Verizon 5G Home will begin accepting new customer orders for its in-home wireless broadband replacement as of this Thursday, Sept. 13, with a scheduled service launch date of Oct. 1.

The new high-speed wireless service will be available in select parts of Houston, Indianapolis, Los Angeles, and Sacramento.

Verizon CEO Hans Vestberg is calling the service part of Verizon’s 5G Ultra Wideband network. Initial reports indicate speed will range between 300-1,000 Mbps and existing Verizon Wireless customers will get a $20 price break on service — $50 a month instead of $70 for non-Verizon Wireless customers. We are still waiting word on any data caps or speed throttle information. Verizon informs Stop the Cap! there are no data caps or speed throttles. Service is effectively unlimited, unless hidden terms and conditions introduce unpublished limits.

Interested customers can determine their eligibility starting at 8 a.m. ET on Thursday from the Firston5G website. If you are not eligible initially, you can add your email address to be notified when service is available in your area.

Early adopters will be awarded with a series of goodies:

• Free installation (a big deal, since it could cost as much as $200 later. An external antenna is required, as well as in-home wiring and equipment.)
• 90 days of free service (a good idea, considering there may be bugs to work out)
• 90 days of free YouTube TV (a welcome gift for cord-cutters)
• Free Chromecast or Apple TV 4K (a common sign up enticement with streaming cable-TV replacements)
• Priority access to buy forthcoming line of 5G-capable mobile devices

Customers in the first four launch cities will be using equipment built around a draft standard of 5G, as the final release version is still forthcoming. Verizon is holding off on additional expansion of 5G services until the final 5G standard is released, and promises early adopters will receive upgraded technology when that happens.

Verizon is clearly providing a greater-than-average number of enticements for early adopters, undoubtedly to placate them if and when service anomalies and disruptions occur. Although Verizon has done limited beta testing of its 5G service, it is very likely the 5G network will get its first real shakeout with paying customers. Unanticipated challenges are likely to range from coverage and speed issues, unexpected interference, network traffic loading, the robustness of Verizon’s small cell network, and how well outside reception equipment will perform in different weather conditions, particularly heavy rain and snow. With a large number of freebies, and no charges for 90 days, customers are likely to be more forgiving of problems, at least initially.

Chromecast

Verizon’s 5G network depends on millimeter wave spectrum, which means it will be capable of providing very high-speed service with greater network capacity than traditional 4G LTE wireless networks. But Verizon will have to bring 5G antennas much closer to subscribers’ homes, because millimeter wave frequencies do not travel very far.

Verizon will combine a fiber backhaul network with small cell antennas placed on top of utility and light poles to reach customers. That explains why Verizon’s initial 5G deployment is unlikely to cover every customer inside city limits. There are substantial deployment costs and installation issues relating to small cells and the optical fiber network required to connect each small cell.

Verizon’s existing FiOS network areas will offer an easier path to introduce service, but where Verizon does not offer its fiber to the home service, it will need to bring fiber optic cables deep into neighborhoods.

AT&T sees a similar challenge to 5G and is openly questioning how useful wireless 5G can be for urban/suburban broadband service, considering it can simply extend fiber optic service to those homes and businesses instead, without a costly 5G small cell deployment.

Verizon introduces 5G wireless in-home broadband in four U.S. cities and starts taking new customer orders on Thursday. (1:00)

Article updated at 6:28pm ET with information about data caps and speed throttles provided by Verizon.

This week, the FCC announced bidding has finished for the latest Connect America Fund (CAF) broadband subsidies auction.

Once again, the FCC gave first priority to incumbent phone companies to bid for the subsidies, which defray the cost of expanding internet access to homes and businesses otherwise unprofitable to serve. Nearly $2 billion was left on the table by disinterested phone companies after the first round of bidding was complete, so the FCC’s second round opened up the leftover money to other telecom companies.

