Scarcely a week goes by without a new would-be standard emerging for the internet of things (IoT). The main thing they have in common is that they promote the political agendas of their founding members, as a number of major players from the web, mobile, industrial and operator communities all seek to play a defining role, and ensure an important place in the evolving value chain. However, it is easy to become distracted by the power games of the overlapping membership rosters, and forget to ask whether any of these groups are doing something more useful than furthering a particular company’s favored technology.
The latest addition, the Thread Group, immediately arouses scepticism because of two of its founders – Nest, the smart home gadgets maker owned by Google, and Samsung (others are ARM, Freescale, Silicon Labs, Yale Security and ceiling fan maker, Big Ass Fans).
Samsung and Google both have devices as their entry point to the smart home, and from there the broader IoT, but have ambitions to influence the whole stack, using ‘open’ vehicles to try squeeze mutual arch-rival Apple back behind its garden walls.
Nest is currently being run at a distance by its new parent, to avoid conflicts of interest in the Android smart home ecosystem and to calm fears about the search giant snooping on users via their smoke alarms. However, its price tag is nothing to do with some neat home appliances and everything to do with Google wanting to define and drive the whole smart home stack, as it does the mobile stack via its relationships with handset makers, chip providers and developers.
So getting a firm grip on the personal area networking protocol which will tie millions of things together in the home or office will be important to Google, and this is the area Thread addresses ‘ in contrast to some other recent groupings, such as the AllSeen Alliance, based on Qualcomm’s AllJoyn technology, and the Open Interconnect Consortium, led by Intel. These two work at a higher, software level and are concerned with automatic discovery and linking of ‘things’, regardless of the underlying network. Indeed, Samsung is in OIC as well as Thread.
By contrast, Thread is looking to standardize the physical links between objects in a home around the IEEE 802.15.4 radio specification, to ensure interoperability for sensors and devices from all vendors. Several standardized and proprietary network protocols run over 802.15.4, most famously ZigBee and 6LoWPAN, and the 802.15.5 low power mesh. Thread is initially heavily focused on 6LoWPAN, because it is already used by Nest, and because it supports IPv6, important to ensure the IoT is future-proofed against running out of address space. 6LoWPAN is effectively a version of IP for the embedded space, providing a compression format for IPv6 that is optimized for low power, low bandwidth wireless links.
The protocol uses AES encryption at the 802.15.4 MAC layer, and claims it will offer a new and powerful authentication technique which uses “pieces of existing technologies wrapped into something new’, with all traffic encrypted, and all nodes authenticated by default. The protocol operates at the UDP level and supports any IPv6 application layer protocol, and could also act as a bridge between 802.15.4 networks and WiFi PCs or mobile products.
But the new body also hopes to lure the larger base of ZigBee developers, claiming many ZigBee devices could be upgraded to support Thread with just a software update. Attracting a home-focused ZigBee company like GreenPeak would be a valuable endorsement in the first major target market, the smart house and indeed, without ZigBee support, Thread will have an uphill battle in terms of gaining critical mass quickly. To do so, Thread will add software to the 802.15.4/IPv6 foundation, for functions such as routing, set-up, security and device wake-up, to standardize these capabilities and reduce power. The Thread group will provide testing and certification for its specifications, emulating WiFi and Bluetooth rather than the more splintered ZigBee.
The use of well understood technologies is a positive, as is the recognition that different markets will require different approaches to IoT networks ‘ Thread will limit its work to residential systems, saying:
“We are not trying to solve for industrial use cases, because they have other requirements.” That will avoid conflicts ‘ or changes to cooperate ‘ with consortia working on standards for other segments, such as GE’s Industrial Internet effort.
But Thread is not just here to put more commercial impetus behind a rather under-the-bonnet standard. It would hardly be an industry group if it did not add some sauce of its own ‘ which, of course, will be contributed freely but, of course, will provide its originators with a new level of influence. It claims this is essential because each of the existing standards it evaluated had at least one big failing. Wi-Fi has single points of failure and is unsuited to mesh; ZigBee Pro lacks native IP support and is limited to meshes of 32 nodes (rather than 250 in Thread); Z-Wave also lacks native IP and was considered too closed and complicated; even 6LoWPAN, which came closest to ticking all the boxes, lacks some of the security features Thread will add.
