It’s fair to say that IoT has seen a fair few buzzwords roll through over the years, but there is a fresh batch of cutting-edge technology that will really change the way businesses and consumers alike interact. We’ve picked out eight critical areas where this change will be most marked and disruptive to the IoT networks we know today:
LPWAN is one of the darker horses in the IoT stable; however, there is about to be much more noise about it. As one analyst firm put it recently, because of the unique characteristics of LPWA technologies, of the 50 billion devices estimated to be connected to the IoT by the end of 2021, it is expected that more than 60 percent of these devices will be connected with LPWANs. Therefore, the global low power wide area network market is expected to rise with a CAGR of 70% during the forecast period of 2018-2023.
Indeed, given the likely contention and network noise created by that 50 billion devices, the desirability if migrating them to LPWAN and off existing networks is a necessity. If that wasn’t enough, LPWAN also offers lower costs, better security and longer lifespans for in-field sensors, due to the low power requirements. T-Mobile already has networks live in the US, and it is certain other infrastructure providers are racing to rollout too, albeit hampered by the current MLCC shortage – watch this space.
Another imminent arrival, AX (or 802.11ax) is the successor to the current 802.11ac Wi-Fi standard, and should be formally ratified in late 2019. However, both consumer and enterprise WLAN markets are already beginning to put out pre-standard products in advance.
AX offers much improved data speeds, as well as better performance over the existing standard. Top speed promises from vendors are up to an impressive 10 Gbps, from a theoretical total of 14 Gbps – which when you consider that the 1999/2000-era 802.11b hit 11Mbps in ideal conditions you can see how far we’ve come. Early iterations are focussed around ultra-dense deployment scenarios, which not only includes public use deployments like airports or railway stations, but also IoT-intensive applications.
Printed polymer sensors
If there is one challenge that unites IoT implementations, it is sensor design and positioning to ensure reliable data is produced. From pressure sensors monitoring oil and gas compliance, through to thermostats in the home, difference environments and different mediums call for varied solutions. However, new research work on substrates for printed polymers holds out the hope of being able to print a wide range of sensors onto a range of mediums. One example being QTSS, based on electron-tunnelling between shaped particles of ‘magnetite’ buried in a range of elastomers. One Yorkshire-based startup, Quantum Technology Supersensors, has created materials that cover 16 orders of resistance and can take almost any size, from covering entire floors down to tiny touch pads.
WPA3 is the latest security standard for Wi-Fi networks, and is part of a raft of improvements to the ageing WPA2 standard, which most new Wi-Fi devices should support. WPA3 aims to simplify wireless security in a variety of ways, as well as improving the strength of the encryption. Of particular note for IoT will be the Wi-Fi Easy Connect feature, which will allow new devices to be added to wireless networks by simply scanning a quick response (QR) code with a mobile phone.
WPA3 will begin rollout late 2018, and is expected to hit mass adoption in late 2019, although precise timings will vary between manufacturers.
There’s been plenty of buzz around wonder-material graphene, but gradually useable devices and products are emerging. One team at The University of Manchester has designed graphene sensors embedded into RFIDs, built up by layering graphene-oxide over graphene to create a flexible heterostructure. The resulting humidity sensor can connect to wireless networks, and requires no batteries as it harvests power from the receiver. The researchers predict various monitoring applications in moisture-sensitive manufacturing processes such as food safety, healthcare and nuclear.
Multi-functional polymer composites
Although the name is a bit of a mouthful, the technology is vitally important. While current IoT applications might include sensors embedded in substances such as reinforced concrete, that can alert when the concrete is dry, and later on if structural damage is detected, the new wave of composites will have this type of behaviour built-or baked in.
For example, structural members in aircraft will not only be lighter, stiffer composites, but these materials will increasingly be inherently able to detect delamination, cracks or other signs of mechanical fatigue. An early indication of this is the embedded de-icing technology developed by GKN on the Boeing 787, and it is estimated that the market for synthetic fibers used in composites is forecast to exceed $9bn by 2027.
Of course, the major speed upgrade coming for mobile networks is 5G, broadly set for rollout in 2019, although that varies by country and manufacturer. Although consensus on the technologies is not entirely complete, predicted data speeds could be between 10 to 20 times faster in the real world.
Rollout is planned to be in parallel with existing 4G networks, and potential applications for the hugely improved bandwidth include much improved control of drones (making them viable for search and rescue, for example), and enabling inter-autonomous vehicle communication, exchanging real-time traffic and mapping data, for example.
The vast data flows that this new array of sensors, running on the faster next generation networks, could create will be a challenge to manage by non-automated means, which is where many hope that AI and machine learning tools will take much of the strain. Using the analytics from this biggest of big data pictures will undoubtedly be a major factor in IoT networks of the future, identifying new patterns from the sensor data, and then making intelligent adjustments or flagging the results to a human operator. The future of IoT is firmly bound up in AI and machine learning – that much is already clear.