Winning bidders will receive their portion of $198 million annually in 120 monthly installments over the next ten years to build out rural networks. In return, providers must promise to deliver one broadband and voice service product at rates comparable to what urban residents pay for service. The winning bids, still to be publicly announced, will come from rural electric and phone cooperatives, satellite internet providers, fixed wireless companies, and possibly a handful of cable operators. But much of the money overall will be spent by independent phone companies rolling out slow, copper-based, DSL service.

Because the total committed will take a decade to reach providers, rural Americans will likely face a long wait before what purports to be “broadband” actually reaches their homes and businesses.

While many co-ops will spend the money to expand their own homegrown fiber-to-the-home services, most for-profit providers will rely on wireless or copper networks to deliver service.

Telefónica Spain

Overseas, broadband expansion is headed in another direction — expansion of fiber-to-the-home service, with little interest in investing significant sums on furthering old technology copper wire based DSL and fixed wireless services. The expansion is moving so quickly, Verizon made certain to sign long-term contracts with optical fiber suppliers like Corning in 2017 to guarantee they will not be affected by expected shortages in optical fiber some providers are already starting to experience.

Virtually everywhere in developed countries (except the United States), fiber broadband is quickly crowding out other technologies, despite the significant cost of replacing copper networks with new optical fiber cables. If a provider is brave enough to discount investor demand for quick returns and staying away from big budget upgrade efforts, the rewards include happier customers and a clear path to increased revenue and business success.

Not every Wall Street bank is reluctant to support fiber upgrades. Credit Suisse sees a need for optical fiber today, not tomorrow among incumbent phone and cable companies.

“The cost of building fiber is less than the cost of not building fiber,” the bank advised its clients. The reason is protecting market share and revenue. Phone companies that refuse to upgrade or move at a snail’s pace to improve their broadband product (typically DSL offering 2-12 Mbps) have lost significant market share, and those losses are accelerating. Ditching copper also saves companies millions in maintenance and repair costs.

Canada’s Telus is a case in point. Its CEO, Darren Entwistle, reports Telus’ effort to expand fiber optics across its western Canada service area is already paying off.

“We see churn rates on fiber that are 25% lower than copper,” Entwistle said. “35% lower in high-speed internet access, and 15% lower on TV — 25% lower on average. We’re seeing a reduction in repair volumes to the tune of 40%. We’re seeing a nice improvement in revenue per home of close to 10%.”

Telus promotes its fiber to the home initiative in western Canada as a boost to medical care, education, the economy, and the Canadian communities it serves. (1:31)

Telus’ chief competitor is Shaw Communications, western Canada’s largest cable company. Fiber optics allows Telus to vastly expand internet speeds and reliability, an improvement over distance sensitive DSL. Shaw Cable has boosted its own broadband speeds and offers product bundles that have been largely responsible for Telus’ lost customers, until its fiber network was switched on.

In economically challenged regions, fiber optic expansion is also growing, despite the cost. In Spain, Telefónica already provides service to 20 million Spaniards, roughly 70% of the country, and plans to continue reaching an additional two million homes and businesses a year until the country is completely wired with optical fiber. In Brazil, seven million customers will have access to fiber to the home service this year, expanding to ten million by 2020.

Verizon and AT&T regularly ring alarm bells in Congress that China is outpacing the United States in 5G wireless development, but are strangely silent about China’s vast and fast expansion into fiber optic broadband that companies like Verizon stopped significantly expanding almost a decade ago. China already has 328 million homes and businesses wired for fiber and added another five million homes in the month of June alone. AT&T will take a year to bring the same number of its own customers to its fiber to the home network.

The three countries that are most closely aligned with the mentality of most U.S. providers — the United Kingdom, Australia, and Germany — are changing their collective minds about past arguments that fiber to the home service is too costly and isn’t necessary.

