Creating the connectivity infrastructure for smart cities requires an open-minded approach to technology, reports Barnaby Page
Smart cities are founded on data – and the sheer quantity of it is staggering.Around 50 billion devices are expected to be connected by 2020, not coincidentally the year targeted for the widespread launch of one of the most important new communications technologies, 5G wireless (more of which below), with volumes of data traffic perhaps as much as thousand times higher than now.
Microsoft, for example, says it collects 500 million data points from its 120-odd premises in the Seattle area every 24 hours: scale that up from a single employer (albeit a very large one) to a whole urban region’s building stock as well as all the other elements that can be connected to a smart city’s intelligence (ranging from weather sensors to streetlights) and it’s apparent that quite apart from storing, analysing and acting on the data, simply moving it around is a challenge that will make significant demands on the digital infrastructure.
Providing fast, high-capacity, robust data connectivity has thus become one of the most important tasks for those in the public and private sectors charged with developing smart cities. To a large extent, this is the infrastructure that underpins all the others: so important, in fact, that smart city visionary Gordon Feller – now with Cisco’s Internet Business Solutions Group – has said that broadband should be regulated as a fourth utility along with electricity, water and gas.
As Microsoft, again, puts it in pointing out how other smart city projects depend on data: “The evolution of energy infrastructures won’t be achieved without information technologies. All cities will need to leverage systems that are connected and supported by a large volume of real-time data from diverse public and private sources.”
Those will, according to Gartner, include hundreds of millions of connected devices in five major locations: homes, commercial buildings, transport, utilities and public services (in that order). Indeed, the need for smart city connectivity and the type of solutions it requires are inextricably tied in with the growth of the Internet of Things. The connectivity requirements of these individual devices is low but their massive numbers and the frequent need for real-time communication make them demanding in connectivity terms.
Therefore, according to software company Escher Group, “the first step for any policymaker is to foster the development of a rich environment of broadband networks that support digital applications, ensuring that these networks are available throughout the city and to all citizens.”
But the challenge is more than just building further-reaching, faster networks. Not only does the smart city require a great deal of connectivity, it also calls for many different kinds of connectivity to match the multifarious different forms in which its data comes, and the sources that provide it.
For example, there are constant, trickling real-time feeds from sensors and actuators, as well as much higher-bandwidth video. Then there are are data sets in the cloud, not updated in real time but possibly far bigger, such as historic energy consumption or traffic information. Data may come from physical plant owned by the city itself (road traffic sensors, for example), from businesses (whether they are utilities providing city services or other companies consuming them), or from households. Conventional IT networking is still important, but it is likely to be overshadowed in terms of complexity and sheer numbers by the requirements of devices like sensors.
All these users of connectivity need to be linked in ways that are cost-effective but still provide them with enough performance to function usefully; they need to be prioritised where bandwidth is limited, which it is often likely to be as the number of connected devices explodes, and in order to ensure that critical systems never go offline; and of course they need to be monitored, managed and secured. Connectivity for the smart city is not about investing in a single product but about provisioning an infrastructure that meets demands as variegated as the city itself.
Complicating the task is the fact that in terms of connectivity as elsewhere, smart cities must build on the foundations of existing ones, and do that incrementally: it’s not simply a case of rolling out a brand-new connectivity network that can tick all the boxes discussed above. There will be legacy networks, and legacy equipment, in place – much of which may not be easily compatible with newer tech.
With this in mind, there’s a strong tendency to conceive smart city connectivity as a mix of technologies and architectures rather than a single, all-encompassing solution, and to focus on providing the flexibility to meet the widely varying demands of different applications rather than simply throwing bandwidth at the problem.
As Huawei, the Chinese networking and telecommunications company which is among the leaders in the field, has said: “The future network will focus on the different business applications and user experience other than just the pursuit of the greater bandwidth and volume. This will raise the requirement to build service-oriented networks to quickly and efficiently respond to user needs, as well as to offer consistent and high-quality services for different use cases.”
