M2M/IoT Communication Services

The Enterprise IoT project manager will face many uncertainties, and no two Enterprise IoT solutions are likely to be the same, or even similar. However, there is one concept that is fundamental to any analysis of our connected future, and that is the need for connectivity. At some point, all Enterprise IoT project managers will need to consider how remote assets can be connected to a wider enterprise backend. In some cases, appropriate communications connections will already be in place, but, more often than not, the Enterprise IoT project manager will need to consider new options.
In this section we consider alternative approaches to providing connectivity for remote assets, so that these assets can be integrated into an Enterprise IoT solution. Fundamentally, there are two types of connectivity that can be considered: Managed connectivity and unmanaged connectivity. We devote the bulk of this section to managed connectivity, that is, connectivity provided as a service by a third party. An Enterprise IoT project manager may of course choose to deploy analogous technologies and operate their own networks, but the technology selection and management considerations are similar.

Service Overview

We have defined six key M2M/IoT Communication Services:

  • Cellular (2G, 3G, 4G)
  • Low Power Wide Area (LPWA)
  • Metropolitan Area Networks (MANs)
  • Satellite
  • Fixed Line
  • Power Line Communications

These technologies differ significantly in terms of the reach of connectivity, the ability to support mobile assets and also the levels of data throughput that can be supported. We summarize these key considerations in the figure below.

Key technical considerations of Managed Communications Services

Key technical considerations of Managed Communications Services

Of all these connectivity options, probably the most interesting are the wireless technologies. In general, wireless technologies have the potential to support a relatively homogenous connectivity environment (so that all remote assets of a particular type can be connected using the same technological solution). In turn, this allows for relatively simple monitoring and fault resolution of remote devices from a central location.
Fundamentally, wireless communications technologies are always a compromise between a relatively limited set of constraints such as data throughput rates, battery life, remote device cost, network cost, and the laws of physics. The main difference between the various wireless technologies is that they have been “optimized” in different ways, and these are the differences that we discuss in the following bullets:

  • Cellular (2G, 3G, 4G, 5G): “Mobile” is the technology that currently dominates the IoT in terms of wide area connectivity. Realistically, it is the only technology that can support a range of intrinsically mobile applications (such as vehicle platforms and eCall). This group of technologies also benefits from very wide ranging geographic coverage and a reasonable level of homogeneity across national borders (potentially allowing for international “roaming” of connected devices). Data throughput rates are reasonable (240 kbit/s for 2G, 42Mbit/s for 3G, 326Mbit/s for 4G, and faster for 5G ), and this group of technologies is well established in the M2M/IoT space. Given their importance, we have dedicated much of the remainder of this chapter to a detailed discussion of Cellular M2M Communications.
  • Low Power Wide Area (LPWA): In many ways, LPWA networks are very similar to mobile networks, but with significantly compromised data throughput capabilities which are traded in return for significantly lower price points and potentially much longer battery life. The raison d’être for these kinds of technology is the fact that the vast majority of assets that will be connected to the IoT will, in fact, generate very little data and not very frequently. Battery lives can extend to 10 years and beyond, powered by a single AA cell. Such a performance envelope opens the door to many possibilities, such as the deployment of perimeter security sensors for industrial locations that can be distributed as a kind of smart dust, continually monitoring events for up to 10 years. LPWA is currently a nascent technology, but widespread deployment of such technologies can be expected in coming years, and also LPWA-type capabilities are likely to be incorporated into 3GPP (cellular) standards within the next couple of years.
  • Metropolitan Area Networks (MANs): This category includes a range of wireless technologies that can be deployed within an urban environment. Typically deployments are homogenous and relatively ubiquitous within a single urban conurbation, but generally vary between different urban conurbations. This makes MAN-type solutions particularly suitable for smart cities solutions (such as controlled street lighting, traffic signage, and refuse collection), but less suitable for solutions that extend beyond the limits of a single (or limited number of) urban conurbation(s). Wireless MAN technologies can vary considerably in terms of technical capabilities, ranging from technologies that are equivalent to LPWA through to technologies that are equivalent to Wi-Fi.
  • Satellite: Satellite is the most flexible of all wireless connectivity solutions, but at a price. Although it is possible to support multiple MB streams over a satellite link to a moving “connected” asset, the cost is likely to be prohibitive for the vast majority of potential IoT applications. At the other end of the scale, there are satellite solutions currently under development that will potentially match current 2G (cellular) modem capabilities and price points. As a general rule, satellite tends to be the “best option” for connecting IoT devices when it is the “only option.” Use cases include the monitoring of refrigerated shipping containers (“reefers”) on the high seas, although even then it is likely that that container ship has deployed some kind of local onboard connectivity (for instance 2G cellular), so that the cost of satellite communications can be shared between multiple containers.

