The idea of “smart cities” has been around for a while, with burgs like Barcelona and Los Angeles taking an early adopter role in programs like smart parking, smart streetlights, city-wide free Wi-Fi, electric vehicles and more. But as the Internet of Things (IoT) starts to become a real phenomenon, making more cities smart and optimized, affordably, seems to be in reach—but a few significant technology hurdles linger.
Already, cities are adopting new capabilities that offer great potential to enhance citizen safety, streamline and improve customer services and reduce overall expenditures. These are leading the way towards the adoption of IoT technologies that will connect widespread sensors through the cloud to harvest relevant data and automate decision-making processes. Think smart lightbulbs, smart thermostats, HD video surveillance for law enforcement with facial recognition, autonomous parking meters and self-driving cars, to name but a few.
Along with this, security, management and creating ecosystems of innovation with developers will all be critical challenges for operators and other stakeholders as the space develops.
A Growth-Mode Market Drives Complexity
Devices and communication networks are expected to grow exponentially as more and more of these non-human-controlled sensors and devices come online. MarketsandMarkets expects as many as 35 billion new IoT devices to be sending transmissions to servers and to each other by 2020. According to IDC, that will translate, in part, to a smart cities market that will grow at a 17 percent CAGR from from $655.8 billion in 2014 to $1.7 trillion in 2020.
Devices, connectivity, and IT services will make up the majority of the IoT market in 2020. They are estimated to account for more than two-thirds of the market in 2020, with devices like modules and sensors alone representing nearly a third of the total.
All of this additional multiplicity means that while smart cities bring great promise, there is also risk introduced through that additional connectivity and intelligence. Managing all of the data from disparate hardware and operational support systems, and applying control, resilience and security, will be the toughest challenge going forward.
“Humanity has decided to create a Utopia we call the smart city,” said Mohamad Amin Hasbini, a contributor to the Securing Smart Cities industry group and security expert at Kaspersky Lab. “This Utopia will advance our growth and performance to unprecedented levels, enabled by information systems and technologies. Its creation is an immense commitment and challenge for everyone involved, from citizens to organizations. We need to best develop, operate and defend our resources, in the most effective way, by selecting the most efficient and secure technologies.”
Addressing Top Line Concerns: Security
Perhaps no greater issue has gotten as much ink in the “smart-city challenges” space as cybersecurity. And no wonder: As cities quest to build a sustainable, more environmentally-friendly infrastructure, to be used to support smart transportation, smart resource management, smart utilities and smart physical security, to name a few—it has struck most players that much of the data generated will be rather critical.
A hacker accessing the control system for the electric grid, for example, could certainly wreak havoc, up to and including fatalities. And consider the immense amount of personally identifiable information (PII) and credit information held in things like smart-meter billing systems and even parking meter and law enforcement servers. License plate information cross-referenced with location data and credit history would be a treasure trove for fraudsters and “phishermen.”
The attack vectors are not that exotic, either: In just one example, a study from Nexusguard, a specialist in distributed denial-of-service (DDoS) security solutions, recently showed how IoT devices could be exploited during software updates and used as proxy servers to target businesses, which can then be extorted for payment.
To address these concerns, Securing Smart Cities recently released guidelines jointly developed with the Cloud Security Alliance (CSA) for the adoption of smart-city technology. The guide provides organizations with an overview of the key elements needed in order to implement the best technological solutions with a lower risk and exposure to cyber-threats.
The guidelines are for public and private organizations when planning and organizing the selection and validation of smart city technologies. It describes the types of testing and assessments that need to be considered when selecting the best and most secure vendors and technologies.
“Technology is already at the core of our cities as the world becomes more connected, and it’s extremely important that it is well protected and secure if we don’t want to end up in chaos,” said Cesar Cerrudo, CTO at IOActive.
The Need for Real-Time, Big Data Management
Security aside, all of that data being generated also brings up a management challenge. To call it “significant” is, well, a significant understatement.
To illustrate: In the ideal world, a smart grid system would monitor, in real-time and constantly, the energy consumption coming from any given house or office location. All of those locations’ data would be pooled into neighborhood buckets; and the grid would in turn use those buckets to identify how much power is needed to support the population’s demand at any given moment. In turn, it asks power plants to generate only as much energy as is necessary for that just-in-time moment---thus eliminating loss and greening up the process. A smart-grid should also be able to adjust for power surges, and could re-route capacity as needed, all on the fly.
In practice, of course, achieving this requires navigating a morass of technological complexity, not the least of which is the implementation of Big Data analytics and the building of a more efficient data center infrastructure to make somewhat affordable the high-powered information processing that will be required to make this dream operational. And it will all have to be done at immense scale.
