New services promise exceptional performance. Just not everywhere. And not soon.
After the successes of 3G and 4G mobile services, something called 5G was inevitable. It’s still a ways off, but the outlines are taking shape.
The hallmark of 5G mobile service will be blindingly fast data speeds, possibly in the gigabit-per-second range, faster than most U.S. wired Internet services today, along with extremely short response times (“latencies”) measured in thousandths of a second.
There will be a downside. Very high data speeds take up a lot of radio bandwidth. To find sufficient bandwidth, designers will have to look toward the upper end of the usable spectrum. The FCC is now evaluating the 28, 37, and 39 GHz bands, and may add 24 GHz and possibly other bands. The problem: frequencies in this range do not propagate well, and in particular do not penetrate the materials used in commercial buildings. The old model of a cell tower serving both indoor and outdoor users for miles around – even blocks around – will not work here. A 5G mobile will have to be within a few hundred yards of its base station with nothing substantial separating them.
Think of mini-base-stations mounted on light poles or the sides of buildings, perhaps a few in each city block. Office buildings may need one or two base stations installed on each floor. Stadiums, shopping malls, hospitals, amusement parks, college campuses – anywhere people gather – will need enough base stations that every user always has one in sight. There will also be a lot of places where 5G mobiles won’t be able to reach a base station – e.g., anywhere the density of people is too low to justify the infrastructure, which promises to be expensive.
The needed spectrum may be expensive as well. Nobody knows yet how the economics will play out. Today’s 4G customers pay a steep monthly bill for moderately fast service in most of the places people go, including small towns, suburbs, and highways. How much will 5G customers pay for extremely fast service that, unless they live downtown, may be out of range much of the time?
Possibly 5G will serve as a supplement to 4G, which might hand off high data loads in places where 5G is available. In that model users would not see 5G as a separate service, but rather as part of a data package that works better at some locations than others.
These basic questions won’t have answers for a while. The FCC only recently began thinking about the spectrum issues. Its decisions are probably at least a year away, and spectrum auctions will take another year or two after that. Standards bodies don’t expect to agree on technical specifications before 2020 at the earliest. Handset and base station manufacturers can’t finish tooling up until those standards are in place.
There have been demonstrations in Europe for the last six months or so. Verizon and AT&T have each announced upcoming trials in the United States. But without firm specifications in place, chances are these will not closely resemble the eventual deployments.
But when 5G finally does arrive, those cat videos will download in a flash.