As mentioned earlier, fixed connectivity options are in general significantly less interesting. We discuss Fixed Line and Power Line solutions in the following paragraphs.

  • Fixed Line: If available, fixed line solutions can be very good options for connecting a remote asset. However, if fixed line connections are not available, the cost of deploying such connections specifically to connect an IoT asset is likely to be prohibitive. In cases where assets are connected using fixed line infrastructure, the actual final connection to a remote device is likely to be an Ethernet connection, and so potentially extremely high connectivity speeds can be supported.
  • Power Line: This is a niche communications solution, generally only suitable for electricity smart metering. The technology works by multiplexing a “signal” onto a “carrier” which is, in fact, an electrical power cable. This works well for electricity metering, since the electricity utilities deploying the solution generally have access to both “ends” of the power supply to a building (i.e. the smart meter and the corresponding electricity distribution sub-station), but few other potential users of power line communications enjoy such access.

Focusing once more on the key cellular (2G, 3G, 4G) technologies, it is worth highlighting a fundamental market structure consideration. Established mobile communications markets are highly evolved, and include players with many different market positionings. Two key such positionings are that of the Mobile Network Operator (MNO) and the Mobile Virtual Network Operator (MVNO). From a (human) customer perspective, these entities are pretty similar, in that they can both provide mobile communications services, including voice and data. In traditional telecoms markets, there really is very little to differentiate between the products and services offered by an MNO and an MVNO, other than brand considerations (and perhaps a niche market positioning for MVNOs). From a technical perspective, however, an MNO and an MVNO are quite different, since an MNO owns and operates its own radio access network (RAN), whereas an MVNO “piggy-backs” on a RAN that is owned by an MNO. In both cases, it’s the MNO that actually provides the radio access network and carries the calls and data, although in the case of an MVNO, the entity that actually faces the customer is something other than an MNO. Essentially, an MVNO focusses on marketing, customer acquisition, and billing and wholesales MNO-provided connectivity. There are different “flavors” of MVNO too, ranging from “light” (these MVNOs do little other than marketing, customer acquisition, and billing) to “heavy” (these MVNOs also build and own some core network infrastructure elements).

This market structure has been carried forward into the M2M world, with both MNOs and MVNOs offering to connect machines. The key distinction in the M2M space, however, is that an M2M MVNO can establish wholesale relationships with more than one MNO in each territory in which it operates. There are two main consequences of this potentially one-to-many relationship. The first is that MVNOs can to some extent select different MNO carriers for each connection that they manage, potentially selecting the MNO that offers the best signal strength in a specific location. The second  consequence is that MVNO-provided M2M connections can potentially be a little more footloose in terms of migrating to a MNO network that offers better rates. In fact, leading-edge MVNOs can now offer a level of flexibility in terms of association of a single connection with different partner MNO networks that the overall solution could almost be characterized as a virtual-shared RAN. The price to be paid for that flexibility, however, lies in the cost, and in the flexibility and homogeneity of interfaces to partner MNOs. For some M2M applications, this will be a trade-off worth making. For others, it won’t.

Mobile M2M Communication

Having discussed different M2M/IoT Communication services, we will now take a more detailed look at Mobile M2M Communication, since this is one of the most widely used and complex technologies for remote connectivity today, and will continue to be so for the foreseeable future.
We start by introducing some basic concepts of mobile M2M Communication. We will then discuss common challenges, followed by an example for an M2M Communication platform that can help address many of these challenges.
For the remainder of this section, we will use the term M2M to refer to mobile M2M Communication. See also the discussion on M2M vs. IoT in the Overture.

Basic Elements

For mobile M2M, some of the most important elements are the M2M SIM cards, the communication modules that use these SIM cards, and the M2M devices that use the communication modules:

