The world may not be prepared for microwave mirrors, street light hot-spots, and towers visible from every backyard. But if 4G’s business model keeps heading the way it’s heading, 5G may be the only alternative.
Video: Who are the players in the battle over 5G and why do we care?
In the building of a railway the first thing is to make the surveys and locate the position of the intended road upon the ground, and to make maps and sections of it, so that the land may be bought and the estimates of cost be ascertained. The engineer’s first duty is to make a survey by eye without the aid of instruments. This is called the “reconnoissance.” By this he lays down the general position of the line, and where he wants it to go if possible. Great skill, the result of long experience, or equally great ignorance may be shown here.- Thomas Curtis Clarke, The Building of a Railway, 1889
Two equally, critically important events in the evolution of wireless communications technology took during the third week of May 2018, at or around the same point on the globe. The 3rd Generation Partnership Project (3GPP), a consortium of the world’s principal telecommunications equipment and service providers, met in South Korea to formalize the final standards documents for 5th Generation wireless (5G). From this point forward, these member organizations will at least pretend that the technical disputes regarding the content of 5G standards, have been settled.
The second was the US President’s hair-trigger cancellation of the Singapore summit bringing together the leaders of North Korea and America, presumably to discuss a roadmap for the former’s denuclearization. At the time of this writing, the summit may yet happen, but the negotiations are being played out like a reverse poker game where everyone sees the cards but the players.
Both of these seemingly unrelated summits have set 5G on an irreversible course either to global adoption or to geographical decimation. The latter outcome would ensure that Asia, Europe, and the Americas will operate somewhat different, perhaps technically incompatible, networks going forward.
If any one institution truly determines whether or not a technology or service belongs under the 5G Wireless umbrella, it is 3GPP. There are some critical elements of the future 5G ecosystem whose very definitions are up in the air, and 3GPP members are disputing them right now. Still, agreement by 3GPP members on the roadmap for Release 16 and Release 17 of its wireless technology standards may give these disputing parties a means of resolving their differences outside of public scrutiny. It is 3GPP that lights the candles in the window for 5G.
Should the Singapore summit be called off permanently, or should talks break down over the obvious issues, those candles could be snuffed back out again. The world’s largest economy — China — is North Korea’s principal ally. Not only would the Chinese constitute more than half of 5G’s consumption base by 2020, by one analyst group’s estimate, but Chinese engineers are actually responsible for having created the catalytic technology that led to 5G’s very existence. China Mobile pioneered the radical simplification of antenna design that distinguishes the 5G roadmap from the pre-existing plan for 4G. AT&T’s all-in embrace of China Mobile’s C-RAN architecture — which may have prompted Verizon to follow suit — is evidence of how serious that company is when it declares its current model of wireless transmitter deployment “unsustainable.”
Should North Korea fall back further on China for economic and moral support, and political relations between China and the US become more strained than they already are, American companies may become prohibited by US law from doing business with the very companies that forged the ORAN Alliance upon which all of 5G depends. This, in turn, would force China (keep in mind, Chinese companies and the Chinese government are, collectively, “China”) to move forward with its own variant of C-RAN and 5G (clue: The “C” in “C-RAN” doesn’t really stand for “cloud”) which would use, instead of semiconductors from Qualcomm, some “variant” of them. And that outcome would essentially force the world’s other huge Asian economies — India, Taiwan, and South Korea — to follow suit.
5G as we know it in North America would not go forward, because three-fourths or more of the world’s 5G consumers would have picked up their ball, gone home, and made their own game of it. Without the infusion of Asian capital, it’s doubtful that Europe’s 5G plans would continue in the same direction. As the US unilaterally decides to impose import tariffs on steel and aluminum from the E.U., Canada, and Mexico, the E.U. may yet choose to retaliate beyond its threatened counter-tariffs on such items as blue jeans and bourbon. And at this point, who knows whether a Brexit-ed UK would choose to follow the US’ or Europe’s lead.
We could have begun this series of adventures in ZDNet Scale at the edge of a cliff. As it stands, we find ourselves on a railroad line where we could derail at any moment.
Usually, our metaphorical maps tell a story of traveling to a particular place, or campaigning toward a specific goal. Trains in stories have a way of arriving someplace other than was intended, or taking an unplanned route, or in many cases not arriving at all. Even the best train songs ever written or sung — for instance, by Woody Guthrie — are about hopeful journeys rather than guaranteed destinies.
