Your browser is not supported. Please update it.

13 November 2015

NB-IoT get its own Forum, but Nokia wants to unify far more than the RAN

In public at least, harmony reigns between the two rival camps which were looking to define an LTE-based standard for internet of things (IoT) networks. Despite continuing insider talk of technical differences and behind-the-scenes politics, officially all the major wireless network vendors, and a group of major operators, have come together to form the NB-IoT Forum. This will seek to accelerate the development and adoption of that platform, and in doing so, hope to starve the challengers in the low power wide area (LPWA) space of oxygen.

This was a hard-won step towards unified platforms, and if it took so much diplomatic effort to achieve consensus on one strand of LTE Release 12/13, doing the same for the entire IoT, from short range connections all the way to analytics engines, would seem an impossible task. Yet this is what Nokia says it is “kickstarting”, claiming a standardized umbrella framework is needed to allow interoperability between every element of the IoT. The Finnish company has appointed itself as the scourge of IoT siloes and is calling for a cross-industry collaborative effort to establish standard interfaces.

In this grand vision, NB-IoT will be just one cog in the machine, but it is still an important focus for the mobile industry right now. At the 3GPP plenary meeting in September, a compromise was officially reached between Huawei’s Cellular IoT and Ericsson’s NB-LTE proposals, though it seems that the bulk of the ensuing specifications will actually come from the latter, which has a far greater degree of backwards compatibility with current, full power LTE networks than Huawei’s ‘clean slate’ approach.

Now the 3GPP working group is concentrating on ironing out the technical details (some of them thorny, say insiders) so that a draft set of specifications can be presented by year end and included in the next LTE Release. Vodafone is taking a leading role and is one of the most vocal operators in supportin NB-LTE (which it insists was “formerly known as Cellular IoT, perhaps because it was an early ally of Huawei’s).

With fellow MNOs China Mobile, Etisalat, LG Uplus, China Unicom and Telecom Italia, it will support the establishment of five NB-IoT labs worldwide, to carry out development, interoperability testing, certification and other activities designed to accelerate uptake of the new standard and formation of an ecosystem. These operators are all founders of the new NB-IoT Forum, which also contains Ericsson, the GSMA, GTI, Intel, Huawei, Nokia, Qualcomm and Telefonica.

The NB-IoT Forum will be hosted within an “existing industry level organization,” the founders said, without revealing which one. The group has a range of goals, all feeding into the overall objective of accelerating availability and uptake of the future standard. The Forum will also promote interoperability and push the technology into vertical markets.

It is certainly in the interests of these operators to speed up the process, and ensure that targets of commercial deployments in 2017 are met, or they may see other networks like Sigfox and LoRa gaining entrenched positions for certain key applications, especially where they are being adopted by MNOs and potentially moved from unlicensed into licensed bands.

Vodafone and its allies are steering away from such solutions and argue that a unified platform with full compatibility with existing networks is essential to drive strong economics and interoperability. While the NB-IoT air interface will reuse existing RAN and transmission protocols, it will have its own dedicated core network, heavily virtualized. Vodafone says the NB-IoT access layer will connect seamlessly to its existing M2M network, and that such solutions do away with the need for private networks or for specialized IoT systems.

Meanwhile, Nokia seems somewhat conflicted in its approach to the IoT. For the access network, it is as adamant as Vodafone that a unified air interface for LPWA connections is essential as is compatibility with existing networks. Stefan Kindt, head of technology marketing, asked in a recent blog post, “Is Sigfox/LoRa the new WiMAX?”, arguing that WiMAX lost to HSPA and LTE not because it was an inferior radio standard but because it did not interwork with the 3G networks in place round the world at the time.

Yet with its new initiatives to create a broad framework, it is talking about standard interfaces which will allow all kinds of connectivity, devices, apps and security mechanisms to work together seamlessly – something which could apply to LTE and its upstart rivals too.
The ‘system of systems’ approach, with a set of common standards and open interfaces, is often discussed in relation to 5G, and if Nokia’s dreams become reality, they are most likely to be part of that phase of IoT development rather than anything nearer term, because there are so many uncertainties about how IoT technologies and markets will look. However, in Nokia’s view, that is precisely why this is the best time to put frameworks together, as long as they are open and flexible enough to support any possible outcome in terms of use cases and standards.

