Smart and safe city projects hold much promise, but aren’t for the faint-of-heart. Sam Fenwick gathers some advice for would-be developers
Urbanisation is one of the most visible trends of the late 20th and early 21st centuries. Anyone looking down from space at night-time can’t miss the tight constellations of street lights piercing the darkness. The shift towards city living has been a big factor in the massive increase in the living standards enjoyed by the average person today, and city dwellers consume less resources per capita than their rural equivalents.
However, there is no doubt that cities come with their problems. Crime, congestion, waste and natural disasters can all conspire to choke the life from even the most thriving metropolis. Developments such as LTE, cloud computing, video analytics and the Internet of Things (IoT) have given rise to the concept of smart and safe cities.
Thomas Lynch, director – Safe Cities and Security at IHS Markit, describes the difference between the two. He notes that ‘safe cities’, as might be expected, covers the use of technology such as CCTV, ANPR and traffic monitoring to increase the efficiency of public safety organisations within a city and to allow them to work better together. However, he adds that such activities also fall within wider ‘smart city’ projects but are typically not their main focus, with much more attention given to things like improving the efficiency of street lighting, smart ticketing or improving the flow of traffic.
Lynch adds that the low-power wide-area (LPWA) technologies that are expected to play an increasing role in smart city projects are not suited to the bulk of safe city deployments, due to the emphasis on video (which he still sees as mission-critical LTE’s ‘killer app’) and the resulting need for broadband levels of bandwidth.
According to IHS Markit research, the global safe city market was worth more than $15bn in 2017, with video surveillance and analytics, command and control, broadband LTE, ANPR and computer-aided dispatch (CAD) and physical security information management (PSIM) software being key components. Lynch adds that this figure is projected to rise to more than $20bn in 2020, with Asia accounting for nearly half of that market currently, due in part to the preference for the installation of completely new systems as opposed to upgrades to existing infrastructure. He adds that as far as the latter is concerned, there is a much more even split between the three major regions: EMEA, Asia and the Americas.
In the case of the MEA region, IHS Markit values the safe city market at around $1bn. Lynch notes that the drivers behind safe city adoption vary from region to region. “In North America the drivers are about sustaining economic growth, protecting against security threats and terrorism, but also crime – it’s a major issue. Whereas in the Middle East and Africa, crime isn’t a major issue, but security threats and terrorism are, while the push for greater prosperity is a major factor in the Middle East, but less so in Africa.”
He adds that the push towards greater data sharing between agencies is also a key driver behind safe city take-up in these regions at the moment, noting that in some countries there can be many different, disparate organisations, including almost paramilitary-style groups. Lynch also highlights the fact that while countries in the MEA region may well be behind their European counterparts, they often aren’t subject to the same financial constraints and therefore “will push ahead over a period of time”.
As safe city projects roll out, one of the considerations for their designers is where to have the majority of the intelligence on their network(s) – either on the network edge (typically in user devices) or on the cloud. Lynch says the answer to this question heavily relies on the technology being used, adding that this is a big question for video: “A lot of it boils down to cost and the user’s exact requirements. If you need real-time intelligence, then for a small network, intelligence on the edge probably works quite well, but if you’re pulling in lots of data feeds in a big city, having the intelligence on the cloud makes more sense.”
He gives the example of the recent search in Barcelona following the terrorist attack in mid-August. “They were searching restaurants, public buildings, marketplaces, roads, underground passes, that type of thing. In that example, edge-based technology might not have been as effective as being able to have a centralised resource capable of cataloguing and monitoring all of that information at once.”
Lynch adds that while intelligence on the edge can be used to make the total amount of data going to the cloud manageable in a large-scale deployment scenario, because much of the intelligence required in a safe city environment is focused on supporting real-time intelligence, being able to better manage operations is typically more of a priority than efficient usage of over-the-air data.
