Anthony M. Townsend - Smart Cities - Big Data, Civic Hackers, and the Quest for a New Utopia

Breaking the wireless network up into cells had the added benefit of reducing the amount of power needed for phones to talk to the tower. Rather than send a signal to a tower a dozen miles away, your phone would talk to an antenna just down the street. Less power per call meant smaller batteries, paving the way for much more portable devices. The brick-sized Motorola phones of the 1980s, though they seem immense to us now, were at the time a huge breakthrough in portability and convenience.

The first generation of cellular networks improved capacity by an order of magnitude over the earlier radiotelephone system—from tens of thousands to hundreds of thousands of subscribers. Prices fell rapidly too, as regulators introduced competitive licensing for different frequency bands, further stimulating demand. But once again, the density of demand in cities pushed the system to its breaking point. On Wall Street, in Hollywood, and inside the Beltway, the nation’s business and political elite, with their incessant chattering, quickly exhausted the new capacity. And so, in the late 1980s, having already sliced up the city geographically, engineers began slicing the airwaves in time.

First-generation cellular networks, which you may recall as “analog” cellular, worked like the old Bell telephone system. When you dialed, you took over an entire channel for the full duration of your call. Second-generation cellular networks, rolled out in the early 1990s, used digital signaling, which only took up a channel when you were actually talking. When there was nothing being said, part of someone else’s call could be smartly shoved into the gaps in transmission. A channel that once carried a single analog call could now carry six or more calls. Digital signal processing brought other benefits—it eliminated the echoes, static, and interference that plagued analog networks and employed strong encryption to put an end to illicit snooping; again it required less power to transmit, further shrinking battery bulk.

Of course it still wasn’t enough. Demand kept growing, as millions—entire city populations—could untether. On top of voice traffic, data traffic from wireless e- mail, web browsing, and media uploads and downloads exploded. A third generation (“3G”) of infrastructure with more frequencies, and more advanced compression schemes that squeezed more bandwidth out of them, were launched. Engineers took out their scalpels and sliced up existing cells into ever-smaller “microcells” and “picocells” so that the same spectrum could be reused hundreds or even thousands of times across a city.

Despite its slow and often painful evolution over the last century, our untethered infrastructure’s greatest challenges lay ahead. The unexpected success of smartphones and tablet computers has placed huge strains on carriers’ data networks, as they suck down screenfuls of data from the web. The launch of the iPhone in 2007 overwhelmed the feeble cellular networks in cities with dense clusters of early adopters like New York and San Francisco. Since then, global mobile data traffic has doubled every year.

Video communications may be the killer app for smartphones, but they are also killing the networks, which may be unable to keep up with demand. As 3G networks are upgraded to even faster 4G specs, streaming video to a high-resolution device like the iPad 3 can burn through a subscriber’s monthly data allowance in just a few hours. Ericsson, a maker of both cellular handsets and network equipment, reported in 2011 that “the top 5 to 10 percent of smartphone users are willing to spend up to 40 minutes a day watching online video.” As a result, AT&T projects that its network will carry more data in the first two months of 2015 than in all of 2010. By then, wireless carriers could be spending over $300 billion annually to satisfy our thirst for bandwidth (not including the actual cost of building the networks), a sevenfold increase over 2010. This assumes they can obtain the needed frequencies—with the concentration of such high-bandwidth users in dense cities, it may be physically impossible for wireless carriers to keep up. “If you had a quarter of the population of Manhattan watching a video over their handset,” explains telecom policy scholar Eli Noam, “it would take approximately 100,000 cell sites, or a huge amount of additional spectrum.”

Another potential black swan for our untethered grid is the Internet of Things. As of yet, there are few killer apps for connected things that could compete with video as a source of data traffic. But wireless will be a natural medium for connecting itinerant things to the cloud, for the same reasons it appeals to people. Even for stationary things, hooking into a wireless network is now faster, easier, and cheaper than stringing a wire. When New York City wanted to deploy a real-time traffic control system in 2011, it didn’t string fiber-optic cables to all twelve thousand-plus traffic lights. Instead, it simply piggybacked an uplink to its half-billion-dollar public safety wireless net, NYCWiN.

The future of mobile networks isn’t all doom and gloom. Up until now, every time wireless data speeds have taken a step forward, there’s been a new bandwidth- hungry app incubated in the world of desktop computers ready to overwhelm them. The fact that light waves traversing a fiber can carry far more information than radio waves in the air has meant there’s always a huge speed gap between the two media. But as we move into a world where wired connections are a thing of the past, and instead of having two classes of broadband, we may only have one, will that drive innovation in services that can live within the more restrictive bandwidth diet of wireless networks? The evolution of mobile apps, which deliver huge value even while volleying relatively fewer bits back and forth to the cloud, seems to point towards that scenario. Or will some new scheme to expand the capacity of untethered networks break this historical pattern of scarcity?

As uncertain as the future for our public untethered networks is, new investment is likely to help ease the crunch. According to IDC, a market research firm, the cellular industry could be spending as much as $50 billion annually by 2015. Governments are moving to free up more spectrum by reallocating bands abandoned by television broadcasters. Still, we are reaching limits on how much smaller cells can get. In big, dense cities, cell sites are often only a few hundred feet apart. At that scale cellular networks will begin to blur with the vast but fragmented constellation of Wi-Fi hot spots. But most mobile devices now have two radios, one for talking to cell towers and one for talking to Wi-Fi hot spots. In the not-too-distant future, as we move through the city our devices will silently shop around, switching between cellular towers and nearby Wi-Fi hot spots if we linger in one place too long. Wireless carriers in several countries have already deployed such technologies, and Cisco is leading a push for Hotspot 2.0, a new standard for global cellular-to-Wi-Fi roaming. And new smart-radio technologies will increasingly allow our devices to make use of frequencies occupied by older wireless technologies without interfering with existing signals.

Cities concentrate demand for mobile bandwidth, but the tyranny of physics constrains the amount available. They push the data-conveying capabilities of our radio technologies to their limits. Yet while untethered networks are the weakest links in the plumbing of smart cities, they are the most valuable. They free us from the terminals of the industrial age, the typewriters and the telephones that morphed into personal computers but kept us chained to our desks. Instead they allow us to merge with our devices; as sociologist James Katz puts it, they are “machines that become us.” This indispensable, intimate, and problematic piece of digital infrastructure will broker our every connection to the systems of the smart city.