The government of Martin Turnbull’s cost concerns forced a modification of the ambitious proposal by the previous government to deploy fiber to the home service to most homes and businesses in the country. That decision to spend less is coming back to haunt the country after Anne Hurley, a former chief executive of the Communications Alliance involved in the National Broadband Network (NBN), admitted the cheaper NBN will face an expensive, large-scale replacement within a decade.

ABC Australia reports on findings that the country’s slimmed-down National Broadband Network is inadequate, and parts will have to be scrapped within 5-10 years (1:37)

Turnbull’s government advocated for less expensive fiber to the neighborhood technology that would still rely on a significant amount of copper wiring installed decades ago. The result, according to figures provided to a Senate committee, found only a quarter of Australians will be able to get 100 Mbps service from the NBN, with most getting top speeds between 25-50 Mbps.

Despite claims of technical advancements in DSL technology which have claimed dramatic speed improvements, Hurley was unimpressed with performance tests in the field and declared large swaths of the remaining copper network will have to be ripped up and replaced with optical fiber in just 5-10 years.

“If you look around the world other nations are not embracing fiber-to-the-[neighborhood] and copper … so yes, it’s all going to have to go and have to be replaced,” she said.

In the United Kingdom, austerity measures from a Conservative government and a reluctant phone company proved ruinous to the government’s promise to deliver “superfast broadband” (at least 24 Mbps) over a fiber to the neighborhood network critics called inadequate from the moment it was switched on in 2012. The government had no interest in financing a fiber to the home network across the UK, and BT Openreach saw little upside from spending billions upgrading the nation’s phone lines it now was responsible for maintaining as a spun-off entity from BT. In 2015, BT Openreach’s chief technology officer called fiber to the home service in Britain “impossible” and too expensive.

Two years later, while the rest of Europe was accelerating deployment of fiber to the home service, the government was embarrassed to report its broadband initiative was a flop in comparison, and broke a key promise made in 2012 that the UK would have the fastest broadband in Europe by 2015. Instead, the UK has dropped in global speed rankings, and is now in mediocre 35th place, behind the United States and over a dozen poorer members of the EU.

What was “impossible” two years ago is now essential today. The latest government commitment is to promote optical fiber broadband using a mix of targeted direct funding, “incentives” for private companies to wire fiber without the government’s help, and a voucher program defraying costs for enterprising villages and communities that develop their own innovative broadband enhancements. The best the government is willing to promise is that by 2033 — 15 years from now — every home in the UK will have fiber broadband.

Deutsche Telekom echoed BT Openreach with claims it was impossible to deliver fiber optic broadband throughout an entire country.

Deutsche Telekom’s dependence on broadband-enhancements-on-the-cheap — namely speed improvements by using vectoring and bonded DSL are increasingly unpopular for offering too little, too late in the country. Deutsche Telekom applauded itself for supplying more than 2.5 million new households with VDSL service in 2017, bringing the total number served by copper wire DSL in Germany to around 30 million. The company, which handles landline, broadband and wireless phone services, is slowly being dragged into fiber broadband expansion, but on a much smaller scale.

In March, Telekom announced a fiber to the home project in north-east Germany’s Western Pomerania/Rügen district for 40,000 homes and businesses. The network will offer speeds up to 1 Gbps. In July, Telekom was back with another announcement it was building a fiber optic network for Stuttgart and five surrounding districts Böblingen, Esslingen, Göppingen, Ludwigsburg, and Rems-Murr, encompassing 179 cities and municipalities. But most of the work will focus on wiring business parks. Residents will have a 50% chance of getting fiber to the home service by 2025, with the rest by 2030.

In contrast, the chances of getting fiber optic broadband in the U.S. is largely dependent on which provider(s) offer service. In the northeast, Verizon and Altice/Cablevision will go head to head competing with all-fiber networks. Customers serviced by AT&T also have a good chance of getting fiber to the home service… eventually, if they live in an urban or suburban community. Overbuilders and community broadband networks generally offer fiber service as an alternative to incumbent phone and cable companies, but many consumers don’t know about these under-advertised competitors. The chances for fiber optic service are much lower if you live in an area served by a legacy independent phone company like Frontier, Consolidated, Windstream, or CenturyLink. Their cable competitors face little pressure to rush upgrades to compete with companies that still sell DSL service offering speeds below 6 Mbps.