  • M2M SIM Card: The Subscriber Identity Module, or SIM card, identifies the subscription through which a mobile device can attach to a network and access services. SIM cards are available in a variety of physical form factors including the standard plastic UICC mini (2FF), micro (3FF), and now nano (4FF) SIMs familiar to smartphone users and the more specialized SON-8-chip form factor, which is an electronic component that can be soldered into circuit boards. The choice of SIM type for M2M use depends on the specific application. Key considerations include the lifetime needed and the environment in which the SIM must operate. The service life of many M2M applications exceeds the rated lifespan of standard SIM cards. Ruggedized SIMs intended for the M2M market support more read-write cycles and longer data retention periods in order to extend their useful lives. In addition, the SON-8 form factor typically offers greater resilience against temperature and vibration.
  • M2M Communication Module: The Communication Module is essentially a mobile phone. In the early days of M2M, actual phones were used. Today, it is a small circuit board containing all the pieces needed to communicate with a mobile network using the subscription provided by the SIM. In effect, it is a specialist modem, designed to be easy to integrate into an M2M device and provide it with a complete communication service. Communications modules must be compatible with the types of mobile network that the M2M service will use, taking into account regional standards, radio frequencies, and the generations of mobile technology to be supported. This is a critical decision affecting the cost of the module, network coverage, and the longevity of the service.
  • M2M Device: The M2M Device processes data from the real world and exchanges it with the backend application via the services provided by the Communication Module. M2M devices are generally specific to individual types of application, such as the smart meters used in utilities or the telematics units in the automotive industry. In some cases, the M2M device contains the sensors and actuators that carry out measurement and control while in others they communicate with them by means of a local connection and specialist communications protocol such as the Controller Area Network (CAN) developed for automotive.

The M2M Communication network enables communication between M2M devices and the backend. Important elements include the Radio Access Network (RAN), Mobile Core Network, Access Point Name (APN), and Backhaul Connectivity:

  • Radio Access Network: The Radio Access Network consists of the base stations owned by the local mobile network to which the M2M device attaches in order to get service. Its purpose is to provide the radio connection to the device and pass traffic back to the Mobile Core network belonging to the M2M service’s communication service provider. Several generations of Radio Access Network such as GeRAN (2G), UTRAN (3G), and E-UTRA (4G) are in common use.
  • Mobile Core Network: The Core Network is in the “home” network of the M2M Communication Service Provider (CSP) and acts as a hub between the customer’s backend systems and the local mobile radio access networks to which their devices attach. The core network consolidates traffic and generates usage records allowing the CSP to carry out billing. It is also the location at which the mobile subscriptions are registered. It allows individual devices access to specific services such as voice or data and also controls the radio networks that they can attach to. When a device attempts to attach to a Radio Access Network, the RAN contacts the device’s Home Mobile Core Network in order to verify the identity of the device and check the services to which it should be granted access.
  • Access Point Name (APN): The APN can be thought of as a virtual data network that the M2M device joins when it sets up a mobile data connection. Important aspects of APNs include IP addressing schemes, authentication mechanisms, and whether the APN is public or private. The APN is hosted by the Mobile Core Nework and may be associated with a particular backhaul connection. Shared APNs, such as Internet APNs, are used by devices belonging to many different M2M businesses, and typically offer quick access and minimal setup in return for limited functionality and security. Private APNs are purpose-built for individual services and offer additional functionality such as private IP addressing, greater security, and separation from other traffic.
  • Backhaul: Backhaul connectivity moves traffic between the Mobile Core Network and the backend system. While not technically part of the mobile core network, such connectivity is usually provided by the CSP as one element of a complete end-to-end solution. A variety of types of backhaul are in common use ranging from private leased lines through to secure tunnels over the Internet.
    If a SIM is roaming the RAN and core networks are provided by different mobile network operators, the RAN is part of the “serving network” while the Core is part of the “home network.”

Challenges

“M2M Communication has very different characteristics to those of traditional mobile communication. M2M businesses and Communications Service Providers alike must deal with issues such as new commercial models, global service management, SIM logistics, system operation on a very large scale, and management of traffic different patterns. M2M is distinct from both IT and traditional mobile communication and requires its own specialist methodology and solutions.” – Mike Prince, Principal Product Manager for M2M Platforms, Vodafone.

In the early days of M2M, many MNOs assumed that M2M connections were simply data connections, and that little needed to be done to serve this new market other than issue what were then regarded as “standard” data SIMs and contracts. As Mike’s quote highlights, the reality of the M2M market proved to be somewhat more complex. Some of the more immediate and significant differences between M2M connections and human-centric voice and data connections are listed below:

