The Future Of Connected Devices
by Adam Fleming-Group Chief Technical Officer and Co-Founder|Fri Nov 13 2020
Connected devices have become a popular topic in recent years and it seems that 2020 is no exception. We chatted to our Chief Technical Officer, Adam Fleming, about the connected devices market and where it’s heading.
What does ‘Connected Devices’ mean in 2020 and beyond?
The term ‘connected devices’ is going through a reformation of sorts at the moment. Historically, connected devices fall into two camps: consumer or industrial. On the consumer side, you’ve got things like wearables, and your mobile phone is the very definition of a connected device, but increasingly you’re seeing things like TVs, fridges and cars all connected to either an information service elsewhere or the internet at large.
The other side of the market is industrial. It could be something as simple as a valve in a dairy farm, where it’s effectively going to be operating unmaintained and unattended for 365 days a year – that’s becoming a connected device, because the guys installing it and maintaining it want to know whether the valve is doing its job properly or if it’s likely to fail.
We’re now also seeing the internet of things (IoT) coming into the home, with things like the Google Home Hub or Alexa. With all these smart hubs, it means lots of smaller devices connected to a central hub that then connects to the internet – allowing something else, like a smartphone, to control everything at once.
Will wearable technology and connected devices overtake more traditional phones and laptops?
The challenge with wearables, particularly small devices, is that you have two main problems to solve, one is power and the other is heat. Generally, if you’re going to have a highly capable device, it will generate a lot of heat and need a lot of power. As processors get better and faster, they don’t necessarily get more efficient.
There’s a lot of work going on at the moment trying to make more efficient and less power-hungry processors. There’s also been a lot of work in the mobile world which is starting to translate to the wearables market, where you’ve got devices that are very clever at doing as little as they possibly can.
I think, as it stands, if you’ve got any significant computing requirements, you still need something with a slightly larger form factor. The general rule of thumb is that the smaller it is, the less you can do with it.
The other big element, when you’re talking about wearables, is usability. It’s perfectly possible to conduct all your phone calls on your watch – I don’t know many people that do it, but it’s perfectly possible.
That’s not to say the capabilities of connected devices aren’t growing. If you look to sport for an example, when I go out on my bike I have a Garmin head unit on the handlebars. Then I’ve got sensors all over the bike, and all over me, so there are about six or seven different devices all communicating with each other, going back to this one centralised unit seamlessly with all the information I need. It all depends on what the user experience needs to be.
Will things change as we advance and less space is taken up by power sources or heat sinks, freeing up more screen space?
It really depends on how you look at it. The other area for connected devices that I haven’t really talked about, of course, is medical.
I’m diabetic so I wear a glucose monitor; I can get a glucose pump and there are various other things that you can do, and they’re all another form of wearable technology. There’s no display there and I wouldn’t really want a display, but you’ve got the same challenges with all those kinds of things.
There’s almost a tendency to assume that by ‘wearable’ you think something with a screen or something you interact with – while that is broadly true, you also have a lot of devices that you don’t consciously interact with. For example, the glucose sensor that I wear, I don’t really want to think about until it’s time to change it every 10 days or so, I just want it to work quietly in the background.
What is IoT (Internet of Things) app development?
IoT (Internet of Things) is a term that was coined quite a way back, and it’s come to mean anything to pretty much anyone who wishes to use it. What I consider it to be is putting enough intelligence into edge devices designed to do a particular job. The device then becomes an IoT device where, in addition to its core function, it’s also able to report on itself to a larger system. This then creates a swarm of devices all reporting into a central control.
The classic problem of IoT is that, in the standard format, each of the edge devices has to be able to talk to other devices in the network, which means they have all got to have a physical connection or some kind of cellular radio, wifi or satellite radio – all these things are power-hungry, expensive and error-prone.
What’s happened more recently is that you end up seeing more of a hub-based approach with a single gateway attached to lots of things that will then communicate with a system’s backend.
