75B IoT Devices by 2020: Do Predictions Matter?
The 75B prediction comes from Morgan Stanley: Cisco pegs it at 50B; Ericsson’s used to forecast the same figure, but in June they brought it down to 26B. A reduction of just fewer than 50 percent is a clear indication that these figures are guesstimates — wild guesstimates. What would be interesting, and useful, is to know the percentage split between B2C and B2B. Common sense tells us that the B2B figure has to be much lower than B2C: Matt Hatton of Machina Research puts the split at 23 percent / 77 percent.
The 77 percent sector includes wearable devices, impulse purchases in many cases. The 23 percent sector comprises multi-device IoT solutions, significant considered investments that will determine the future of the organization. Lumping them and hyping up the figure can direct attention away from the rock-solid business case for the IoT. In addition, the Net is disseminating a lot of ill-informed consumer-centric nonsense and that doesn’t help the cause.
Where are we now?
In September 2014 the IoT topped Gartner’s Hype Cycle Special Report, which implies that the descent into the ‘trough of disillusionment” has started. This may or may not be true for the consumer sector, but there is ample evidence that at we are at the beginning of a rapid expansion of the Industrial Internet of Things (IIoT). For example, the McKinsey Global Institute predicts that the IIoT will drive productivity growth of 2.5 to 5 percent over the next ten years, which translates into cost savings of $900 billion a year just for the manufacturing sector. That is a meaningful prediction.
We’re also witnessing a slew of enabling technology developments. Processing costs have declined dramatically in recent years and so-called multi-core chips allow intelligence and advanced computing power to be embedded in devices. Apps, software and operating systems can function autonomously and/or connect to other machines in a synchronized manner. GE (General Electric) calls them “Brilliant Machines”.
The business case for the IIoT is compelling; quite soon it will be overwhelming because it leverages the technologies that are taking us into a new 21st century era, one that’s based on the transition from a product to a services economy, on distributed intelligence and intelligent devices, as well as new business models, paradigms and concepts.
Where are we going?
Apart from onwards and upwards, which is a safe prediction, the m2m / IoT industry is increasing the emphasis on the deliverable —information on which real-time C-level decisions can be made — and away from the way data is delivered, the so-called value chain. Telit and other leading vendors have been building bottom-up solutions that go from the edge of the network, where the devices are located, through to the Cloud and enterprise environments. But the market is using top-down services, so we can expect to see this reflected in the way that vendors market their offer in future.
“Real-time C-level decisions predominate the IoT industry which is in turn based upon information about the value chain. By virtue of this deliverable, real-time intelligence is enhanced and extended.”
There isn’t going to be any significant change to the value chain — data will continue to be acquired, aggregated, transmitted and processed — but we will witness the value of real-time intelligence being extended and enhanced.
The Internet of Wisdom
Organizations such as iMinds, which is a digital research center and business incubator, aim to develop and demonstrate solutions and concepts that improve performance beyond today’s state-of-the art technologies by at least one order of magnitude. One such development is “The Wisdom of Things.”
Right now we can handle data and information in a structured and machine-readable way, but longer-term we need a deeper, semantic understanding of all that aggregated data. And when I say we, I mean users and high-level applications. When information is enhanced in this way the intelligence is actionable and can be used, for example, by decision support systems. We then move from the Internet of Things to the Internet of Wisdom.
Semantically annotated m2m resources provide more meaningful descriptions: they provide consistent data translation and data interoperability between heterogeneous applications. Semantic descriptions are, in fact, necessary to secure interoperability between applications. This is work in progress and it forms part of the mission of the European Commission’s new Alliance for Internet of Things Innovation.
Self-learning systems are another interesting development. The main idea is to get computers to “learn” rules that are difficult for a human programmer to define using standard programming methods. It leverages massive amounts of data, internal, external, sensor, social and so on to identify and define associations between data sets.
It’s worth noting that machines don’t solve real, complex business problems. That’s what people do. Machines can compute at incredible speed. People can think creatively and in context. Put them together and allow people to query the way intelligent machines work, to suggest changes, and the intelligent software will evolve on its own.
Making it fit for purpose
Today’s cellular networks are not truly fit for purpose. Most m2m traffic has a payload of less than 30 bytes and only needs a throughput of 100 bps. In addition the communications protocol, IP, is not an efficient way to handle low-volume traffic. The communications overhead can be 500 to 600 bytes: well over an order of magnitude higher than the payload.
That issue has led to the development of lightweight protocols, such as 6LoWPAN (IPv6 over Low power WPAN), a somewhat clumsy acronym and Message Queue Telemetry Transport (MQTT), which is another light weight protocol, but it’s also a technology. IM-type messages can be used and files exchanged, which means that transportation is payload agnostic.
A related development is the trend toward the use of LTE / 4G networks, which can transmit data at rates of up to 300 Mbps. That looks like a massive overkill. However, LTE represents the future for mobile network operators and this versatile technology has the potential to deliver significant benefits for the IoT. The efficient use of spectrum reduces the cost of delivering services and LTE future-proofs IoT solutions against earlier generation 2G networks being discontinued.
Companies who recognize that LTE technology will be the global standard and who are marketing solutions that have long life times, ten years or more in key segments like automotive, cannot rely on 3G since it too might reach decommissioning status within this timeframe.
LTE MTC (Mobile Type Communications) is another upcoming development. It defines a new, low-complexity user equipment category having a relatively low, but more than adequate, 10 Mbpstransmission rate.
The last word
The last word goes to Apple co-founder Steve Wozniak, who recently warned that the Internet of Things, or IoT, market could be reaching a “bubble phase” comparable to the dot-com bust in the late 1990s. One way to gauge bubble-like growth is through irrational hype, which is happening.The other occurs when share prices are over-valued and P/E ratios start going through the roof. That hasn’t happened. It would be good if the hype bubble were to burst: good for the B2B sector and good for the economy.