fotohansel - Fotolia
When IT pros from around the world come together to solve a problem with IoT for smart cities, everyone learns something new. Itron's Smart City Challenge drew on the IoT expertise of tech startups from more than 20 countries to help the cities of London and Glasgow solve difficult challenges.
London's challenge winner I-Tech created a detection system of IoT sensors and lifebelts combined with Itron's standards-based developer tools to detect people jumping into the Thames River. I-Tech's detection system had to be capable of identifying the precise location of people who have fallen from the bridge even in thick fog to save lives and improve response times.
Glasgow's challenge winner Noesis developed acoustic sensors on lampposts to reduce noise pollution during conferences and events with large crowds in the city. The organization can update the acoustic sensors over the air. In the future, the sensors could be used for situations such as traffic management or public safety.
Itai Dadon, Itron's director of smart cities and IoT, assisted the startups and cities with creating the technology that would solve their problems. The organization creates technology systems and provides basic infrastructure, networks, security and data platforms for cities to better manage energy and resources. Dadon discussed the lessons learned from using IoT for smart cities.
What inspired this smart city challenge and what did you learn from it?
Itai Dadon: What we found out is that every city has slightly different needs and different challenges, [and] that one portfolio -- even though it's flexible and customizable -- doesn't necessarily answer all the different issues that different cities see in their community. These challenges are interesting in that they apply to many cities around the world.
For example, the problem for London was people, unfortunately, jump off bridges. Because of the temperature of the water, they have minutes before hypothermia sets in, and they just die. Knowing at the second when one such event happens -- being able to get to the rescue and understanding that the equipment is there when it's needed -- that used together with the police of London and the city of London made creating this technology an extremely important project.
We found a great level of creativity from the solutions that we got. In this case, they used instrumentation technology with special optical laser waves around the edges and deck to detect whether it's just an object falling off the bridge -- someone's coffee cup -- or if someone actually fell. We didn't know how to solve this problem either, so we wanted to hear from the community and from people around the world about how they thought of solving these problems. It's amazing how small this entire world is in terms of how we could reach companies with solutions from the Middle East, Asia Pacific, Europe and here in North America.
This at the end of the day is really about creating a new technology, not just for the sake of technology, but for the sake of transforming people's quality of life, whether it's around the social, environmental or economic challenges that they face.
What difficulties did you find from this challenge that the IoT industry still needs to address?
Dadon: For specifically industrial IoT, our challenges were around creating a solution that falls into the economic reality of the situation. That means that the total cost of the solution -- whether it's the products, software, installation or maintenance -- all fits within a business model that makes sense for eventually the customer or, in this case, the city. To achieve that viable business case, you really need the technology to push the envelope on the life of the product.
For example, we have devices that we need to install and put in the field for 20 years without ever being able to touch them. Products would need a network that allows updates of devices over the air, and over time, basically scale the functionality and fix the security programs. In each solution, we had to guide our partners to understand all the components of what it means to take a device into the field. They had to think about all the aspects of configuration and installation to really make those devices mature enough for actual field productivity.
What advice would you give to companies that are interested in implementing IoT?
Dadon: Never underestimate the importance of security and securing the device. Maybe your customers are not demanding this, but this is what will make or break and any deployment in the future. It's our job as professionals in the IoT industry to push security forward. However good your security solution is, the moment you put that device in the market -- the second after that -- that solution is already obsolete. Someone's already probably found some weak link in the solution. Being able to update the software of those devices over the internet is critical in any IoT technology.
Point number two is really thinking about the lifecycle of the device and thinking about all the tools needed around the device itself -- for provisioning, configuration and setup -- and how do you manage the lifecycle of that device from the moment it leaves the manufacturing line to the moment you want to decommission it from your network. Think about all these things with an understanding that you probably don't know today what the functionality of this device will be in a few years from now. Take into account the technology that you're using in terms of how long that technology will live. Is the technology going to be backward and forward compatible with the next generation?
The last thing is the reliability. Sometimes you have devices that you don't really care how reliable the communication to them is. If it's a cattle monitoring application, if you don't get a message of where exactly the cow is that very moment, nothing critical will happen. If you take the other side of the spectrum, when we're deploying methane gas leak sensors, if one of these gas detections senses a leak and that message will not reach its destination, there's people's lives at risk. Making sure that your network capital keeps the level of reliability and uptime and the ability of devices that the situation calls for is extremely important.
Are there any challenges of IoT for smart cities that you run into with services -- such as energy or water -- that you might not find in manufacturing, for example?
Dadon: The idea that wherever you put the device, it will just connect if you have a network is plain wrong. It's a challenge that is tough to overcome. On top of that, another dimension of complexity is that urban environments are not static; they're always changing, buses are going around, and they could put a whole new building up. These are things you cannot control. That's one of the main reasons why we chose the mesh network topology, because one of the interesting characteristics there is that it has self-healing and self-forming capabilities. Whenever a device is working, the connection links to the network, if that link is lost because a building was built in the middle, our devices will automatically find alternative ways of connecting to the network via another device. All that happens automatically and transparently.
When you're in a manufacturing plant, you're probably dealing with different challenges, like the super critical latencies you need to achieve. But you can connect all the devices to a very high throughput network through an Ethernet cable, fiber optics or Wi-Fi.
Where do you see the future of IoT devices?
Dadon: IoT sensors are becoming more sophisticated, improving costs and improving power. We're really creating a transformation. The real disruption is when we're bringing the ability to use information from multiple types of sensors and solutions together. Add on top of that using machine learning or AI to drive insight that we have no other way of getting.
For example, Itron showcased at IoT World in Santa Clara a scenario of a natural disaster and explained how we created resiliency in cities facing those natural disasters. With climate change, we're seeing more and more natural disasters, unfortunately. We showed an earthquake and how we can, in real time, get a lot of information of what's going on in the field during an earthquake and immediately after an earthquake. We can see the shape and the magnitude of them thanks to the sensors that we have in every smart meter in every house. We can then understand what's happening with the grid, where lines are going down, if fires are erupting, if gas leaks are happening, how is traffic going, are people running around or blocking streets and have accidents occurred thanks to our acoustic sensors.
With all this information applied to a synthetic model created by our partner Runwithit Synthetics, we created a personalized evacuation scenario for the affected area where each person gets a personalized evacuation delivered to their phone. You and I could be next to each other, but I would get the order to evacuate via one road and you'll get a route in completely different direction. Runwithit Synthetics optimized the evacuation of the population out of the danger zone in a way that is much more efficient than the typical just letting people know to evacuate and everybody will go through the same route and basically just block it. Then it will take hours more. And unfortunately, it will put people in danger. That's just an example of the type of insights and things we could do with that information that's we're quite far from doing today.