Future Tech 2024: An Interview with Dr Dean Economou (Part 2)
Crowd Sourced Traffic Information, Automotive Communications Control Systems, Sensor Networks, Wearable and Ingestible Devices, Innovations in Healthcare Early Detection Systems and Drones
In this Future Tech interview, we’re speaking with Dr Dean Economou who guides NICTA research and development for impact in the Digital Economy, Broadband Applications, Transport, Infrastructure and Logistics. He was former head of CSIRO’s Networking Lab, an architect of the Smart Services Cooperative Research Centre (CRC) and developed core technology for the IEEE802.6 standard. Dean is also currently Business Development Director for Yuruware, a NICTA cloud tools start up.
In Part 1 of the interview we discussed new form factors for computing devices and displays, holograms, automotive interfaces, driverless cars and platooning robot cars.
Crowd Sourced Traffic Information
Shara Evans (SE): There was one other aspect of transportation that I wanted to raise with you, and that’s the use of crowd-sourced information. When we first started to chat about those topics, you had mentioned that there was a new paradigm, and an example of that was an application called Waze that Google has just purchased, where drivers can share information about driving conditions. How much do you think the crowd is going to contribute to this big transport and logistics sensor network — as opposed to things that might be done by the government or the automotive industry by putting sensors both in cars and on the roads where they all mesh together?
Dean Economou (DE): They’re all happening at once. The traditional — well, you come from a telecom background, so can I just say that road plans and traffic engineers are a lot like telecom heads. There’s a particular culture of how it’s done and what the right way to do it is. Meanwhile, new technologies are growing up, which are basically overtaking the traditional approaches. The official approach is that the road authority creates all the safety systems and the infrastructure, but what’s happening is that consumer innovation is such that smartphones and navigation devices are actually growing the ability to communicate, send feedback, and help be part of the coordination of traffic more broadly.
I think all of these things are working in parallel. If you play your cards right, the data you get from GPS devices or phones can actually feed into things like traffic light control. At the moment, that doesn’t happen directly — there are cultural and jurisdictional reasons for that — but eventually you’re going to need to use that richness of consumer data to manage increasing volumes of vehicles, and just people using the transport system.
Automotive Communications Control Systems
SE: Do you imagine that the communication system that puts this all together will be today’s mobile networks or something completely different?
DE: Well, today’s mobile networks are evolving. I’m actually chairing a panel in June here in Sydney for the ICC, which is IEEE’s biggest international event for wireless. Samsung’s talking about 5G technology already, which is looking at being a magnitude faster than the current 4G networks. It’s an evolution of what we currently have.
Now the nice thing about the sort of data you need to transmit in order to coordinate transport with engines moving around is actually not a lot of information. It’s just some coordinates and how fast you’re going and what kind of vehicle you’re in. It’s really not a very large amount of data. There are lots of small points of data that are changing quickly, but compared to video or voice communication, it’s actually rather small. So a very high-value communication can be easily carried on the current network.
SE: Yes, I suppose that’s getting closer and closer to the Internet of Things that we keep hearing about. In my view, though, one the barriers for this to become a reality are the pricing plans that are on the market today, where essentially you’re buying a brand-new contract for every SIM card that you have. If you can imagine having all kinds of sensors embedded in your car or in your clothes or in other accessories and needing a separate mobile plan for each one of these SIM cards, it could get to be prohibitively expensive. What’s your view on that?
DE: There are a couple of things happening. One approach is that you slave everything off the phone, so the car itself doesn’t actually have any mobile data on it, and when it gets near a phone, the phone takes over for you — so that’s one way around it.
There’s also something happening where manufacturers are actually getting the telcos to do special deals on SIM cards, where it’s a very low volume of data over five years or something, and it’s prepaid, so at the point of manufacture, you actually get a very simple data plan. That’s another way to do it.
Then the third approach, which is quite an intriguing one, is that there are companies like Taggle, which is actually an Aussie company. They’re saying what we need is 0.5G, which is a very low bandwidth, very low cost and very widespread network for all these millions of devices which don’t say much very often, but you want them connected and you want to know where they are. Taggle’s actually building entirely new infrastructure that will allow you to locate things like water meters, and tags on sheep, and stuff like that.
SE: So, really, what you’re talking about is sensor networks. Would these run over licensed radio frequencies like the mobile networks do, or would they be looking at potentially using the class licensed “free” bands or both?
DE: Well, obviously, wherever possible, they’d use class licensed bands, but it’s a very, very tiny amount of spectrum because it doesn’t need much bandwidth.