An IoT application is effectively an application that can be used to manage a fleet of edge devices – whether that’s tracking faults or receiving status. Going back to our dairy farm example, an IoT application might involve linking up all the flow rate sensors to let you know where milk is flowing. You could have another IoT application sitting over the same data sources, but what it’s looking at is whether there is excessive flutter in particular valves, which is a sign that the valve is about to fail.
IoT development is a beautifully loose term that can fit into almost anything, but for me, it really is about solving two problems. There’s your straight line of business stuff (e.g ‘Is my dairy farm efficiently shifting milk from point A to point B?’), and then you’ve got the more operational maintenance side of things (e.g. ‘What is the state of my deployed fleet of devices?’).
How do you think healthcare is leading the way in big data usage from connected devices?
There’s one big advantage of connected devices in healthcare. I’ve already talked about glucose monitors and that kind of thing, but you can imagine that being able to continuously collect data at a really high fidelity is infinitely better than asking someone to self-report.
If you ask someone to take their heart rate 20 times a day, what’s actually going to happen is they’re going to do it four times and make the other 16 up. Whereas, if you’ve got a continuous heart monitor, you know what it is and you can see it all the way through with more data points.
Probably the biggest advantage in using connected devices in healthcare, with regards to big data, is the fidelity of the data you collect and the non-intrusive nature of how you collect it.
Consider things like blood pressure – there’s a well-known white coat phenomenon where there’s a large chunk of people who respond to a doctor coming near them with an increase in blood pressure. It’s a problem to such a degree that there are people who have to have it on their medical notes so that they’re not treated for high blood pressure. It’s all that kind of subtlety and taking things in the right context.
How is employee and consumer data capture changing with mobile technology and connected devices?
There are two pieces on this. There’s the technical side of it, where everybody carries connected devices with lots of interesting information on them that we could use technologically to do all sorts of weird and wonderful things with. We’ve even spun off a company called RealityMine which does exactly this kind of stuff, analysing consumer behaviour – with their permission and buy-in, of course.
However, people own their data and there’s this wonderful thing called GDPR which basically says we can’t collect people’s data without asking them. So, while there’s a lot of stuff you can do if people consciously opt into it, there’s both a power and a danger to it.
I think GDPR was primarily written to deal with web tracking and how companies like Facebook were handling data. As you move into the realm of connected devices like Alexa, Google Home or your Smart TV, the amount of data they can record is absolutely titanic and I think there’s a lot of scope for what could be done with that. With the privacy advocate hat on, there’s a lot of stuff going on there that you don’t necessarily have any control over. Legislation always lags behind where technology is.
The technology, to an extent, is moving very quickly, so how legislation keeps up with it and how we, as app developers, make sure that what we’re doing is ethical, first and foremost, but also not about to become illegal, is critical. We frequently have conversations about GDPR – data protection is a big thing.
Finally, where do you see Apadmi’s place in the connected devices market going forwards?
We already work with quite a lot of people who are building or interested in connected devices, so we do quite a lot of the system integration. If I was going to look into a crystal ball, I think connected devices are going to move in two directions at the same time.
First, we’re going to continue to see increasing specialisation. You’re going to see smaller, cheaper and more specialised sensors which you can literally dust all over everything. This will require a lot of bottom-up work around how to take lots of data and use it to build a cohesive picture.
Then, you have the other way down as well, which is managing and orchestrating across large groups of things. Again, using our control valve example, you’re saying “I have a tank of milk at one end and a spout at the other, and I want to pour a glass of milk” – if you have 6,000 valves between the start and end of that process that’s actually extremely complicated.
Just thinking of consumer-grade stuff, Alexa and other voice services, there’s always a need to integrate. There are things like smartwatches and other wearables, that are increasingly making your mobile phone a portable hub. We’re almost seeing a convergence onto the phone as it can run the complicated software while the physical sensors stay compact by handling as little processor work as possible.
If you want to find out how we’re working with connected devices to bring innovation to businesses, get in touch.