The story of 5G, at its heart, is about a grand bargain. Disguised as the previously scheduled program for the evolution of wireless technology, if we’re being honest, 5G is a correction for 4G. While 4G technology itself hasn’t failed, the business of sustaining that technology through both the planned and unforeseen challenges the world has heaped upon it, is itself unsustainable. The greatest of these challenges has clearly been the warming of the planet itself. In some countries the 4G model is already collapsing. It would have collapsed in China, except that its engineers were first-hand witnesses to what was happening with its own 3G, and started steering a course away from what their next evolutionary step would have been.
Fly by wire
Here is the essence of the story of 5G wireless: Operating and maintaining a 4G wireless network has quickly become more expensive than most of the experts (or so they were called at the time) had predicted. A large LTE cell typically covers a 1 kilometer radius, and under the right geographic and climate conditions, could conceivably cover a 3 km radius. But that coverage is far from being evenly distributed. The drop-off is like the slope of a tent pole, and increasing transmitter power can’t change that fact.
There’s mounting evidence that, as this “tent” ages, that slope sags even further.
Although there remains some dispute about the exact figures, here is a fact few in the communications industry will dispute: Cooling is responsible for the single largest expense incurred by today’s wireless transmitters, whether 2G, 3G, or 4G. China Mobile Research Institute — the creators of the revolutionary C-RAN antenna network architecture — points to significant, and perhaps tremendous, reductions in a telco’s capital expenditures and operating expenditures if it were to relocate its transmitter base stations’ processing power from the stations themselves to an off-site cloud data center. Conceivably, according to the Institute’s ongoing studies, it could slash annual OpEx by half, perhaps better. (The data center would have costs associated with it, of course, though these figures have yet to be estimated.)
The 5G bargain is between telcos and their customers. It is this: Telcos must replace all the previous generations of digital wireless transmission with a system that uses much, much less power. The only way to do that without cheating the laws of physics is with more towers — by one industry estimate, about four hundred times more towers. They’d be smaller, perhaps not so ugly, and could be disguised to look like other things, such as trees, church steeples, cacti, and in one extraordinary but very real experiment, buffalo.
The only way to erect 400 times more towers throughout the world is by laying down the one thing you’d reasonably think a wireless technology would exclude: Wire. Specifically, tens of millions of miles of fiber optic cable. It would go everywhere, preferably below ground, but in cases of servicing regions where tunneling is impossible, over it. Towers taller than telephone poles but shorter than 4G cell towers, absolutely require fiber optic connections to deliver backhaul — to bring the full connection capability of all the world’s networks, to each tower’s gateway. Wirelessly connecting the world’s wireless towers isn’t possible. Watch any live discussion between participants via satellite on a TV newscast (one of the few that remain) and you’ll see a live demonstration of why.
In exchange for permitting them to dig up the planet and pock-mark it with pin-pricks of towers like some mad acupuncturist with an artistic vision, the telcos promise to deliver a plethora of new services: Extremely fast Internet connections; flawlessly smooth video communications; networking capability for billions of autonomous devices in factories, streetlights, and homes; and data center services that would seek to rival, if not completely upend, the public cloud as we know it today. In fact, telcos must deliver these services, and do so very successfully, if they were ever to retrieve their investment in the overhaul of 4G.
The telcos try not to sound alarmed or frightened by the potential fate of their 4G business models. After all, they still need their 4G customers — and 5G needs all of those 4G customers. But perhaps the only way to truly communicate the severity of the situation they face, and that they know they face, would be through an analogy:
Imagine if all the world’s ground transportation providers knew — absolutely knew, as a fact — that within five years’ time, an acid that erodes the composition of concrete and asphalt would be carried by the rain across the planet, corroding the world’s highways, bridges, and city streets into an impassable gravel pit. They could forestall this erosion by funding the treatment of every square inch of road with a protective coating. But that coating would only delay the inevitable by another year or two. What’s more, the cost of that magic treatment is rising every year, and some of its chemical components are being banned. Just for the road maintainers to break even, taxes and tariffs would have to double every five years — in a world where no one is willing to pay five cents more in toll fees.
It’s literally like that.
Short ride for “long term”
The ability for current 4G wireless infrastructure in the United States to continue to be upgraded at the present rate, and still be capable of supporting today’s rate of traffic growth — assuming that rate does not itself grow — will cease to exist in as few as 19 months. This according to the telecommunications consultants at McKinsey & Company. The dead end for 4G wireless may end up happening at about the same time in our history when similar studies, conducted just five years ago, projected that the golden era of 4G would just begin.