The other issue for these very large numbers of devices is battery life. You don’t want to be plugging these things into power points. You don’t want to be recharging batteries. You want something which extremely low power, and 3G and 4G tech is actually fairly high-power. This makes the case for an alternative infrastructure, which is specifically designed for some of these devices.
The other clever thing I’ve seen, which may work very well, is that you have an intermediate system where you have a gateway to the same 3G, 4G network, which is a device which has a very big battery and maybe is a bit more expensive. But, it’s got a very low-power radio interface and it can basically act as a catchment for thousands and thousands of very low-powered devices using a special radio protocol that’s designed for low-power-distributed devices.
SE: Would these devices have a relatively short-distance connectivity radius, as opposed to today’s mobile phones that can connect to towers many kilometres away?
DE: There was actually a French company that came to visit us, an interesting one. They build these devices, which are basically wireless gateways. On one side, it looks like a 3G or 4G data device, but on the other, it’s got a special low-power wireless network. The coverage for these devices is a one or two kilometres radius in a built-up area. They were showing how they could carpet Tokyo with about 200 devices, which is not really very many for 36 million people.
SE: No, not at all. I’m trying to imagine this in, let’s say, an agriculture deployment. You mentioned sheep or cows. What kind of distance would the devices cover in a completely rural environment where there might not be much in the way of interference?
DE: I’m not an expert on that, but wireless goes a long way if there’s no water and it’s flat.
SE: Yes, exactly. I’m also trying to imagine logistics and from changing out batteries and tags that are all on livestock. That would be a nightmare, so a very long battery life would certainly be a prerequisite for deploying those types of sensors.
Wearable and Ingestible Devices
Let’s move back to the area of wearable devices for people, as opposed to things. What do you see happening in that field?
DE: Well, this is a field that’s exploding. I think we’re just coming to terms with what’s going to be useful. Clearly, a device like a watch has been acceptable for a long time, so whatever we can pack into the watch is going to be a good thing.
I think something like Google Glass is great. I think we need something for audio that is going to be not very obtrusive and I think will be a very good interface. But I don’t want to talk about that too much because I might give someone a good idea. But I do think audio is really under utilised and it’s already something we wear a lot, so let’s see how that one pans out.
Some of the other interesting things I’ve seen in wearables have been innovative batteries. You can now buy batteries that are designed to go into belt buckles or be belt buckles or integrated into clothing.
We’re going to see very interesting times as people look at what can fit in your pocket, what can fit on your wrists, what can you put in your ear, what you can wear on your face, and how can those things be integrated, and which environments they work best in.
I think what we’re going to come to is that one kind of interface to a computing device doesn’t suit all environments. If you’re driving, I think it’s radically different from if you’re sitting in your office with a desktop. I think these different kinds of devices — wearables, tablets, phablets, and normal cell phones — all have slightly different functions and different environments where they’re useful.
SE: I suppose there’s one last category of wearable, or maybe you can call it ingestible, and that’s where you’re actually swallowing a pill that monitors some of your bio-functions. I’ve been reading a few things about developments in that area, and I’m sure you’ve probably heard a lot more than I have, being in your research field.
DE: Well, there’s some bizarre things. I don’t know about you — I’m not too keen on having active electronics inside me, unless it’s compensating for ever-declining memory function …
SE: Well, there’s the privacy issues that go along with it, too.
DE: Yes. And, what’s happening in the wearable space is that there’s a lot of advances in solid-state sensors that can detect biological functions. At the moment, we’re not so good at doing things like blood tests from the outside. That still requires analysis of bodily fluids and stuff like that. But vital signs and things like that now can be pretty well monitored just with wearable devices.
Innovations in Healthcare Early Detection Systems
Probably more likely, and it does sound a little bit bizarre, there are now projects — there’s one I heard of in Singapore that I know — basically looking at and analysing what comes out of the loo.
SE: Oh, no!
DE: There’s a sample there, and it can be analysed with machinery, see? Basically, your loo’s going to tell you if you’re not doing so well.
SE: The smart loo. When do you think the smart loo will be a product?
DE: Well, in Japan, it already is.
SE: Wow! So we’re not even talking about future tech. We’re talking now tech.
DE: Yes, that technology’s out there. The Singaporean government, for example, is testing what the cost-benefit ratio is. Is it worth putting these things into all the people’s homes to try and relieve the load on doctors and find out early if people have some life-threatening problems? I know it sounds a little bit icky, but in practical terms, it’s much better to detect a lot of problems early.