  • Commercials: With long periods of device inactivity, unique traffic patterns, and very narrow margins in many M2M businesses, it is clear that standard mobile tariffs are unsuitable for M2M. International services, where the distribution of devices across countries is variable, introduce particular uncertainty over roaming costs.
  • SIM Logistics: SIMs are commonly inserted into M2M devices during production of the device itself, often using a chip-based form factor rather than the more familiar plastic component. After initial testing, the SIM lies dormant for an extended time as the device travels through its supply chain before finally being activated. Certain types of M2M device may later be recycled between users, leading to further periods of inactivity. This creates difficulties for mobile operators who must carefully manage and optimize their return on the use of scarce resources. M2M businesses attempting to use standard SIMs in this way face significant complexity and cost in order to maintain their subscriptions.
  • Scale: Some M2M businesses deploy devices in volumes that are otherwise unheard of, with single services running to millions of devices. Large-scale M2M requires a high degree of automation.
  • Operations: Unlike a smartphone, whose owner may bring it back to the shop or call a service desk if something isn’t working, machines must operate independently. Installation and activation must be highly automated and any subsequent troubleshooting carried out remotely in order to minimize costs due to site visits.
  • Traffic Patterns: Machines work in different locations and at different times of day to human beings. They may be static or highly mobile. Each machine “conversation” may involve smaller amounts of data but when things go wrong they can keep trying. As a result, the M2M traffic on a network follows a very different pattern to that generated by other types of mobile device. Close supervision is necessary to make sure that networks are not hit with excessive traffic volumes and customers are not faced with expensive bills should devices malfunction.
  • Device-Originated Communication: Connections for mobile data communication are originated by the device rather than from the network. This serves consumer mobile needs but in M2M it means that a backend system is reliant on devices to initiate connections before it can exchange data with them.
  • Security: Enterprises use M2M to support their critical business processes. M2M applications frequently deal with sensitive data or control important infrastructure. It is essential that M2M communication has an appropriate level of security, bearing in mind that the level of threat is likely to rise over the lifetime of a service as awareness of the role played by M2M increases and attackers become more sophisticated.

Beyond these items, the potentially global (or multi-country) nature of M2M solutions can drive significant complexity in tariffing, service management, and “customer” support. As Mike highlights, “Whereas traditional mobile services are strongly associated with an individual home country, M2M services can be required to operate across regions or even globally. At the time that a SIM is installed in a device its destination is often not known.”

SIM Lifecycle

Because the SIM plays such a central role in M2M, understanding the M2M SIM lifecycle is very helpful in addressing some of the challenges outlined above.
The lifecycle of an M2M SIM is very different to that used in traditional mobile services. In addition to the standard states relating to basic activation and de-activation, a number of other states are needed in order to provide greater granularity of control during the active stage of the SIM’s life. These states help automate control over connectivity during the process of building, testing, shipping, and using M2M devices. By varying the tariff according to SIM state, it is possible to match commercials to the customer’s situation. For example, a given amount of usage or time can be offered for testing.

SIM Lifecycle

M2M SIM lifecycle (Source: Vodafone)

 M2M Connectivity Management Platforms

In order to address the M2M challenges outlined above, carriers must deploy M2M connectivity management platforms. Such platforms are highly scalable multi-tenanted systems, typically delivered as cloud services. They enable a high degree of self service and are optimized for management by exception of large volumes of SIMs and devices.
The connectivity management platform plays a central role in reducing the risk for an enterprise in adopting M2M. The figure below provides an overview of the basic M2M elements and connectivity management platform interaction.

Overview of key M2M elements and M2M platform

Overview of key M2M elements and M2M platform

Key features of an M2M connectivity platform include:

  • Online Graphical User Interface: Allows users to administer and manage their service. Accounts with clear roles and permissions give users access to the functions they need in order to carry out their own tasks while avoiding unnecessary complexity.
  • APIs: Allow the integration of customer systems in order to support end-to-end business processes that depend on M2M Communication
  • Ordering and Provisioning: Efficient generation of high-volume SIM orders and bulk provisioning of subscriptions onto the system
  • Device Wakeup: SMS triggers that can be invoked by a backend system that needs to communicate with a device. The device’s response to a trigger should be to initiate a data connection
  • Security: Measures to restrict access to network resources and guard against fraudulent use. Deep integration with dedicated M2M networks platforms should offer in-depth security
  • Session Management: Detailed real-time control over data sessions including access controls, address assignment, and usage quotas
  • Analysis and Reporting: Broad examination of the records concerning one or more devices in order to build up a picture of service characteristics or produce specific datasets
  • Diagnostics: Detailed drill-down on the behavior of a specific device in order to identify the cause of faults. May examine live data or historic records (For example: protocol trace for voice, data, and SMS communications) or the outcomes of proactive tests
  • Notifications: Pro-active notifications of events of significance to different users in the customer or service provider organizations. For example: usage alerts
  • Business Rules: Configurable toolkit of measures, rules, and actions that can be used to automate processes and deal with exceptions.
  • Audit Trail: Historical record of events and changes in order to provide a full audit trail
  • Rating and Billing: M2M specific tariffs that can be applied to services as required and which take into account the full platform feature set
  • Revenue Assurance: Detailed usage records to allow verification of bills
  • Online Support: Ticketing, documentation, help, and user forums

The following figure provides a screenshot of the portal Dashboard provided by Vodafone to its customers for managing M2M connectivity.