CAF funding from the FCC offers some rural areas a practical path to upgrades with the help of public funding, but with limited funds, a significant amount will be spent on yesterday’s technology. In just a few short years, residents will be faced with a choice of costly upgrades or a dramatic increase in the number of underserved Americans stuck with inadequate broadband. Policymakers should not repeat the costly mistakes of the United Kingdom and Australia, which resulted in penny wise-pound foolish decisions that will cost taxpayers significant sums and further delay necessary upgrades for the 21st century digital economy. The time for fiber upgrades is now, not in the distant future.

Criminals are supposedly having a field day robbing cell phone stores in Canada after regulators ordered all cell phones to be sold unlocked, allowing customers to bring their devices to other carriers.

“There have been multiple instances of armed robberies at our stores targeting unlocked, new devices,” Bell Canada complained in a letter to the Canadian Radio-television and Telecommunications Commission (CRTC). “We believe this trend is attributable to the availability of unlocked devices [that are] more desirable to fraudsters and thieves.”

Because Canada’s three major carrier-cell phone marketplace is seen as less competitive and more expensive than the United States, the CRTC has tried to keep wireless service costs under control by regulating some of the practices of the barely competitive Canadian market. One such initiative is the ban on charging unlock fees on devices, which carriers used to deter customers from changing providers. As of last December, carriers could no longer collect an average of $50 to unlock each device, and new devices had to be sold to customers in an unlocked state, allowing them to be used on any compatible wireless provider’s network.

Rogers, which runs Canada’s largest cable operator and has a major market share of Canada’s wireless market, claims the unintended consequence of the CRTC’s unlock policy is a 100% increase in cell phone thievery during the last six months the policy has been in effect. Rogers reports thieves are stealing brand new cell phones in the mail or off a customer’s front step after the shipper drops the package off. Brazen armed robberies of cell phone stores have been more common in the United States, but providers claim criminal gangs are now taking their business north of the border, holding up stores and running off with dozens of valuable phones.

Both Bell and Rogers warned the CRTC last year thievery would be the likely result of providing unlocked phones. Consumer groups claim both providers have a vested interest complaining about the new unlock policies. In 2016, Canadian telecom companies made $37.7 million from fees related to unlocking smartphones. That was a 75 percent increase in fee revenue since 2014.

Canadian consumers called unlock charges “ransom fees,” and were particularly upset paying fees after they paid off the device.

“You should be able to unlock it [for free] at the very least once you’ve paid off the device. You own it,” John Lawford, executive director with the Public Interest Advocacy Centre in Ottawa told the CBC.

Lawford calls unlock fees an intended consequence of the industry’s own policies. Cell phone companies sell devices manufacturers have to lock at the behest of carriers, and then consumers face fees paid to the same carriers to undo the lock.

Canada’s providers often point to examples of armed robberies and truck hijacking south of the Canadian border as a reason to be concerned about employee and customer safety. In the view of some, an unlocked smartphone worth more than $500 is an invitation to steal.

Bell told regulators things are certain to get worse in Canada.

“It appears that illegal activity may have shifted from the U.S. to Canada as some [American] carriers have begun to lock devices,” Bell officials told the CRTC.

Bell was referring to Verizon’s unilateral announcement it began relocking smartphones in February, despite its agreement not to as part of an acquisition of 700 MHz spectrum in 2008. That prime spectrum came with strings attached, including a requirement not to disable or restrict devices that use the spectrum, something locked phones do. Verizon previously tested the waters on reintroducing locked cell phones during the second term of the Obama Administration, but the idea met immediate resistance from FCC Chairman Thomas Wheeler.

