The Machine to Machine (M2M) / Internet of Things (IoT) sector is currently a Babel of proprietary systems and competing standards. However, several important consortia are working on standards and shared platforms to help combat fragmentation and the preservation of unwarranted data siloes.
Here, Anatoli Levine of Spirent encourages developers to understand the emerging standards landscape and make informed decisions about which standards to adopt.
Metcalfe’s law famously asserts that the value of a network is proportional to the square of the population it connects. The more modest estimate is that the value is proportional to n.logn, where n is the number of that population. In either case, the 2020 suggested figure of around 26 billion connected devices would be astronomically valuable – if were we to assume that everything was connected to everything else.
Of course, we are a very long way from realising this sci-fi vision of a totally connected IoT world. For a start there are systems that will remain locked in their own networks for reasons of security, privacy or practicality. Then there are legacy, administrative or “political” reasons why certain IoTs are spreading independently of others. But there are also many IoTs that would benefit from greater integration, but are trapped in data islands because of incompatible standards or connectivity issues.
Despite all these hurdles to interconnection, there is still a massive potential for the Internet of Things. Technological market drivers include: increasingly ubiquitous IP networks, a proliferation of short-range device connectivity solutions and the emergence of new technologies and protocols for linking low-power, resource-constrained devices.
Market forecasters confirm that this is a fast-growing area, though just how large the market is open to debate. While Gartner has forecast global revenues from IoT products and services of USD263 billion in 2020, IDC estimates that spending on IoT devices, connectivity and services could reach some USD1.7 trillion in 2020.
This range of possible values is partly because we are entering new territory: for every ingenious new IoT application being suggested, there must be thousands of opportunities yet to be discovered. Against that there are the hurdles already mentioned – including the range of competing standards and standards developing organisations (SDOs) currently active within the M2M/IoT sector.
Simpler, better communication between devices, networks and applications will facilitate those innovative applications waiting to be discovered, massively increasing the value of M2M networks. It will also accelerate adoption by encouraging users to deploy M2M hardware without fear of being locked into a vendor-proprietary cul de sac.
The rise of simpler M2M devices
We already have an IoT connecting relatively smart devices such as computers and smartphones. The more radical deployments are those that add a massive and scattered population of sensors, controllers or other low power devices to the IoT.
Battery-powered M2M devices such as sensors and actuators are typically designed to be small, cheap and maintenance-free. This implies limited resources, such as minimal memory and restricted processing power to allow battery life to extend for months or years. Such devices have very different connectivity needs from smartphones, tablets or personal computers: they typically need to transmit low volumes of data intermittently.
So a number of specific protocols have been developed to meet these M2M requirements – in addition to existing, well-established communications protocols such as 802.3 (Ethernet), 802.11 (WiFi), GSM/GPRS, HTTP and TCP/IP.
At the network/transport layer these include:
- CoAP – web transfer protocol for constrained devices/networks
- MQTT – light weight messaging transport protocol running over TCP/IP
- 6LoWPAN – IPv6 over low power wireless personal area networks
At the physical/link layer these include:
- 802.15.4 – low-rate personal area network
- Bluetooth LE – low-power wireless personal area network
- LoRaWAN – low-power wide area network
- SIGFOX – low-power wide area network
There are also device management protocols, such as OMA Lightweight M2M (LWM2M), specifically developed for M2M hardware to supplement existing device management protocols such as TR-069.
How to avoid being trapped in data siloes
M2M systems were first developed for specific use cases and often for particular vertical sectors – for example, a network of sensors and actuators to monitor and control the heating, ventilation ad air conditioning in an office building. Data from these devices feed into an integrated monitoring and control application designed explicitly for that purpose – but would be inaccessible to other, third-party applications without expensive work on systems integration.
This vertically integrated, siloed approach limits the network’s potential value in terms of Metcalfe’s law. It sets limits on innovative applications that might integrate and analyse data different M2M systems. Proprietary solutions also limit third party systems that could add more advanced processing to the existing data network.
So there is a current movement from this “stove-pipe” development towards a horizontal model, in which applications can interact with devices using standard APIs and communications protocols. The aim is to encourage interoperability both across hardware and software from different vendors and also across different vertical sectors and use cases.
One way to achieve this is by adding a “service layer” between the network and application layers. The diagram shows on the left a single application applied to a networked device, and on the right a number of applications simultaneously applied to a range of networked devices via a common service layer.
This service layer defines common service functions such as device discovery and registration and provides an abstracted representation of underlying resources so that applications can communicate across vertical domains, through well-defined APIs.
This shift is opening up the M2M sector to new kinds of innovation. These range from improved city management – using big data analytics across smart parking, traffic management and smart lighting systems – to new business models such as leasing expensive equipment according to actual usage using existing remote monitoring and diagnostic systems. Above all, it will make possible those thousands of new opportunities yet to be envisaged.
The author of this blog is Anatoli Levine, director of Product Management for Developer Tools Business Unit (DTBU) at Spirent Communications.
About the author:
The author is Anatoli Levine, director of Product Management for Developer Tools Business Unit (DTBU) at Spirent Communications. He is responsible for developing strategy and product roadmap for the new M2M and IoT segment for enabling products for developers. Anatoli is also responsible for the portfolio of enabling technology products (WebRTC, VoLTE/RCS, SIP, IMS and many others) for developers of Unified Communications solutions.
Since 2006, Anatoli has been the president of International Multimedia Telecommunications Consortium (IMTC) – an organisation facilitating interoperable implementations of multimedia communication technologies based on the open standards.
Comment on this article below or via Twitter: @IoTNow_ OR @jcIoTnow