5G Fixed Wireless Works for Internet Access Business Case, Nokia Argues

Using 5G (and millimeter wave spectrum) for consumer internet access, a service provider might well have a business case for using fixed wireless for internet access, Nokia has found.

In a study of one European service provider, the business model for using 5G for internet access in a fixed mode works at 30 customers served per cell site, though obviously the model is sensitive to capital investment and average revenue per account.

But the model appears most sensitive to take rate, as you would expect. The business model requires a 30-percent take rate. That is a lower threshold than generally required for a facilities-based fiber-to-home network.

Average revenue per account probably needs to be at least 40€ (US$43) per month, a target most consumer service providers should be able to hit with one service.

In its analysis for one European service provider, a Nokia 5G solution using millimeter wave spectrum allows each base station to serve tens of households.

The 5G short-range fixed wireless access is expected to sustain one Gbps per household. To achieve such a high speed and longer ranges for 5G fixed wireless service in suburban and rural markets requires MIMO antenna and beam-forming, plus use of lower-frequency spectrum.

In other work, Nokia has found that 5G is much more cost effective than earlier generations of mobile networks to support entertainment video, an advantage directly related to consumer internet access demand.

source: Nokia

The business model also will be influenced by ARPU, which for some suppliers might be $130 a month, meaning the breakeven is lower than modeled by Nokia. Also, other benefits, especially the ability to share cost with the mobile 5G infrastructure, might have some impact on the business model, further lowering the payback threshold.

That appears to be Verizon’s thinking, as it envisions using one optical transport infrastructure to support both mobile and fixed access operations.

Fiber Reaches Less Than Half of U.S. Commercial Locations

U.S. business locations with 20 or more employees reached by an optical fiber connection  reached 49.6 percent in 2016, according to Vertical Systems Group. In 2004, optical connections reached only about 10 percent of such locations.

One reason so many business locations, even those with at least 20 workers, are not connected directly by optical fiber is that so many are small locations that do not provide a business case. Roughly half of all commercial buildings represent about 10 percent of total floor space, showing how small most sites are.

The vast majority of commercial buildings are relatively small. About half of buildings are 5,000 square feet in size or smaller, and nearly three-fourths are 10,000 square feet or smaller. The median building size is 5,000 square feet (i.e., half the buildings are larger than this and half are smaller), while the average size is 15,700 square feet.

Buildings over 100,000 square feet (from large high schools to hospitals to sprawling distribution centers to skyscrapers, for example) make up only about two percent of the building count but about 35 percent of the total floorspace.

Many of those smaller locations are markets for consumer-grade connections or non-fiber moderate-speed access connections, and may never be amenable to optical fiber access.

 

source: U.S. Energy Information Administration

Offices represent the single biggest category of locations, followed by warehouse and storage locations; service businesses and retail.

source: U.S. Energy Information Administration

Where Will Artificial Intelligence Appear First in Telecom?

Billing systems, self-optimizing networks, customer service and network functions virtualization are some of the areas in which artificial intelligence is likely to appear first in the core telecom business. AI also already is used by content and media firms to personalize content feeds, and some access providers who also own such assets will be using AI that way.

Over time, almost any function or process with big data characteristics will be augmented by AI, as that is a way to wring insights from masses of unstructured data.

Global revenues for cognitive and artificial intelligence (AI) systems will reach $12.5 billion in 2017, an increase of 59 percent over 2016, according to researchers at IDC.

Global spending by enterprises on cognitive and AI solutions will grow at a 54 percent compound annual growth rate (CAGR) through 2020, when revenues will be more than $46 billion.

The largest area of spending in 2017 ($4.5 billion) will be cognitive applications, which includes cognitively-enabled process and industry applications that automatically learn, discover, and make recommendations or predictions.

Spending on cognitive-related IT and business services will be more than $3.5 billion while dedicated server and storage purchase will total $1.9 billion. Each of these areas will experience strong growth throughout the forecast, led by cognitive applications with a five-year CAGR of 69.6 percent, IDC predicts.

On a geographic basis, the United States is the largest market for cognitive/AI spending with 2017 revenues totaling nearly $9.7 billion. Europe, the Middle East and Africa (EMEA) is currently the second largest region, but strong spending growth from Asia/Pacific (including a 107% CAGR in Japan) will move it ahead of EMEA by 2020.

The Cure for High Prices is High Prices

Some observers worry that the growing cable operator market share in consumer internet access means higher prices are coming, as telcos increasingly are relegated to almost-marginal roles.

In a new research note, New Street Research analyst Jonathan Chaplin says cable providers controlled 65 percent of the overall consumer internet access market at the end of 2016, possibly growing to as much as 72 percent by 2020.

Of course, the concern about higher prices assumes new challengers, or new methods of competing, cannot arise. That is likely an unwise assumption. In virtually all markets that are open to competition, high prices and high market share are invitations for competitors to attack.

“The cure for high prices is high prices,” a commodities industry adage suggests. In other words, high prices encourage more production, which leads to lower prices.