In 2018, Verizon found a much more receptive audience from the Republican-dominated FCC under Chairman Ajit Pai, and has gradually returned to locking down devices on Verizon’s network. Last spring, Verizon began locking all smartphones sent to stores, to be unlocked after purchase. Verizon argued this would deter armed gangs from hijacking deliveries or raiding stores to steal phones by the dozens, to be resold to the eager black market.

After meeting little resistance, Verizon announced it would start locking phones for an arbitrary amount of time after purchase, defined in terms of “months, not years.”

If thieves obtain a stolen, locked phone, it cannot generally be activated by the customer unless taken to an authorized retailer. This theoretically leaves thieves stuck with worthless phones, which is why Canadian carriers claim the country’s unlocked phone policy will draw American thieves north. But critics suspect financial motives hold more sway. In addition to charging lucrative fees for unlocking phones, customers unable to take their device with them to a new carrier can effectively deter a provider change, especially for family accounts where multiple devices would need to be moved.

Others claim locking phones is not the best way to deter thieves, because an unscrupulous Verizon employee or reseller can still unlock them for thieves.

The wireless industry already claims to have a voluntary, industry-led initiative to dramatically reduce theft — a national database of stolen/lost phones. Under this system, a would-be customer is denied activation if their device’s unique ID appears on a list of stolen or lost phones.

CBC Calgary reports Canadians no longer face unlock fees on their smartphones and other wireless devices. (3:55)

Among the major wireless companies with spectrum holdings worth billions, few would suspect that the fifth largest (behind Sprint, AT&T, Verizon, and T-Mobile) is the satellite television company Dish Networks.

After spending nearly $20 billion over the last ten years acquiring nearly 95 MHz of extremely valuable low and mid-band spectrum in markets across the United States, Dish is the largest wireless company that isn’t actually providing wireless service. Critics have questioned whether Dish co-founder Charlie Ergen was ever really interested in getting into the wireless business when he could make an even bigger killing warehousing spectrum until it grows in value and can be profitably sold to someone else. One Wall Street analyst thinks there is a strong case for exactly that. Cowen and Company estimates Dish’s holdings are now worth $30.2 billion — a $10 billion profit possible from keeping spectrum off the market until a buyer is willing to make an offer Dish cannot refuse.

Unfortunately for Ergen, spectrum is public property and ultimate ownership rights can never be sold or transferred. Instead, the FCC licenses companies to use the public airwaves, and has provisions to take them back if a company does not put that spectrum to good use. For Dish Networks, the first important deadline is March 2020, by which time the FCC expects Dish to achieve at least 70% market coverage of its 700 MHz “E-Block” and 2000-2020/2180-2200 MHz AWS-4 licenses.

Dish’s “E-Block” spectrum was formerly known as UHF channel 56. Dish has already begun testing the next-generation TV standard ATSC 3.0 on its E-Block spectrum in Dallas, as part of a joint venture with TV station owners Sinclair, Nexstar, and Univision. Dish proposed to use this spectrum, which covers 95% of the United States, as a potential tool for broadcasters. Among the services Dish could offer are broadcast data applications made possible with the ATSC 3.0 standard.

Because time and money is on the line, Dish needs to either build its network quickly or sell/lease its spectrum to other companies before facing possible spectrum forfeiture in less than two years. Analysts say one of the cheapest and easiest ways of placating the FCC is to deploy a modest, narrowband wireless network designed for machine-to-machine communications. These networks rely on short bursts of data to communicate information. Possible applications include exchanging irrigation and crop data collected from wireless sensors and various remote weather and climate measurement tools.

Coincidentally, that is exactly the kind of network Ergen initially envisions, largely operating on the sparsely used AWS bands. Officially called “NB-IoT” in wireless industry parlance, the ‘narrowband Internet of Things’ network would be the first chapter of Dish’s wireless story. It’s a network done on the cheap — constructed with a relatively low investment of $500 million to $1 billion through 2020, adequate enough to keep the FCC off Dish’s back.