Chris Boross, president of the new organization (and, predictably enough, from Nest), said: ‘We wanted to use something off the shelf, but we knew that we had to do something new to make the best products’, adding that some Nest products already use an early form of Thread, rather than vanilla 6LoWPAN. And with those words, Thread was revealed as a de facto standards group like the others, not the purer beast it had briefly appeared ‘ its core technology will be something developed by Google’s smart home division, and Google, as Android shows, is a master of open sourcing its inventions while really keeping them under its control.
So, Thread is not battling AllSeen and OIC, but the established wireless PAN standards, all of which are chasing considerable expansions of their addressable markets via the IoT, especially in smart homes, cities and vertical sectors. ZigBee has established strengths in the industrial sector, and Z-Wave in the home, while Bluetooth Smart is well entrenched in mobile connected devices. The size of that mobile ecosystem makes Bluetooth the one to beat, since it is present in most connected consumer devices and supports Android and iOS.
That means that Thread is seeking to convert ZigBee, as well as the existing 6LoWPAN community, with the promise of a combined 802.15.4 platform, driven by Google’s power and optimized for home (as opposed to mobile) uses to a greater extent than Bluetooth. This would also bring IPv6 to ZigBee, and could help to unify the notoriously fragmented platform, at least in the home environment, even if the trade-off was adopting a’glue’ essentially devised under Google’s auspices.
The conversion of ZigBee is by no means a foregone conclusion. In a statement, the ZigBee Alliance said it was ‘closely reviewing the details of this announcement’ but did not sound as though it was going to leap into bed with Nest too readily. It said: ‘Thread appears different from ZigBee, which addresses both the network and application level necessary for success in the IoT device market and has led ZigBee to be deployed in tens of millions of devices around the world. Thread does not seem to provide the higher-level standardization to ensure interoperability between devices.’
It went on: ‘Thread is the latest in the recent influx of many announcements and initiatives … The ZigBee Alliance will evaluate Thread and its impact as well as the opportunity for the market and our members; we have a long history of collaboration with other organizations that provide specific solutions for a particular market segment. ‘ However, it ended on a more positive note, pointing out that five of the seven Thread Group founders are also members of its own Alliance, and noting that ‘this connection makes it that much more likely to develop a path for collaboration’.
Perhaps the promise of Google’s market making abilities will soften the Alliance’s attitudes, especially if it gains a better chance of overtaking Bluetooth Smart in the home. Bluetooth’s power lies in mobile devices, while in the smart home, where most of the large variety of sensors will not have to move very much, it has some disadvantages, largely its problems with meshing. Bluetooth Smart cannot form meshes that don’t rely on a single device (like a handset) acting as a controller, and though there are efforts to add this capability to future iterations, it will not be a quick task.
By contrast, 802.15.4-based protocols like ZigBee and Thread can form multipoint meshes (of up to 250 items, says the new consortium), which are resilient, because they can reroute connections if one device fails ‘ valuable in applications such as home security, which need to be robust without any human monitoring the system from a controller. This also extends the range of the connections as they can travel through several hops.
Whether or not Bluetooth catches up in this area, Thread will hope to take advantage of this particular headstart to build up a groundswell of support, in particular, it appears, by persuading ZigBee players to add Thread support. The appeal for ZigBee stalwarts – which include Samsung, the only major handset maker to propose implementing the protocol in mainstream smartphones ‘ is that the industry weight of Google might give their standard a boost in the home market, where it has had far less impact than in the industrial world – which could, not doubt, be Thread’s next target ‘ even though it is embedded into most hubs and some set-top boxes (such as Comcast’s).
Hubs are important for new or minority standards, but sufficient critical mass would see ZigBee/Thread incorporated into the devices themselves too. Samsung, despite its flirtations with smartphone ZigBee, has not enlarged on any plans for Thread yet, but the involvement of ARM undoubtedly promises both 6LoWPAN and ZigBee a broad ecosystem, with the associated economies of scale.