In another example, Andy Castonguay, principle analyst for Machina Research, pointed out the challenge that exists just for autonomous vehicles that leverage a mobile network, given that they will likely take a machine-to-machine, crowdsourced approach to navigation and decision-making.
“The densification of the networks would need to be nearly complete, and the radio coverage next to public roads would need to be ubiquitous,” he explained. “It requires designing cars to use vehicle-to -vehicle frequencies set aside by the U.S. government, so cars on the highway can all make decisions together as to where hazards and obstacles are and how to best avoid them. That’s an ambitious amount of information processing to support.”
Some are addressing the challenges, bit by bit. Communications giant Verizon Communications for instance has announced a sweeping IoT initiative, including three end-to-end smart-cities solutions: intelligent video, intelligent lighting and intelligent traffic management.
The solutions are designed to help municipalities integrate disparate systems, monitor traffic and safety conditions in real-time, and manage their systems in a dynamic way in order to improve efficiency and public safety. A control platform, dubbed ThingSpace, meanwhile in theory lets users manage their IoT environments and related data, end-to-end, from device to network to application.
On the network and data center front, it’s expected that network function virtualization (NFV) and software-defined networking (SDN) will provide relief for some of the management challenges, allowing communications providers to take a central role in the enablement of smart cities and IoT in general.
“NFV and SDN are some of the building blocks needed to create programmable networks to cater to different needs—hospitals, banking, operating heavy machinery,” explained Manish Jindal, head of technology and strategy development at Ericsson North America. “They all have different requirements, be it for low latency, or a lot of security. You have to become an agile operator and for that you need an agile network.”
But, making the networks programmable alone isn’t enough. Cities will also require portals to manage their applications, and will want these to feed into the customer relationship management (CRM) system facing the end user.
“Introducing this takes a fair bit of time, so having that next-gen OSS/BSS system becomes even more important than it already was,” said Jindal. “We are starting to see larger operators already consolidating legacy software and shutting that down.”
The challenges are big, but not insurmountable. Getting the management piece right can have profound differentiation perks for the companies involved, too—so t here’s an industry-wide will to address it.
"Continued innovation in smart cities, connected cars and wearables demonstrates that IoT is the future for how we will live and work," said Mike Lanman, senior vice president of enterprise products at Verizon. "Despite the exciting potential, IoT is still too complex, too fragmented, too expensive to connect and too hard to scale. Success in that future relies on a leader that can cut through the complexity and change the IoT model."
The Role of Third Parties and Developers
It’s also critical to understand that smart cities will be developed from the standpoint of an ecosystem. Platforms will be implemented that will allow network operators, private infrastructure owners (consumers and businesses), utilities and contractors, the cities themselves and application developers to share a common basis for innovation, testing, development and implementation of smart-city solutions.
“The IoT platform is becoming more and more important in industrial IoT,” said Jindal. “A lot of specific applications will be vertically integrated. So the platform takes information from sensors in different fields and generates one set of data; that data set can then lead to action in another system.”
As an example: If a car comes within 10 km of its owner’s house, let’s say, then a smart thermostat can automatically turn on the cooling or heating systems inside the residence. That information can then be used by a smart-grid implementation to anticipate changes in energy consumption to produce more electricity, natural gas or other impacted resource.
Jindal added, “This is the new business model in the IoT, once you add analytics and make platforms open and intelligent.”
Some carriers are already looking to accommodate developers for innovative solutions. Telefonica for instance has built a developer-friendly smart-city platform using a standard known as Fiware.
Fiware allows for open access to government data, like bus schedules and location. So, in another vertical integration example, a developer could write a telematics app that incorporates weather data and delays.
“Smart city is where we start to see some of the earliest activity in creating an interoperable platform to allow different data flows to be used by third parties,” explained IHS Research analyst Sam Lucero. He added, “IBM just bought the Weather Company, and the vision is that weather can be applicable in so many contexts as an input into an application.”
Verizon’s platform also allows developers to create applications and build IoT solutions using Verizon's application programming interfaces (APIs) and application-enablement capabilities, and then code and test on the ThingSpace platform.
Not to be outdone, AT&T has an eye on mashups as well.
“We know that there are enterprise platforms out there that we need to live in,” explained Mobeen Khan, associate vice president of industrial IoT solutions at AT&T Mobility Business Solutions. “The industrial IoT data from our network may need to go into SAP or Saleforce.com, so we have to live in a world where those systems are integrated into our Control Center connectivity management platform.”
As the smart-city space continues to develop, so will the answers to the technological challenges that still haunt the sector.