Over in the US, the events of 9/11 have led to the creation of 78 ‘fusion centres’ which, according to IHS Markit, look at big data sets from both traditional police data and many other sources, including video surveillance streams, social media, arrest records, warrants and mug shots. These centres provide intelligence packages for patrol briefings, and operations. They also support real-time intelligence cells, which are typically attached or embedded into control rooms to provide intelligence to support either command and control decisions or deployed officers. The firm believes that these will become increasingly important given the growing emphasis on predictive policing and that the systems that underpin this approach require a great deal of information to work effectively.
The analyst firm predicts there will be 88 large-scale smart cities by 2025, but there will be hundreds (if not thousands) of safe cities by that point. Lynch adds it is fair to say that there is a danger of “death by pilot”, and perhaps smart cities would roll out faster if cities adopted more of a one-size-fits-all approach.
“Projects associated with mobility and transport are more than double that of safety and security, and it’s similar with energy and resource efficiency. However, we have started to see a lot more safety- and security-orientated smart city projects.” He returns to Barcelona, noting that while it has smart city projects of its own that have focused a lot on non-public-safety projects, he is “sure that public safety will be a major concern going forward”.
He explains that while there are many small-scale smart cities at the pilot stage or that are currently operating with low revenues, safe city projects involve command and control upgrades or completely new installations, often a new critical communications network and the roll-out of thousands of cameras. He cites Shenzhen as possibly a good example of this – by the end of 2017, it will have deployed 1.3 million cameras, many of which are planned to be connected through a public video cloud solution. This will be one of the largest unified security systems in the world, with the Chinese government passing legislation to support a unified approach to security.
“Ultimately, in terms of investment, rolling out a safe city is a much higher-level project, requiring more significant ICT infrastructure to support frontline operations, than a typical smaller-scale smart city project,” Lynch concludes.
Have a plan
Jane Rygaard, Nokia’s head of marketing, advanced mobile network solutions, says that smart city projects “where there’s a clear strategy from the beginning, not for the individual use cases, but for the city as a whole, are always the ones that go quicker and experience fewer obstacles”.
Rygaard explains that the reason for this is that a lot of smart city use cases go across silos, requiring data sharing between public and private organisations, as well as the general public. She adds that “simple things like being able to share data from one department to another seem really trivial, but they’re not”, due to barriers created by legislation and working practices.
“If you don’t think that into your strategy, there’s definitely places where it holds up deployments. I can give examples from the public safety side of places where data needs be to be handed over by law in writing with a signature.
“If the laws and legislation have not been updated for the world of IT we’re now living in, you can plan many smart city use cases but you can’t implement them, because the data is what builds the use cases,” she adds.
“It’s not all within the municipality or city domain because some of the organisations might be private entities. We need to have much more of some of these discussions as an industry. There’s a need for action on this issue because otherwise we might end up not being able to make the technology work for us and we’ll be hindered by the silos we’ve created.”
Shifting focus to safe city projects, IHS Markit’s ‘The Benefits of Safe Cities’ white paper also emphasises the need for a clear strategy: “The projects which have been successful had one major element in common: the various stakeholders worked together and constructed a consolidated project team and dedicated safe city initiative. Projects fail because the different agencies do not identify unified objectives, are unwilling to collaborate, or do not engage the city’s citizen stakeholders effectively.”
It additionally states that successful projects also have a steering committee that takes the lead and secures funding, strong partnerships between a range of agencies, and specific project goals in terms of security and the technologies that will be implemented.
Rygaard says there needs to be more sharing of the lessons learnt by smart city projects, both within individual countries and globally. She adds that while there are some things that can’t be replicated from two of her favourite smart cities, Dubai and Seoul, due to their regulatory environments, there is also “a lot of good learning” that other smart city developers can apply to their own projects.
She adds that one of the best types of learning to share are concrete use cases, giving New Zealand’s flooding prevention system as an example. This uses sensors to measure the amount of sludge in individual drains, so that maintenance teams can respond by cleaning those that need cleaning the most, rather than sticking to a rigid schedule.