ARM demonstrated strong interest in 6LoWPAN last September, when it acquired Finnish software house Sensinode. Some of the latter’s work in security and connectivity for microcontroller IP is sure to find its way into Thread, since the company has been a major contributor to 6LoWPAN standards for some years, and has also built protocols based on COAP (Constrained Application Protocol), which enables connected devices to talk to cloud-based servers. The Finnish firm has also been a contributor to the IETF, ZigBee IP, ETSI and OMA standardization efforts. ARM has made the code for Sensinode’s NanoStack and NanoService M2M offerings available for evaluation through its mbed project, which lets engineers test products online.
Both the Sensinode purchase and the involvement in Thread sees ARM increasingly moving up the stack from its processor and microcontroller IP and tooks, into key software areas where interoperability is required, and where chip-level implementation may be helpful ‘ such as security.
And of course, like the other big founders, ARM has its commercial rivalries to pursue ‘ while Google and Samsung try to wrongfoot Apple and Microsoft, the UK company understands how a relatively small firm can harness standards and broad ecosystems to weaken a giant like Intel. The fourth big name on the initial list is Freescale, itself an IoT powerhouse in the making thanks to its major position in microcontrollers and connected cars, and its numerous M2M-focused industry partnerships, including those with both ARM and Intel, plus one with Oracle for IoT Java. Freescale expects to provide a version of Thread with many of its home automation chips and may well get to market first, once the certification program is established in the first half of 2015. The release 1.0 specs, and full terms of membership, will be published later this year. Members will pay an annual fee and get access to the specs on a royalty-free basis.
There are some definite positives to Thread, when compared to some IoT alliances. At least it is using existing technologies, some of them fully standardized, rather than proprietary inventions or vague promises of future specs. And at least, with the focus on 6LoWPAN and IPv6, this is actually about the ‘real’ internet, unlike many approaches to the so-called IoT which have very little to do with that. In addition, Thread may bring the collective weight of its founders to bear behind a useful standard which has, in the past, failed to get the same profile and investment as Bluetooth Smart, but which may have some different qualities to offer.
‘Existing wireless networking approaches were introduced long before the IoT gained ground. The Thread protocol takes existing technologies and combines the best parts of each to provide a better way to connect products in the home,’ said Vint Cerf, VP and chief internet evangelist for Google and advisor to the Thread Group.
The big fear is that, with so many groups proposing different routes to the nirvana of interoperability, the result will be fragmentation and politically motivated islands. However, there is a real question over how far a single standard is needed, and whether that would be a hopeless lowest common denominator. There is no reason why the same protocol should be optimal for the smart home (where 802.15.4 mesh looks promising) as the mobile IoT (wearables etc, hooked up by Bluetooth or perhaps low power LTE), or various industrial segments, many of which may have their own specific networks. To drive maximum benefits from the IoT, especially for firms driven by big data and cloud services, like Google, there clearly needs to be a common umbrella technology. But as long as those different standards stay in the local/personal area, the data from their gadgets can still be fed to the cloud, or exchanged with other local systems, via those umbrella networks, almost certainly WiFi with some cellular thrown in.
Google will be fighting to dominate that layer too, but with a different set of partners and a different set of protocols and tools. And the bigger question ‘ whether the IoT will thrive if it is unified by a few large vendors rather than official standards bodies ‘ will only get bigger. Oleg Loginov, chair of a new IEEE effort for IoT network standards, P2413, defended the ‘proper’ standards process, despite its problems with responsiveness and speed to market. He told EETimes that he believes groups like the IEEE must unify the efforts of the various corporate alliances, saying: “Any major technological shift, and IoT is probably the most significant one in history, starts with islands of innovation.These early activities have to transform into global standards enabling the economy of scale and vibrant ecosystems.” Many will hope that standards bodies can shake up their practices in order to enable those ecosystems, rather than letting control slip into the hands of a few corporate giants.