M2M Dashboard

Screenshot M2M Portal (Source: Vodafone)

MCS and Ignite Asset Integration Architecture

The following figure describes the AIA for MCS.

3MCS and Ignite Asset Integration Architecture

AIA for MCS

Recommendations and Outlook

Building on the discussion so far, this section focusses on the future development trajectory for M2M/IoT Communications Services. We start by summarizing the role of (M2M) platforms to date. We then define “best practice” for (IoT) connectivity platforms and discuss new and emerging technologies in the IoT connectivity space. We will end with a set of recommendations for providers of connectivity platform services.

The Role of (M2M) Platforms to Date

“The key role for M2M and IoT platforms is to reduce the levels of friction that exist in today’s IoT markets. This friction is generated by a wide range of factors including high entry costs, lack of standardization in specific segments, limited skills and awareness, different and varying market conditions between countries and geographies, and regulatory policies. As a consequence, CM Platforms play a central role in reducing the risk for an enterprise in adopting M2M,” Mike Prince, Principal Product Manager for M2M Platforms, Vodafone.

Thus far focused almost exclusively on cellular-based technologies, connectivity support platforms have made significant contributions to the growth of many M2M and IoT solutions. This is the case particularly where different types of connection are required at different stages of the lifecycle of a cellular connected solution. For example, in the automotive industry, a significant market for cellular connections, the ability to activate and deactivate SIM connections is necessitated as a result of various production, test, and launch stages.

In addressing the M2M requirements of industry verticals, the design, development, and building of M2M solutions can be compared to complex IT solutions. The use of connectivity support platforms allows enterprises to standardize connectivity management, and extend the functionality within solutions. The current limitation is that very few connectivity support platforms have been designed to be flexible or adaptable in terms of different connection technologies. Once applications have been developed, it requires significant time, effort, and money to change devices, add connectivity technologies, or adopt and integrate new application requirements with new data models. This generally leads to multiple, specifically designed solutions with limited re-use or the integration of connectivity support platform capabilities.
With M2M applications becoming ever more advanced and complex, coupled with the emergence of the IoT, connectivity platforms will need to evolve. Where previously connectivity support platforms allowed connection to a narrow set of devices and primarily cellular connectivity options, future requirements can be characterized by increasing agility and flexibility.
We have already seen a similar trend in application development where abstraction has become a preferred approach for the emerging range of M2M/IoT application platforms. We expect similar developments in the connectivity support platform space, characterized by an ever increasing technology agnosticism. This combination of abstraction and agnosticism allows for the scale and heterogeneity (of devices and protocols) to be managed through fewer platforms. It also enables developers to focus more on application development rather than specific communications technologies or device characteristics.
But the simple idea of a “technology-agnostic” connectivity support platform belies the complex and challenging task involved in managing the characteristics of multiple connectivity options. For example, an M2M solution for container freight tracking might require some combination of satellite, cellular, and short-range connected devices. The ability to support these different connectivity technologies with a single platform solution could be a significant benefit and differentiator for enterprises and operators, system integrators, and providers of M2M/IoT Application platforms alike.
Managing different connectivity technologies is a complex task. Providers of multi-technology connectivity platforms face the challenge of working across different protocols, managing multiple billing, real-time data and reconciliation functions, and ensuring secure and resilient communications across a range of communications technologies. Each connectivity technology will behave differently when it is “working properly”, and may require different actions when an error status occurs. Accordingly, the ability to offer a well-defined and managed connectivity solution drawing in multiple connectivity technologies can be a significant competitive differentiator. For enterprises, such a solution removes much of the difficulty of integrating new connectivity options, and opens the door to new tariffing and billing options.