Rygaard doesn’t believe smart cities are being held back by the pace at which massive IoT features are being standardised within 3GPP, and perhaps for good reason, given that the world of IoT is just beginning to get to grips with narrowband IoT (NB-IoT) and the latest version of LTE-M. She says the range of wireless tech that can be used by smart city projects requires smart city applications to be access-independent. There is also the need for IoT platforms to detect network behaviour abnormalities, especially as “you cannot assume that all the connections you have out in the network are secure endpoints, as you can in a controlled network”.
Finding problems, creating solutions
Ali Raza, professor at Rochester Institute of Technology (RIT) University and principal consultant at Norconsult Telematics, warns that some organisations are aware of the technologies associated with smart cities such as NB-IoT and blockchain and the need to adopt them to “stay ahead of the curve”. They don’t know how to use them to add value to their operations, highlighting the need to gather feedback from end-users so they can be used to address real-world problems.
At CCW, he gave a good example of one such problem: Dubai Corporation For Ambulance Services’ struggle to get through the city’s traffic to reach patients and get them to hospital within an acceptable timeframe. According to the Khaleej Times, 49 babies were delivered in an ambulance between the start of 2016 and 14 November of the same year. Raza says the situation is partly due to Dubai’s high number of construction projects. The expansion of roads, flyovers and bridges is a positive sign that the Dubai Road and Transport Authority is focused on addressing these issues.
“After doing a bit of research on traffic analysis with my students, Raina Zakir and Jhealyn Samson, we found that once you come towards a red traffic signal, because drivers are static, they do not know where to move, and that’s where ambulances get stuck the most,” Raza says.
“We took on this challenge and tried to solve this problem in a secure way with 3GPP device-to-device proximity services and narrowband IoT sensors. When a D2D beacon is detected from an ambulance, they transmit that information using LTE’s NB-IoT over a macro-coverage area to command and control, and it can tell the correct traffic light to go green and all the other traffic lights in the vicinity to go red, and that way the traffic can move, cars can spread out and the ambulance can make it through.”
Away from the headlines about robot policemen and Bugatti Veyrons, there’s no escaping the fact that Dubai has heavily invested in safe city projects. IHS Markit reports that an increase in government investment in this area of 29 per cent has made its citizens safer from individual risk and property risk by three per cent and 31 per cent respectively, while also resulting in other significant social benefits.
One issue that a city’s fire department raised with Raza and his team is how to accurately locate its firefighters when they are responding to incidents in skyscrapers. “A fire chief doesn’t want to know whether his firefighters are on the first or second floor, he needs to know where exactly they are so they can be rescued if necessary.”
He notes that doing this with TETRA or LTE requires a lot of investment in in-building coverage. “So, after giving it a bit of thought, we looked at a concept very similar to Bluetooth Low Energy but which doesn’t use the ISM band. We’re looking at using a very small part, 200kHz, of spectrum that is allocated by governments to mission-critical and public safety entities to be used over LTE.”
He explains that this approach uses small low-powered beacons that do not implement the complete protocol stack – they only send out a small beacon signal that has the primary and secondary synchronisation signals and the cell ID.
“When a firefighter is on a particular floor on their mobile phone, they can immediately download a floorplan. With that floorplan, we have the exact location of these narrowband beacons, so the cell IDs are already known to the device. The device looks up the signal from the different beacons, which we can do frequency reuse with, to pinpoint the location of a firefighter to within 23 centimetres.”
Raza adds that he and his team are “looking to implement this in a very low-cost manner, in terms of investment, deployment and power usage”.
In the rush to take advantage of new technology, it can be easy to lose focus. It seems anyone under pressure to push ahead with either a smart or safe city should take a deep breath and come up with a plan based on the needs of end-users and that addresses real problems, and then execute it. A great deal can be done with today’s technology if you are prepared to get creative.