With connectivity, as with smart city IT in general, mixing up solutions in this way also means working with multiple vendors, keeping the big picture in sight, and not falling into the trap of viewing the data network as a single answer to a single problem.
Schahram Dustdar and colleagues at the Vienna University of Technology are among those who have expressed their concern: “Current smart-city services are typically provided in single domains – for example, building management, transportation, and health care, among others. With such services, domain-specific application requirements drive all system-component design and determine most technical choices, ranging from sensors and smart devices to middleware components and computing infrastructure.
“The service-delivery process is rigidly orchestrated by domain solution providers who develop applications and integrate subsystems from various vendors. This model leads to many closed vertical systems with tightly coupled hardware, networks, middleware, and application logics. Scalability and extensibility are intrinsically limited in such systems, and the closed relationships between stakeholders stifle the creation of novel services.”
The lesson, of course, is to avoid this by taking a higher-level view and emphasising integration as much as technical solutions to individual problems.
How, then, does all this translate into practice? There are competing technology approaches, and though nearly everyone is agreed that wireless has a very major role to play (in part because it’s comparatively easy to install), it may not be the whole answer. Sierra Wireless, for example, advises that smart cities don’t need to throw out their connectivity legacy: they should look at making better use of existing public networks, building centralised networks to integrate what is there, and unifying different communications technologies.
This heterogeneous philosophy applies to technical solutions too. Escher, for example, advocates a combination of cable, optical fibre, and wireless networks rather than relying on any one comms technology.
Some characterise this concept as “demand attentive networks” (DAN), which instead of blindly maximising network speed at all costs, instead intelligently supply just what’s needed to each user. The Institution of Engineering & Technology sums up the DAN vision: “[The] key underlying principle is that it is not necessary to have…high speed capacity available everywhere at all times; it is enough if the device or appliance which is using the connectivity finds that it is not constrained in its demands by the network infrastructure in place. This can be achieved by a combination of technical standards, network architecture and smart regulation which work together to organise the demand for bandwidth in real time.”
Of course, speed and capacity will still be important for many smart city applications, and these are two reasons why virtually all concur that one new technology in particular holds many of the keys: 5G, the next-generation wireless system that will follow on from today’s 4G and which should be rolled out around 2020.
5G has many advantages for smart cities. It is, needless to add, faster than 4G, providing data rates up to ten gigabytes per second. Using a much wider band of spectrum, it opens up scope for more wireless networks to coexist in the same area. It will support very large numbers of simultaneous connections (up to a million per square kilometre), which are crucial for smart cities since they are not only linking computers and phones but also all those sensors.
It promises energy efficiency, which is important given the vastly increased quantity of network infrastructure equipment and wireless-connected devices that can be expected. And, perhaps as importantly as speed and capacity, 5G will enjoy low latency – in other words, a minimal delay between a signal leaving the sending device and arriving at a receiving device.
This is not the same as bandwidth; though that is often colloquially referred to as the ’speed’ of a network link, bandwidth really refers to how much information the network can – theoretically – convey in a second, whereas latency is the actual delay time suffered by any signal
On 3G the latency should be one millisecond, far lower than the 50 milliseconds on 4G or 100 on 3G, and this has profound implications for smart cities because it opens up the possibility of monitoring and controlling remote devices, over the network, in what to all practical purposes is real time.
Not all intelligence will have to be built into the device, whether it is a self-driving train, a smart traffic light or an air conditioning system: that low latency will mean it is feasible for devices to be controlled by software on a central server (or, more likely, in the cloud) without experiencing unacceptable time lags in communication. And as a bonus, keeping to a minimum the amount of processing conducted within the device also prolongs battery life, an important consideration when dealing with hundreds of thousands of IoT sensors.