Defining “Best Practice” for (IoT) Connectivity Platforms

Twelve technical and commercial capabilities and features, as illustrated, define the “best practice” elements for connectivity support platforms. Delivering these capabilities and features will ultimately reduce friction in the marketplace, enabling significant growth in connected devices, and providing improved ROI, which in turn will open new markets and application development opportunities. In this space, technical and commercial capabilities and features are closely related, and are now converging to bring about a single and comprehensive connectivity proposition.
However, it is clear that the “best practice” capabilities and features listed are not the same as those typically cited for cellular connectivity support platforms. Such capabilities would typically include connection provisioning, usage monitoring, and some level of support for network fault resolution. By contrast, the best practices listed here typically address higher-level commercial application development, application management, and implementation considerations. Ultimately, the pursuit of best practice for a connectivity support platform will result in its being repositioned as a “Connectivity Platform” that provides actual multi-technology connectivity.

Connectivity Platform

“Best Practice” elements of connectivity platforms (Source: Machina Research 2014)

We discuss these “best practices” in more detail in the following tables. The first table details commercial best practices, whilst the second expands on technical best practices.

Six best practic elements

Six commercial best practices for connectivity platforms (Source: Machina Research)

Six best practice elements technical - platform

Six technical best practices for connectivity platforms (Source: Machina Research)

Outlook

The outlook for the connectivity space is dynamic and diverse. There is a wide range of developments expected, including the enhancement and evolution of existing technologies, the introduction of new technologies, and the blurring of technologies at the periphery. We will briefly discuss these trends in the market in the following paragraphs.
The highest-profile change in existing technologies is the introduction of the soft SIM. Specifically, this development represents the evolution of the physical removable SIM card towards an “embedded,” “component” or “M2M-Form Factor” (MFF) SIM that is more robust, less sensitive to heat or vibration, and more secure, not least because it cannot be removed and used in another device. These embedded SIMs should also be cheaper and ultimately help to simplify the supply chain.
Such “soft SIMs” are managed by cellular operators over the air (OTA) and are shipped with a “bootstrap” mobile operator configuration (the IMSI number) that allows for initial connectivity (via inter-carrier roaming.) Once such devices connect to a local network in the country in which they are activated, it is possible to implement a new IMSI, automatically and over the air, which “re-homes” the device onto a new (local) carrier network. Ultimately, it is the soft SIM that will allow you to buy cellular airtime for your iPad from the App Store.
Clearly, today’s technologies are also evolving. The inevitable “5G” cellular technology is in the works already, and new capabilities are being developed for 4G. However, for once, it seems that “faster is not necessarily better” for some of the latest iterations of today’s cellular technologies. Currently, one of the key thrusts for 4G and 5G is the development of cheap, low-bandwidth connectivity that is very light in terms of power consumption (potentially enabling very long battery life).
This development has been driven particularly by mobile operators in response to the perceived competitive threat from new technologies such as SIGFOX. These technologies offer nationwide, out-of-the-box connectivity as a service (such as as you might expect from a mobile operator), at minimal cost (cost of hardware device) and with up to 10 years battery life from a single AA-cell. The price paid for this extreme battery life is in bandwidth: SIGFOX can only support 144 (short) messages per device per day, but that’s more than adequate for many M2M applications. A host of other players are currently emerging in this space, often with very different go-to-market strategies. Emerging players include the likes of M2M Spectrum, NWave, and SemTech and many more. Generically, these new providers are referred to as Low Power Wide Area (LPWA) network providers. Huawei has also muscled in on the act through their acquisition of Neul, and these companies are now jointly developing a “clean slate” Cellular IoT (“CIoT”) solution aimed at deploying a LPWA service in old GSM spectrum that has been freed up by the migration of data traffic to 3G and 4G networks .
As ever, there is also a range of innovative technologies waiting in the wings, pining for recognition as accepted (de facto) standards, at least within specific niches. These include the likes of Zigbee, ZWave, and VLC (Visual Light Communications, or Li-Fi) and a perennial favorite: HALO (High Altitude, Low Orbit) platforms. And many more.
And, of course, more and more things will become connected to the IoT, leading to the widespread adoption of Home Area Networks (HANs), Vehicle Area Networks (VANs), and Personal Area Networks (PANs). And no doubt many other yet-to-be imagined area networks, which we could potentially term xANs. Ultimately though, the “things” at the very periphery of the IoT will be only intermittently connected, or even simply “sensed” as opposed to communicating in any meaningful way. This will result in a blurring of technologies at the periphery, particularly where NFC (Near Field Communications) can be substituted by a range of other technologies (such as 2D barcodes, RFID, and WiFi direct) which could be termed near-field communications technologies.

This type of market development is further illustrated by the following interview with Nigel Chadwick, CEO of Stream Technologies.

Jim Morrish: Can you describe the approach that you take to providing connectivity for IoT?