When a courtroom’s expert witness is discredited, it’s difficult to want to hear more testimony from that person. It’s easy to blame one consulting firm when a market projection goes wrong, such as McKinsey for its 1980 forecast for the US cell phone market in 2000 — some900,000 subscribers, the firm said, when the final total was well over 100 million. In absolute fairness, 4G was not created by a consulting firm, or even a consortium of such firms. It was sold to analysts and to reporters based on criteria, much of which never came into play when these firms, along with the rest of the world, came to realize 4G was unsustainable.
The second hardest thing to do is to know what it is we don’t know. The hardest, perhaps by far, is to admit we don’t know it.
The crux of the 4G wireless portfolio was a radio technology that was dubbed Long Term Evolution (LTE). It was sold not only as fast but as sustainable. “Although most LTE markets are already saturated and most operators deployed adequate nationwide coverage,” admitted analyst firm ABI Research as recently as April 2017, “there is still room to grow.” Indeed, many countries have only just begun to deploy LTE, one of the largest being India, where ongoing LTE rollouts may be suspended to make way for 5G.
While growth in adoption for 4G and LTE has risen worldwide, growth in revenue has not. IHS Markit analyst Stéphane Téral reported in March 2017 the worldwide mobile infrastructure market was declining in revenue by over 10 percent annually. That decline became 14 percent the following year. It wouldn’t seem to make much sense: Adoption of LTE services worldwide is high, the uptake of 4G-enabled smartphones has catalyzed a huge market for devices, and the consumption of data itself is rising beyond what the least conservative analysts estimated. And yet the infrastructure business is squeamish.
In a paper published in July 2017, a research team combining the University of Cambridge’s Judge Business School and Universidad Politéchnica de Madrid’s ITC group suggested a critical element that 4G analysts evidently missed, which projecting the costs and revenues for LTE-based services: They did not model what they call the spatio-temporal dimension: that is, they forgot to account for the aging of the equipment, of the systems they formed, and of the business model that contributes to supporting those aging systems.
Common sense might tell you that, in an environment where maintenance costs were higher than projected, the market for the equipment that sustains that environment would be prospering. In a sense, it is — but not for the equipment anyone anticipated very seriously.
“While certain politicians may pretend that global warming isn’t happening,” stated Jake Saunders, vice president of research for the Asia Pacific region for ABI Research, during a May 2017 webinar, “there are a number of ongoing studies into global warming. And this will shape and shift the weather.” A 1.5 percent global average temperature rise since 1870, he noted, has magnified the damage potential for storms around the world. It’s also compounded the damaging effects of non-severe weather on outdoor equipment. And the world’s array of wireless transmitters lives outdoors, in what is truly an electrified forest.
“That is a challenge. You know, if you’ve got tens of thousands of base stations,” Saunders continued, “you want to make sure you’re minimizing the number of cell site engineers going out to the sites. And there are also constraints. With many operators, about 30 percent of them only have one antenna mount per sector; 40 percent only report two antenna mounts. With these challenges, you therefore do need to take into account a number of considerations when you’re looking at your overall total cost of ownership.”
Now that McKinsey & Company has more of these spatio-temporal metrics in hand, it’s making much starker projections for the upkeep of 4G, including in the TCO department. For one European country the firm studied (its identity is revealed in its final report), 4G network capital expenditures for its telecommunications firms were expected to rise a full 60 percent in the period between 2020 and 2025, with TCO for its 4G networks effectively doubling over that period.
“We’ve done some business modeling, return-on-investment modeling, looking at the operator’s point of view,” stated Joe Madden, founder and principal analyst with Mobile Experts, speaking to the 2018 Brooklyn 5G Summit last April. “If you invest $10 billion in a 5G network, how do you make money back?”
Madden noted the underlying trend for the decline in revenue in the infrastructure market: Telcos’ average revenue per user (ARPU) during the 4G era has been flat at best, as this six-year quarterly chart for T-Mobile’s US revenue (courtesy Statista.com) demonstrates. At the same time, each user’s consumption of data is rapidly increasing.
“How do you deal with that?” he continued. “If you just keep adding more LTE base stations, more small cells, you have a certain cost profile for delivering that additional data. And you never catch up. You basically have to bring down your cost of delivering each gigabyte of data faster than your price per gigabyte is dropping.”