Ergen reports the radios have been ordered and in a sign of serious intent, Dish has now signed master lease agreements with cell tower companies that will allow Dish to place its transmission equipment on tens of thousands of cell towers around the country. The company has also hired experts in tower permitting and network design and planning. Those contracts are an important indicator for some skeptics on Wall Street who believed Ergen would not show seriousness of intent until he signed paid, binding commitments to begin network buildout.

Ergen would disagree that Dish has been foot-dragging its wireless network deployment, despite a decade of accumulating wireless spectrum that has gone unused.

“It’s all about timing; too early you are roadkill, if you get it just right you have a chance,” Ergen said. “We missed the 4G shift because of the regulatory reasons. The next big paradigm shift is 5G.”

Ergen

Unfortunately for Ergen, he will be late to that paradigm shift, admitting his dream of a national 5G network isn’t possible right now.

“We’re […] going to spend at least $10 billion or more on a 5G network,” Ergen said, while also admitting, “we don’t have that kind of capital on our balance sheet today.”

Ergen promised that sometime in the future, Dish will begin a “second phase” that will “build a complete 5G network.” But Ergen’s vision of 5G is somewhat different from Verizon and AT&T, which are focused on the consumer and business voice and data markets. Ergen envisions a robust 5G network designed to support IoT applications like smart cities, artificial intelligence, and autonomous vehicles, and does not seem interested launching a fifth national cell provider.

Ergen quit in December 2017 as CEO of Dish’s aging satellite TV business to refocus on Dish’s mobile future, and to recast the venture as a glorified startup, much like his early days in the home satellite television business where he got into the business manufacturing 10-foot C-band satellite dishes for consumers and then sold the programming to watch on those dishes. From money earned in that business, Ergen launched Dish Networks, which relies on today’s familiar small satellite dishes and competes with DirecTV.

Ergen’s satellite TV venture only had to compete with one other satellite provider. His wireless network will have to compete with at least four established national wireless companies, plus emerging competition from the cable industry and regional cellular providers. Ergen tried to turn that obvious business challenge into an opportunity:

“We have two disadvantages; We don’t [have many] customers and we are not as knowledgeable as other people in the business, but we don’t have the legacy of 2G, 3G, 4G networks,” Ergen said. “We have a clean sheet of paper with 5G. It reminds me of 1990 when we decided to reinvent ourselves from the big dish business to small dish. It took five years to design and build that system with not one penny of revenue, and we obsoleted the business we were in. When we got into satellites, we didn’t know anything about it, but neither did anyone else. It is the same with 5G/IoT. We are not the world’s experts, but neither is anyone else.”

What Ergen lacks in experience he makes up for in enthusiasm, laying out plans for Dish’s wireless future. By the time he activates 5G service, Dish expects to use its combined 95 MHz of spectrum in the 600 MHz and 2 GHz range for that network. That will take until at least July 2020, because many of the 600 MHz frequencies he needs are still occupied by UHF television stations that are in the process of migrating to a more compact UHF band.

Dish has spectrum holdings that reach almost every corner in the U.S.

Ergen may also consider acquiring additional millimeter wave spectrum if he deploys small cell technology. He has even decided to keep small cell and larger traditional “macrocells” found on traditional cell towers on different frequencies, claiming sharing the frequencies would create interference issues.

Ergen also hopes to convince the FCC to repurpose little-known Multichannel Video Distribution and Data Service (MVDDS) spectrum located between 12.2-12.7 GHz for 5G wireless applications. That solid block of 500 MHz of spectrum could be an important asset to power small cell 5G networks, because it can support faster speeds than the typical smaller blocks of frequencies most companies control. MVDDS also lacks a significant constituency to protect it, having been woefully underutilized in the United States. Only tiny Cibola Wireless, an ISP in Albuquerque, N.M., licenses MVDDS technology for its wireless internet service, selling Albuquerque residents up to 50 Mbps speed for $79.99 a month. Users claim the service does not suffer the latency problems of traditional satellite internet access, but can still slow down if too many users are online at the same time.

