FL, USA
Internet of Things World
16-18 May 2017
Santa Clara, CA, USA
IEEE IoT Vertical & Topical Summit for Agriculture
21 May 2017
2nd Convergent Internet of Things (C-IoT) Workshop
21-25 May 2017
IEEE International Symposium on Circuits & Systems (ISCAS 2017)
28-31 May 2017
Baltimore, Maryland, USA
2017 Global Internet of Things Summit (GIoTS)
6-9 June 2017
The 1st EAI International Conference on Smart Grid Assisted Internet of Things (SGIoT 2017)
11-13 July 2017
Sault Ste. Marie, Ontario, Canada
Call For Papers
IEEE Internet of Things Journal
Multimedia Big Data in Internet of Things
Deadline: 31 May 2017
Emerging Social Internet of Things: Recent Advances and Applications
Deadline: 15 June 2017
Trust, Security and Privacy in Crowdsourcing
Deadline: 1 July 2017
Feedback Control for the Internet of Things
Deadline: 15 July 2017
Software Defined Networking for Internet of Things
Deadline: 1 August 2017