“While the 3GPP is standardizing cellular-based wide area IoT connectivity and several technologies exist for standardized short range communications, standards for the raft of other technologies needed for IoT are missing,” Nokia said in its official call to arms. “Areas such as applications enablement, analytics, security, location mapping, indoor positioning and smart sensors are all highly proprietary, which does not support interoperability.”

One project which may be of interest to Nokia in its quest comes from Evrythng, which has submitted a ‘Web of Things’ proposal to the standards body, the World Wide Web Consortium (W3C). This aims to create a web-based standard to improve IoT interoperability with a global open application layer.

Following three years of collaboration between the W3C, IBM, and the Barcelona Supercomputing Center, under the auspices of the European Commission’s COMPOSE research program, Evrythng was tasked with turning the group’s findings into a usable platform – which is now the Web Thing Model that has been submitted for consideration.

The COMPOSE project was begun in November 2012, and was intended to enable new services that could integrate real and virtual worlds by converging the IoT with the ‘Internet of Services’, with the goal of creating an “open and scalable marketplace infrastructure, in which smart objects are associated to services that can be combined, managed, and integrated in a standardized way to easily and quickly build innovative applications.” The result of this ambition is the Web Thing Model.

Evrythng’s CTO, Dominique Guinard, and fellow co-founder and EVP of R&D, Vlad Trifa, headed up the company’s work on the proposal within the W3C interest group. The goal is to create another open web standard, like HTTP and WebSockets, in order to bring physical objects into the cloud applications that will power the IoT. Essentially, this is an API for linking physical things and the cloud.

Key to the project is network and transport protocol independence, meaning that the standard would be PHY-agnostic, and therefore support all the usual suspects in the IoT – from cellular, WiFi and Bluetooth, to ZigBee, Z-Wave, 6LoWPAN and Thread, and onto LPWANs like LoRa and Sigfox. However, such projects do highlight, as Nokia does too, how wireless standards efforts will not live in isolation in future, but will need to take account of their counterparts in the IP and web worlds like the W3C and IETF.

The second main thrust of the new proposal is the use of web protocols, including HTTP, WebSockets, and JSON, in order to achieve that transport layer agnosticism. Citing these web protocols as the critical building blocks, Evrythng advocates that the third component of the proposed standard is a clearly defined open framework that models physical objects and their interactions – so that a common global language emerges that allows the different players to cooperate more easily in their designs and operations.

Guinard said that the W3C would likely standardize the specs in early 2016, and that products would then begin appearing around 12-months later. He likened the approach to an open version of Apple’s HomeKit or Nest’s Weave – application frameworks that support multiple physical layers, and ensure interoperability by defining the application layer interactions. He added that the W3C has no interest in defining another PHY.

Guinard explained: “The architecture of the IoT is currently a complex labyrinth of fragmented standards and alliances. This makes neutral institutions such as the W3C vital to ensure that no single, commercial interest dominates the space. Universal standards are essential; not only to provide a fair marketplace, but also to help the IoT achieve its full potential. By operating in silos, devices are making it much harder to work together to deliver a coherent, valuable user experience. This is why Evrything has proposed an entirely web standards-based approach.”

Trifa added: “The adoption of a simple and uniform taxonomy that provides a set way of describing Things would enable all applications to discover and communicate with a myriad of things without the need for individual custom codes. Standardization would mean an IoT without barriers where anything is possible.”

Ingenu has big plans for US LPWA:

Back in the LPWA debate, the specialist network providers are not losing heart. Last week we reported on Sigfox’s expansion in the US and LoRa’s big win in India. This week, a smaller player, Ingenu (formerly OnRamp) announced that its Machine Network, running in the 2.4 GHz band, will be deployed in 30 US cities by the end of next year, by which time its networks will cover about 100,000 square miles.

Like LoRa and others before it, Ingenu has shifted from concentrating on private networks only – a space where other providers like Telensa still sit – to public systems, which in the absence of a partner with licensed frequencies, means living in a licence-exempt band. Its first public IoT service will be in Phoenix, Arizona, followed by Dallas/Fort Worth, both of them metro areas where Ingenu could make some of its existing private networks available to public IoT deployments (it is already doing this in Italy). However, eventually it wants to make most of its revenues from licensing and services.

The firm’s aim is to deploy about 600 towers to cover 70% of the US, which it claims would be highly economical, calculating that LoRa would need 11,000 towers for the same coverage and Sigfox 43,000 (even though they run in the lower frequency 900 MHz band).