Back in 2010, MVDDS technology was seen as a potential competitor to companies like Dish and DirecTV, as well as satellite internet providers which share similar spectrum. Like satellite internet, MVDDS can transmit and receive data over a small dish. But instead of pointing it to a satellite 44,000 miles away, MVDDS systems target a ground-based transmission tower much closer nearby. The technology never attracted much attention, and will now likely be displaced by 5G in the United States, although it has done modestly better abroad, serving a limited customer base in the United Arab Emirates, Ireland, France, Vietnam, Greenland and Serbia.

AT&T is rolling out mobile 5G service for its wireless smartphone and tablet customers in a dozen U.S. cities by year’s end, starting in parts of Atlanta, Ga., and portions of Dallas and Waco, Tex.

“After significantly contributing to the first phase of 5G standards, conducting multi-city trials, and literally transforming our network for the future, we’re planning to be the first carrier to deliver standards-based mobile 5G – and do it much sooner than most people thought possible,” said Igal Elbaz, senior vice president, wireless network architecture and design. “Our mobile 5G firsts will put our customers in the middle of it all.”

AT&T’s mobile 5G will work differently from the fixed wireless home broadband service Verizon is launching this year using small small cell neighborhood antennas. But like Verizon, AT&T is taking a gradual, incremental approach to the next generation of wireless technology.

In 2017, AT&T announced what it calls “5G Evolution” service in almost two dozen cities, although this branding was derided as “fake 5G” in the tech press because, in reality, it is just an improvement of today’s widely deployed 4G LTE service. Similar technology is also in place at T-Mobile. In the fall of 2017, AT&T introduced 4G LTE-Licensed Assisted Access (LTE-LAA) technology in Indianapolis and parts of Chicago, Los Angeles and San Francisco. This network lays the foundation to offer gigabit speed wireless service, and is especially useful in areas where AT&T’s spectrum holdings are tight.

AT&T’s initial 5G rollout will serve parts of:
A – Atlanta, Ga.
B – Waco, Tex.
C – Dallas, Tex.

AT&T is preparing its existing wireless network to permit gradual migration to the completed 5G wireless standard over both existing and new spectrum.

This year, AT&T plans to launch some 5G service using millimeter wave spectrum, which is very line-of-sight and offers a more limited service area. But the technology will support very fast wireless speeds and offer plenty of bandwidth. AT&T could deploy this technology initially in dense population areas and places like stadiums, malls, and convention centers.

“Ultimately, we expect to reach theoretical peak speeds of multiple gigabits per second on devices through mobile 5G,” AT&T wrote in a press release. “While speed is important, we also expect to see much lower latency rates. With higher speeds and lower latency rates, our mobile 5G network will eventually unlock a number of new, exciting experiences for our customers.”

If past precedent means anything, AT&T will likely only initially offer 5G service in selected parts of each city. It needn’t hurry, because equipment designed to work with the new spectrum isn’t expected to become widely available until 2019. A gradual transition will also please shareholders by keeping network upgrade costs predictable over the next 3-5 years.

AT&T isn’t expected to use 5G technology anytime soon as part of its taxpayer-funded, rural wireless broadband deployment. AT&T currently uses its 4G LTE technology to power its fixed wireless rural broadband service. AT&T claims this service was designed to assure download speeds of at least 10 Mbps, although customers using it report speeds are often lower, although sometimes higher. AT&T does not offer and network performance guarantees, stating, “service performance may be affected by your proximity to a cell site, the capacity of the cell site, the number of other users connected to the same cell site, the surrounding terrain, radio frequency interference, applicable network management practices, and the applications you use.” That will also be true of AT&T’s forthcoming 5G network.