5G could break mobile network operators free from this conundrum, Madden went on, provided those spatio-temporal costs fall within the parameters of the new model. He suggested that operators place strategic investments in 5G upgrades, initially selecting those regions of their service area where traffic is growing the most. If everything works out as planned, the medium-term cost of deploying and maintaining a 5G system could be significantly less than continuing to maintain the existing 4G LTE system.
This is the real appeal of 5G wireless, from a telco’s perspective. A year ago, during an investor’s conference, AT&T President of Technology Operations Bill Hogg declared, and repeatedly reiterated, its existing business model for operating its wireless antenna network was “unsustainable.” Although he brought up that word in the context of real estate costs, one did get a sense — especially from his references to the increasing pressure to squeeze the most out of its ARPU — that Hogg had several spatio-temporal variables on his mind at the time.
“We think about this holistically,” stated Igal Elbaz, AT&T’s vice president for ecosystem and innovation, in an interview with ZDNet Scale. “I really think that our advantage is our ability to build an efficient network that, for once, has the right cost structure — because it’s software, it’s cloud, it’s distributed. And at the same time, unleashing new businesses and use cases that will drive the right story for 5G in our eyes.”
Other US telcos, including Verizon and T-Mobile, are on record as intending to continue providing LTE service alongside 5G even beyond the foreseeable future, wary that 5G does not present the same benefits of downward compatibility that 4G technologies offered for 3G. The problem is, they may continue to face a steeply inclining expense curve for 4G, so long as customer demands on 4G networks’ capacity continue to rise, unless they discover a formula — perhaps collectively — for offsetting those costs. Such a formula may involve shifting the traffic burdens to 5G systems in fits and starts.
“There’s a lot of — I’m not going to say hype, [ but] expectation that 5G is going to solve all of our problems,” remarked Dr. Thyagarajan Nandagopal, deputy division director at the US National Science Foundation, speaking at the Brooklyn 5G Summit. “We have a huge feature list. Some are going to happen now in the first release in R15; for R16, we have a wish list for things to do. But the way I see it, there’s only a limited bandwidth that companies can do in any given time. Money is going to be a big constraint for everything.
“Today, we are in the early deployment phases,” Dr. Nandagopal continued. “Over the next few years, R16 and R17, we’re going to see a lot of those features come to life. But ultimately, if you look at the true wish for 5G — that anyone can be connected anytime, anywhere, with these massive gigabit pipes with minimal latencies and cutting-edge applications — that vision is going to take a while to realize. There are a lot of other issues at play. It’s not just about deploying the network. It’s also trying to make it affordable, accessible, open so you can have a wide variety of ecosystems on top of it. A network is just a pipe, in some sense.”
Imagine if California had not been discovered and surveyed prior to America’s construction of the Transcontinental Railroad. Put another way, think of how history would proceed if the builders of this nation’s, or any nation’s, infrastructure were to just plow ahead in the general direction of their goals, knowing they’d have to work together to make it across high and uncharted terrain, but without a certain vision of how things would work once those goals were reached.
The conception of 5G wireless technology is mostly complete. If the world’s telcos continue working together, they would be following a common itinerary, and their strategies would be more or less in sync. Assuming politics (their own, or someone else’s) doesn’t shatter their fragile alliance in its early stages, they do have a framework for reaching 5G deployment.
So the first few stages of our Scale journey in 5G track the progress of these courageous industrialists, while they’re still on our side of the mountain range. At Waypoint 2, we’ll give you a taste of the resistance already under way, in city after city throughout the United States, to the rewiring of neighborhoods in the name of wirelessness. It might seem a tepid resistance at first, but circumstances have conspired in a few cases to render it surprisingly effective.
Then at Waypoint 3, we’ll begin to study the convergence of cloud computing technology and the radio access network, necessary for the principal component of 5G to function effectively. It takes place at the edge, which is a place we’ve visited before here in Scale, but not from this angle.
From there, our journey westward will be fraught with uncertainties and strife. Will it be reminiscent of the Oregon Trail (the migration), The Oregon Trail(the video game), or something else entirely? Are telcos ready to become data center providers? Will an Internet of Things be supplanted by an entirely new and more cost-effective model, before it gets off the ground? And are telco equipment manufacturers truly on board with this notion of smaller, simpler antennae everywhere? What’s in it for them?
There will be a Waypoint 4, but from that point on, there be dragons. Until we pick up the center of the action again, hold tight.