As Paul Bradley, 5G and transport director at the telecoms firm Gemalto, puts it: “In terms of the practical applications, that low latency is what is going to provide the essential real-time and ultra-reliable interactivity for any smart services using the cloud such as, for example, self-driving connected cars, healthcare monitoring systems, AI robotics and other industrial applications. Autonomous drones and driving networks are going to need incredibly fast response times, yet won’t need particularly fast data rates.”
All the same, 5G may not be the whole of the story. There are other technologies emerging specifically to address the needs of smart cities and the IoT, including many different flavours of low-power wide-area network (LPWAN) which allow limited-bandwidth but long-range, low-power wireless communication among devices.
If smart city connectivity is not a product in a box, nor is deploying it a single project. Instead, it will grow with the city, again necessitating a flexible approach. As the Smart Cities Council explains, “cities have several alternatives for connecting to their smart devices citywide. They can ask their IT department to hardwire the connections one at a time. Or they can hire an integrator/consultant to do all that messy work. Or they can turn to a connectivity platform – a collection of tools and service optimised to connect to any and every device. You talk to the platform in a standard way. The platform then translates your requests across multiple communications networks to multiple devices from multiple vendors.”
Not only does this address the worries of those like Dustdar and his Viennese colleagues who warn that a narrow view of the technological challenges will result in narrow, limited solutions. It also provides some future-proofing.
The vendors of smart city connectivity tech are likely to include many of today’s big names in IT and telecoms – as we see in our section on suppliers, below – and the array of devices and technologies they provide will no doubt include many that are already familiar. But the advantage of taking a platform-based approach rather than concentrating on linking specific pieces of technology is that new ones will surely emerge too (there is already early research being conducted into quantum-based 6G networks) – and so, as the theme of mixing it all up that we’ve seen throughout this exploration of connectivity suggests, the smart city will be one that leaves its options open.
Connectivity in practice
Widespread Wi-Fi was the first highly visible sign of connectivity making cities smart – ranging from public-facing projects like the LinkNYC scheme in New York City, which has put free Wi-Fi kiosks on the site of old phone booths, to more infrastructure-oriented rollouts like that in the new city of Songdo near Seoul, where wireless is used to connect everything from street lights to utility meters.
Many other projects have involved leveraging mobile networks. Dubai, for example, has enlisted the help of UAE telecoms operators while Tainan in Taiwan has launched services based on FarEasTone’s cloud infrastructure.
Guadalajara in Mexico, a poster child of the smart cities movement, is providing fibre-optic links to homes at the same time as configuring older networks used by services such as police and weather forecasting to work with new mobile and sensor infrastructure.
Meanwhile, 5G pilots are springing up all over the world. Estonia will be one of the first to pioneer a large-scale 5G mobile network this year. Verizon plans to switch on its U.S. pilot in 2017, while in South Korea and Japan, SK Telecom and NTT DoCoMo respectively have promised 5G in time for the Olympics they are hosting in 2018 and 2020.
Both will use light fidelity (Li-Fi), a form of communication based on light waves from LED bulbs rather than the radio of Wi-Fi.
Vendors to watch
Smart city connectivity and, particularly, 5G are unsurprisingly attracting the attention of nearly all the big names in technology and telecoms. Some are focusing on the nuts and bolts of building the networks, while others are more oriented toward applications, particularly those involving the IoT.
Ericsson intends to become a major player in 5G tech, as do Huawei and Nokia. Verizon is focusing on smart cities applications which will likely heavily rely on 5G, including intelligent lighting and traffic management, while Australia’s Telstra and Germany’s Deutsche Telekom are other telcos committed to 5G.
Further heavyweights likely to make their presence felt include Alcatel-Lucent, Cisco, Orange, Qualcomm and Samsung. AT&T, meanwhile, is positioning itself at the centre of smart city connectivity through its Smart Cities Framework. It will lend its expertise on networking to IoT developments involving big-name partners such as GE, Hitachi, and IBM, with early projects planned for Atlanta, Chicago and Dallas.