Nigel Chadwick: It has long been popularly, but wrongly perceived, that the connectivity layer pertaining to a connected ‘thing’, is a generic commodity.  The reality is that this connectivity layer is highly differentiated in terms of technical parameters such as resilience, as well as fragmented in terms of footprint and communications protocol ‘type’.

Ultimately, we think that the IoT will be best supported by what we term a ‘Unified Access Connectivity Environment’ (U-ACE).  Such a platform should be adaptive to multiple connectivity protocols –  including cellular (2G, 3G, LTE, CDMA), satellite, wi-fi, low power wide area radio networks (LoRa for example) and others. This is particularly valuable to established wireless carriers, evolving LPWA network operators and enterprises, since such a solution can provide third parties with a powerful yet simple, low cost – fast deployment solution to monitoring, managing and monetizing connected Things.

Jim Morrish: What’s the key difference between this approach and the typical approach that might feature in the market today?

Nigel Chadwick: To date, connectivity ‘platforms’ designed for the IoT, have tended to evolve and be deployed as proprietary or closed to specific network operators and service providers. They are also largely designed and built to manage a singular wireless layer (largely cellular but also some satellite).

With this in mind, there are several key differences when compared to the unified connectivity environment that I described earlier:

  • A U-ACE should be technology agnostic and be designed to work with any type of wireless protocol. In turn, this means that multiple network connectivity layers can be fully managed from a single platform.  The U-ACE therefore becomes a one-stop solution that provides a full range of connectivity types and geographic coverage and can support the extension of powerful connectivity management tools outside of the ‘on-net’ reach of existing carriers  – for example extending management capability of cellular carriers into LPWA, or satellite operator capabilities into cellular.  It also ensures that connectivity management is effectively ‘future-proofed’.
  • A U-ACE should be what we term ‘technically light and non-invasive’. Through a simple set of API’s a ‘virtual state’ of the host network connection can be created, relying on the underlying integral but external infrastructure, to provide real time and granular monitoring and management of each connection. This enables a fast and low cost deployment resulting in minimal time to market for both network operators and client organizations.
  • The use of API’s is important for another reason. A U-ACE should be able to co-exist with any existing platform used to manage connections; either above or below in the vertical chain of connectivity management. This enables extension of wireless connectivity and presentation of a single user interface for those interacting with the platform thereby avoiding the multiple platform view and engagement with all the inherent complexity and duplication, and associated risk and resource cost.

Stream Technologies has developed such a U-ACE. It’s a platform that we call IoT-X. IoT-X has fast evolved into an IoT connectivity based ecosystem, reflecting Stream’s vision of a the kind of Unified Access Connectivity Environment that I described earlier.  IoT-X is about much more than just connectivity.  There is a growing number of integrations into IoT-X including other platforms that enable device management, data exchange management and other aspects that are often critical to the successful implementation of an end-to-end IoT solution.  Examples of integrations completed so far include ARM’s Mbed, ThingWorx, and wot.io.  This means that enterprises and wireless carriers (and their end customers) can ‘plug into’ the already extensive and growing ecosystem inherent within IoT-X.

Jim Morrish: Isn’t managing all those technology choices a huge task, just in and of itself?

Nigel Chadwick: Stream has continuously evolved IoT-X in a Darwinian way over a 10 year development timeframe.  Given the scaling requirement, as well as the breadth of the platform capability, there has been a focus on how best to remove ‘friction’ in the process of integrating into third party wireless networks, and also from end user perspectives in terms of ease of use and functionality.  The company has evolved to a situation whereby it is pretty much a software development house, with a fully developed platform that automates connectivity monitoring, management, and billing.  We also have in-house specialists in each of the core wireless technologies and infrastructure we deal with – cellular, satellite, LPWA, backhaul infrastructures and coding.  We have provided connectivity to end customers since 2000 so carriers and enterprise organizations engaging with us to adopt IoT-X are increasingly relying upon the technical expertise vested in the company to help them figure out how to deploy vertical solutions, thereby moving up the value chain.

Jim Morrish: How should a U-ACE cope with intermediating between the limitations of any specific device connection and customer needs and demands?

Nigel Chadwick: A U-ACE should be agnostic and capable of accepting data via any connectivity protocol.  Off the shelf adaptors for most of the current common (and not so common) protocols should be readily available. Given the U-ACE is purely dedicated to management of the communication layer there is essentially very limited intermediation necessary.

Data is data regardless of where it comes from. At Stream Technologies, we’ve always tried to keep everything as simple as possible to reduce complexity and chances of failure. Our approach has been to make all networks comply with a defined structure rather than trying to make each individual network work together. What we’ve created is an abstract concept of how a network should function and we adapt networks in to this. One of the benefits to this is that as soon as a new network is configured in it instantly gets access to all our other services and integrations.

