« A mighty wind | Main | Insecurity by the dozen »

Fifth element

To hear the opening speakers at this week's 5G Huddle, the next generation of mobile standards (5G? do we even really have 4G yet?), will do everything from getting you breakfast in bed and attending conferences on your behalf to directing your car to deliver the groceries your refrigerator has ordered, hopefully without the "help" of some hacker who's decided you need 150 bags of kumquats and a bottle of Worcestershire sauce.

All of which was very slightly undercut by the next question: what exactly *is* 5G? Or, as Philip Marnick, the director of Ofcom's spectrum policy group put it, "What is the problem we are solving?" A little later, discussing a variant of the same question - what is broken that we need to fix? - Dan Warren, senior director of technology for the GSM Association, ran through previous generations quickly: 1G was all voice, analog and insecure; 2G brought digital and improved security; 3G was mobile broadband; 4G brought needed enhancements. But 5G? Is anything broken now or is what we have good-enough?

There was general agreement that speed isn't the key factor, although there was a nice bit of aggrieved competitiveness on display when Rahim Tafazolli, director of the Centre for Communications Systems Research at the University of Surrey noted that since GSM the speed gap between fixed lines and wireless has been growing, and "We must fix this".

The rest of his talk focused on more pressing problems: reducing latency, improving reliability, vastly improving energy efficiency, a vague something about security and privacy, and improving capacity. Capacity in particularly looks like a key issue because these folks are doing the scary arithmetic. Say 5G is ready for deployment in 2020, as planned, and predominates until 2030 or 2040. If current trends hold, the mobile networks will be handling 1,000 times as much traffic in 2020 as in 2010, rising to 1 million times the traffic in 2040 - *not* including the data rush of the Internet of Things. Wifi helpfully siphons off some of this - but not that much. At this rate, said Tafazolli, "Most cellular operators in London will run out of capacity before 2020".

Warren focused instead on expanding coverage to places where humans don't go, a key necessity for the Internet of Things: sensors monitoring temperature and humidity in the middle of planted fields, remote smart meters, and running driverless cars on roads through unpopulated areas.

"From Day One the entire road network must be covered. If not, you'd better keep tight grip of the wheel." So for him, coverage, more than bandwidth - "LTE, when you have it, is outstanding" - is the thing that's broken. That's investment, not so much new technology.

It was all sounding mostly harmless: what shall we build and how shall we build it? Massive bandwidth hoping people will find uses for it? Or imaginative, new services to justify the technology? Until the really big question arrived: how do we pay for it? The internal alarm sounded when Alistair Urie, architecture strategy director in the wireless CTO office at Alcatel-Lucent, said: "We won't be able to offer low latency on a best-effort basis. That has to be solved." With? "Sponsored data."

Sponsored data is the wireless equivalent of Internet fast lanes. It creates a two-tier system favoring the already-large and successful. It surfaced in February as a proposal from AT&T, the company that started started the network neutrality battles back in 2006. The idea is that certain sites and providers could make deals paying to exempt their traffic from inclusion in subscribers' data caps, advantaging them when the subscriber chooses which to access. Mobile networks have an advantage here: the FCC has long treated them differently from their fixed cousins.

The instinctive reaction: did he just urge the mobile world to declare war on the free and open Internet?

Wireless versus the Internet is a deep cultural clash. Asok Chatterjee, the executive director of standards for India's new standards body, TSDSI, described standards-making this way: "It's not philanthropy. It's driven by greed, and when everyone walks away dissatisfied that means it's a good standard."

The room filled with the warm laughter that signals the recognition of a truth. But the ethos could not be more different from that of those who created the Internet: when those standards were being devised, as someone said to me at a conference in 1998, if you had said something would be good for your company you would have been booed off the stage. What mattered was the goal of creating the most universally accessible network. Made over lunch, this comment was immediately shot down by a GSM guy: more people use GSM than use the Internet, he said, though the numbers may be closer than he thinks - GSM at 3.6 billion, Internet at 2.9 billion, both rising rapidly. I argued that this was temporary. GSM is an access medium that will gradually be replaced; the Internet is the thing people want to access.

That was Tafazolli's slide: "The killer application is the Internet." While everyone is wrangling over network neutrality for fixed lines, 5G could be the killer forming in the forest.

Wendy M. Grossman is the 2013 winner of the Enigma Award. Her Web site has an extensive archive of her books, articles, and music, and an archive of earlier columns in this series. Stories about the border wars between cyberspace and real life are posted occasionally during the week at the net.wars Pinboard - or follow on Twitter.


TrackBack URL for this entry:

Post a comment

(If you haven't left a comment here before, you may need to be approved by the site owner before your comment will appear. Until then, it won't appear on the entry. Thanks for waiting.)