Will a lack of standardization slow IoT development? - Embedded.com

Will a lack of standardization slow IoT development?


The idea of the internet of things (IoT) has moved rapidly from being marketing hype to market reality, as the concept has been implemented in everything from fitness trackers to remote weather monitoring stations. As the concept has matured, there have been efforts to promote over-arching standards for IoT devices, in a belief that strong standardization will ease the rollout of the technology and strengthen the network effects at its heart.

Some proponents of 5G, the latest mobile communications standard, suggest it should become the preferred communications channel for IoT devices. And there are good reasons for using 5G to implement some IoT use cases. 5G offers much higher data rates, lower latencies, and support for many more devices per unit area than previous cellular standards, making it easier to implement widely distributed ecosystems of sophisticated IoT devices.

Standards don’t always provide equal benefits for all

In response to those who argue that 5G’s functionality is, in fact, overkill for less sophisticated IoT applications, the 3GPP standards body has incorporated NB-IoT and LTE-M technologies, two lower-cost, lower-functionality approaches to providing wireless connectivity for IoT devices, into 5G. The latest mobile communications standard is therefore equipped to become the Swiss Army knife of wireless communications, providing appropriate connectivity options for a very wide variety of applications – including the IoT.

The idea of a standard can be seductively tidy, a kind of oasis of Marie Kondo organizational certainty in an otherwise chaotic development process. If we just stick to a standard approach rather than brewing our own solution, then we’ll fit in with what everyone else is doing, benefit from the resultant economies of scale, be able to focus our resources on developing our true differentiators rather than enabling technologies ­– and so on. But not all standardization efforts are benign and adopting the resultant standards doesn’t always provide equal benefits for all.

Some standards efforts are launched in good faith, and then are talked into the ground by powerful participants who are happy to drag out the formalization process for as long as possible in order to protect a piece of proprietary technology that happens to control access to a market. This often happens when complex systems are partitioned into functional blocks: owning the definition of the interfaces between those blocks can be a powerful way of controlling access to market opportunities.

Some standards efforts are presented as a way of ensuring fairness, but rapidly devolve into efforts to sustain a technological advantage. We see this in Formula 1 motor racing, where the intensity of wheel-to-wheel racing at 200mph is matched by the intensity of discussions over the sport’s rules. Each year the FIA, the sport’s governing body, sets out to update its rule book to level out any gross technological advantages that individual teams may have developed. The teams, in turn, lobby hard to preserve their advantages and limit their opponents’ freedom of action.

Some standards have become so complex that they no longer offer the ‘tidiness’ that attracts developers. For example, it might seem easiest to choose Bluetooth for an IoT device’s wireless connectivity, but not all Bluetooth implementations are created equal. If the device in development relies on the mesh-networking features defined in the latest version of the Bluetooth standard, it may not play well with devices that are limited to implementing an earlier version. Standards evolve, and in doing so can fragment.

Pursuing different architectures fragments engineering efforts

At an engineering-management level, it’s worth being wary of other forms of fragmentation. For example, an engineering team may have chosen Arm as its preferred processor core for IoT devices, only to be pressured into cutting costs by moving to open-source cores such as RISC-V. The engineering team’s focus will become fragmented by senior management’s demands that it should now work with two standards.

And some standards are never quite the market enablers they were expected to be. We saw this with the development of wireless USB in the late noughties. A wireless version of the ubiquitous wired USB interface was such a no-brainer that hundreds of millions of dollars were invested by both start-ups and established players to make it happen. Somehow, though, it didn’t. A combination of factors including the 2008 financial crash, the technology’s relative immaturity at the time, and questions about its utility meant that the standard did not take off and the ecosystem around wireless USB eventually ebbed away.

Questions to ask when making decisions on IoT standards

IoT straight ahead signpost

The message here is to take care when making decisions about which standards you adopt, especially in a rapidly evolving market such as the IoT. Which standards will best help you implement your current use case in the best way for your current users? Which standards are well proven technically and look like they will remain coherent for the long terms? Which have been shown in the market to open up opportunities rather than making their users beholden to other market participants? Which provide a pathway to participation in wider IoT ecosystems that may emerge, and which are only appropriate for today’s use case and will therefore limit future opportunities?

The good news for IoT developers is that there are plenty of well-proven wireless standards to consider alongside 5G. Bluetooth is widely used in body-worn electronics, such as fitness trackers, to make the hop from the device to its host mobile phone. Zigbee is well established in smart-home applications such as smart bulbs and sockets, usually to make the hop from the device to a wired hub. WiFi is very widely available, mature and yet constantly evolving, providing a bearer channel for all sorts of more complex devices such as printers. Previous generations of mobile standards than 5G can provide ubiquitous, low-cost, and low-data rate connectivity, as evidenced by the WhisperSync service that keeps Amazon’s Kindle e-book reader up to date.

Factors slowing the development of IoT

Will a lack of standardization slow the development of the IoT? It seems unlikely. Two other factors are more likely to slow the pace.

The first is an inappropriate choice of standards. This can be mitigated by considering the issues discussed above, and learning from the many examples of the choices that others have made and how they have worked out in practice.

The second factor that could slow the development of the IoT is extended device development schedules caused by poor specifications, inefficient design processes, and weak sourcing relationships.

The use cases for IoT are so diverse, the potential volumes so large, and the competition so fast-moving that it is these issues that are most likely to catch out IoT device makers and ecosystem implementers. Having good relationships with suppliers can provide the appropriate components for use in multiple end markets through well-established buying relationships, ordering interfaces, and delivery logistics will remove an entire degree of uncertainty from IoT development processes.

When it comes to the impact of standards on the rate of development of the IoT, it pays to think deeply about which kinds of standardization will provide you and your customers true benefits, and which are just for show. Knowing the difference could make all the difference.

Dr Phil Lessner - KEMET CTO

Phil Lessner, senior vice president and chief technology officer of KEMET, joined the company in 1996 as a technical associate in the tantalum technology group. He has since held several positions of responsibility in technology and product management prior to being appointed to his current position in February 2013. He received a Ph.​D.​ in chemical engineering from the University of California, Berkeley, and a Bachelor of Engineering degree in chemical engineering from Cooper Union.

On the subject of IoT standards, and the future of IoT, EE Times, our sister publication, will be exploring the promise of IoT and the impact of 5G at an event next month, IoT & 5G World. Taking place over three days from 22-24 June 2021, speakers will be covering everything from the chip to the network, addressing challenges around security, bandwidth, connectivity, and compatibility – which includes overcoming interoperability issues that are a major cause of current market fragmentation. It will also address how 5G augments IoT.

Related Contents:

Leave a Reply

This site uses Akismet to reduce spam. Learn how your comment data is processed.