SE: Well, I’m still getting my head around the loo detection.
Let’s move now to another area, and it has to do with transport and logistics more generally, and that is drones. I’ve recently been picking up on some of the work that Amazon’s been doing. They have a new service that apparently is in trial mode, and Amazon is planning to use drones to deliver packages to people’s home in these little octopod-type of copters. I’ve also seen some pictures of some of the Australian companies that are also doing research in this space. Have you seen anything here?
DE: Well, my current boss has actually worked on them — we prefer to call them remotely piloted aircraft than drones. These things are very capable, obviously, in the military where there’s a lot more money, and we’ve seen their early use. Now that technology has been commoditised — again, because electronics get cheaper so much faster and the sensors are getting cheaper so much faster — you can now have a very sophisticated commercial device for a few thousand dollars that does what the military one used to do five years ago.
The issue now is not whether the tech can do it. The issue is: how do we integrate these devices into our airspace? How do we make sure they don’t accidentally land on people? How do we manage the traffic that may result from thousands of these things buzzing around? The tech is kind of there. What we need is the systems, the regulations, the processes around it, to make sure it operates well and to everyone’s benefit.
SE: I think there are also privacy issues affiliated with this. Who’s to say a drone wouldn’t have on board video and be able to do all sorts of surveillance activities? Something like that could be used for good or bad purposes. It could be used for, say, advanced scouting of a fire and ensure the safety of fire-fighters before going into a building, or it could used to invade someone’s personal privacy. Have these considerations come up in the various groups that you’re involved with?
DE: Well, obviously, proper commercial operators and government will abide by all the regulations. Those people are unlikely to abuse whatever the law says. But there’s nothing to stop private operators doing that, and I think that’s why we need the regulations and the licensing — so you know that the remotely piloted aircraft up there are actually there officially, and they’re not just some person violating people’s privacy rights. You can imagine the paparazzi loving these things.
SE: Well, I haven’t even thought about that application, but I can see how pictures could be very easily end up on the cover of gossip magazines.
There must be other form factors of drones, aside from the delivery drones that I mentioned earlier on. What kinds of form factors have you seen?
DE: Well, the drones span form factors from micro flying robots the size of insects, which could fly for a few minutes, but still could be quite useful, to absolutely enormous machines that look like light aircraft, which the military used to go vast distances.
In terms of form factors, in the city, the ones that are most useful are the ones that don’t use wings — effectively, helicopter-type devices because they could do vertical take-offs and landings. Devices with wings don’t really have range to take off and land in built-up city areas.
SE: What sort of communications networks would they use? Presumably there would have to be a commanding control system somewhere.
DE: Yes. Well, that’s an interesting issue. Usually, they can get along with a combination of existing cell networks and semi-autonomous operations. If they’re going long distance, these things are actually capable of navigating themselves, following landmarks and stuff like that. Once you get into a built-up area, they can make use of the 3G and 4G networks if they’re not too high off the ground.
SE: Well, we’re moving into such a fascinating world. What timeframe do you think it will be before drones or unmanned aircraft become part of our everyday lives?
DE: Again, I think it’s easily within 5 to 10 years.
SE: Gosh, it’s hard to think of. It just reminds me of so many science-fiction movies, and they all seem to be coming to life.
Telecom Services: The Next 10 Years
Before we wrap up, Dean, is there anything else that you’d like to add that you think will change our world in the next 10 years from a telecom services and technology perspective?
DE: Shara, I don’t see any sort of huge transformational changes. I think it’s all to do with the fact now that telecoms has become a commodity. It’s not really in the way of anything. I think we’ve got a few places where there is friction. It’s still very expensive to ship a little data from a mobile device, and I think that inhibits innovation. It’s still quite expensive to do anything on an airplane. We still don’t have good solutions for high-speed broadband in cars. I’m hoping that the cost of telecoms will become so low that it’s not going to be an impediment to innovation.
SE: Yes, well, certainly in the mobile space, data is usually more expensive than from a fixed-broadband perspective. I did a study a few years back where I conducted a tariff analysis and found that mobile broadband was more than a thousand percent more expensive than a similar fixed broadband service, which is just ridiculous when you consider the cost of infrastructure that’s carrying mobile data. Yes, licensed spectrum is expensive, but there’s, in my view, not the economics to justify that kind of cost difference.
Dean, it’s been such a pleasure speaking with you. Thank you very much for your time.
DE: Thanks so much, Shara.