Jim Morrish: Who do you see as your ideal customers (and partners) within the overall IoT ecosystem?

Nigel Chadwick: We believe IoT-X and U-ACE have the potential to unlock substantial value for a range of customers and partners through enabling connectivity income on a large scale basis and/or reducing the Total Cost of Ownership (TCO).  This means that System Integrators and Solution Providers in other parts of the IoT ecosystem – those with significant numbers of  customers which require device or ‘thing’ connectivity, can significantly remove substantial amounts of ‘friction’ and inertia through removing fragmentation and uncertainty pertaining to the connectivity layer.  The greater the scale of the connectivity requirement in terms of scale, geographic spread and connectivity types deployed, the greater the management risk & cost, and the greater the relevance of IoT-X. This is equally applicable to enterprises for the same reason. For carriers, including (but not constrained to) cellular, wifi, satellite too, the ecosystem we are creating provides a ready-to-go service for their existing and future end customer need.  As a result of ecosystem evolution as well as in response to the requirements of customer using IoT-X, then supplementary services and technologies will continue to be further integrated and this further extends the flexibility and options to all adopters and users in the ecosystem.  Everyone benefits.

Jim Morrish: How do you think that the overall market would develop?

Nigel Chadwick: The Internet of Things is promising billions of connections.  Due to the sheer scale of end point connections, their inevitable monetization will lead to emergent new organisations and clusters or partnerships comprising new service type organisations as well as established players.  The dynamics may change as to where value is created, including a potential shift of value and empowerment towards LPWA solutions, Wi-Fi and cable operators, as well as System Integrator and Solution Provider type organisations.  We are still very much at the early stage of the IoT lifecycle.

There also remains the consumer sector –much anticipated and discussed, but yet to adopt in mass worldwide. Home security, personal asset management, health & wellbeing, are massive markets still to be affected by IoT.  Connectivity management for this sector will equally be required across different wireless layers; security, reliability and data routing, storage, accessibility and sharing will move up the agenda, again introducing new challenges leading to the evolvement of companies and technologies that will usurp traditional legacy companies in the comms field and will likely shift control back to the consumer.

Finally, the all important topic of Total Cost of Ownership – imperative if billions of things are to be connected to the internet. I expect new business models, possibly even some freemium type models, to start to emerge around platform connectivity management.  If TOC is to reduce then management of the connectivity layer is one obvious element whereby automation of systems and processes might result in some level of cost reduction. Conversely, currently ‘free’ connectivity options such as Wi-Fi (when used off the back of an already paid for broadband connection for example), or a public LPWA network, could also start to be simultaneously managed as a ‘private’ network – thereby introducing the possibility of monetizing. This further introduces the somewhat radical notion of the creation of new national networks comprising multiple private networks within the unlicensed wireless spectrums.

Given these ideas, it soon becomes quite clear that the ability to effortlessly manage networks, monitor connections and monetize/bill for connectivity to and from ‘things’ is an incredible opportunity.  And we really are at the very beginning of what is possible.

Recommendations

Ultimately, what the discussion in this section points to is the concept of “connectivity as a service”, and the emergence of entities that support that proposition. This is an emerging, but particularly valuable concept as old M2M markets transition to IoT markets. It is highly consistent with the general technology agnosticism and abstraction that characterizes the IoT.
With that in mind, and with the emerging requirements of enterprises seeking to benefit from M2M and the IoT, connectivity support platform providers should:

  • Continue their efforts in minimizing points of friction in M2M and IoT market growth and development by designing and building platforms that enable enterprises to create, build, and deploy agile, scalable, and flexible solutions for managing devices and connectivity, and developing and managing applications and data
  • Create open and integrated systems that encourage and enable enterprises to deploy and manage end-to-end M2M and IoT solutions
  • Remain aware of the evolving requirements of enterprises, and continue to explore how platforms can enable the strategic expansion of enterprises into new markets and opportunities

Finally, it is worth highlighting that the rise of connectivity support platforms does not relegate connectivity provision to commodity status. In reality, with the twelve capabilities and features outlined above in the “best practice” model, this new model of connectivity platform will be an important addition to the market. While connectivity may become a commodity, the provision of technology-agnostic, seamless connectivity as a service will become a highly valuable proposition.

We would like to thank Mike Prince, Principal Product Manager for M2M Platforms, Vodafone, for his support with this section.