To be sure, many observers of the communications might deny that can happen with the commodity known as internet access, but history suggests the internet access commodity will respond to competitive forces in the future, as it has in the past, bringing additional supply to a market that concentrated, and also featuring “higher prices.”

In principle, several conceivable lines of attack exist. New platforms and new providers both are coming. Though some might doubt the amount of share that can be taken by new challengers on new platforms, several new platforms are coming, ranging from fleets of low earth orbit satellites, possible use of balloon fleets, unmanned aerial vehicles, fixed wireless and 5G mobile substitution.

Some of those platforms will come to market because new providers (Facebook, Google, OneWeb) are sponsors of the new platforms. In other cases, existing providers (mobile operators) will compete with fixed network providers.

What seems clear is that the era when telco fixed internet access networks based on digital subscriber line were competitive with the cable TV platform has largely past. Fiber to the home will be a choice in some cases, but the business case has gotten harder as cable has gained more share (stranded investment risks are far higher as the addressable market shrinks).

That likely means the future competition will come from new platforms and providers (mobile substitution as well as new competitors).

That might seem unlikely with the pause in Google Fiber deployments. Longer term, a big and attractive consumer market with rising prices, where one contestant has 70 percent share, will be attacked. The only issue is “by whom?”

Verizon to Deploy 8,000 to 10,000 Small Cells in Boston

Verizon CEO Lowell McAdam says it will deploy 8,000 to 10,000 small cells in Boston, as part of its One Fiber optical infrastructure in Boston, investing  $300 million over six years. That massive deployment of new cell sites explains why the One Fiber deployment of transport fiber is so important.

Those 8,000 to 10,000 cell sites all will need gigabit per second backhaul.

Separately, he says Verizon will purchase up to 12.4 million miles of optical fiber from Corning each year beginning in 2018 through 2020. The minimum purchase commitment is $1.05 billion, Verizon said.

To put those Boston small cells into perspective, there were about 300,000 cell sites total, in the U.S. market, in 2016, and about 150,000 towers.

source: SNL Kagan

Computing Models Have Business Model Implications

The Open System Interconnect model (and the incompletely compliant internet protocol) is a reference model for the way modern computing works, and illustrates why the global telecom industry faces some common problems.

For starters, OSI was developed because modern computing requires communications across disparate systems. That heterogeneity is one key foundation of modern computing. As computing no longer is monolithic, so telecom no longer is monolithic.

As modern computing assumes (in principle) that “any authorized app” can be used by “anybody who is authorized” on “any authorized device,” so the structure of telecom now makes similar assumptions.

That is far different from yesterday’s world, where “my authorized apps” can be used by “any of my authorized users” on “any of my authorized devices.”

That dramatically changes customer lock-in and therefore supplier power. So all access providers now are “less powerful” than they used to be. Look at the market capitalization of the world’s leading apps, compared to the market capitalization and multiples of telecom and access providers and you will see how “value” translates into valuation. As early as 2011, app providers had leapt to market values far higher than the largest telcos, creating what some have called valuation envy on the part of access providers.  

source: Telco 2.0

But the OSI model (and Internet Protocol) also explains why access providers have lost value in the ecosystem. By creating logical separation of computing functions, business role separation also was enabled.

By definition, it is possible for apps to be created and accessed irrespective of the ownership of access facilities or the actual applications themselves. That is the fundamental opposite of the old model, where applications and network ownership were vertically integrated.

That takes profit margin out of the access business at the same time it enables product substitution that does not require network ownership. Apps now are effectively virtualized and abstracted from access and transport functions and ownership.

Wi-Fi provided an early example of how even access has become more virtualized, allowing users to connect over networks not owned by their “service providers.” More examples of that are coming.

And though some have argued that over the top apps do not compete with telco and access provider revenue, only control of the value chain, one might now argue that is only partly true. OTT providers essentially destroy the former access provider markets (look at voice, messaging, now video).

source: Vision Mobile

source: IDATE

FCC Wants to Streamline Pole Attachment Processes

I cannot prove it is true, but I believe one reason why young digital natives often miss the business challenges of creating access network infrastructure is that it is so physical, so resistant to Moore’s Law improvement and other advantages available to creators of virtual products.

In other words, internet access facilities are construction projects. And, over a five-year period, the cost of getting a pole ready for hanging of one new cable can represent two orders of magnitude most cost than the cost of renting space on the pole for five years.

For such reasons, impediments to construction allow incumbents to delay competition. That has been an issue in access networks as long as I can remember (only a few decades, admittedly). One clear example is disputes over pole attachment rates and rights.

That is among the reasons the U.S. Federal Communications Commission wants to make the process of gaining pole attachment rights easier.

Federal Communications Commission Chairman Ajit Pai of the has proposed a Notice of Proposed Rulemaking (NPRM), Notice of Inquiry (NOI), and Request for Comment (RFC) to remove barriers to infrastructure investment by streamlining pole attachment processes.

Reforming the FCC’s pole attachment rules will make it easier, faster, and less costly to access the poles, ducts, conduits, and rights-of-way necessary for building out next-generation networks.  

The FCC chair also hopes the rulings will provide greater regulatory flexibility and expedite the process of retiring copper facilities by streamlining the regulatory process by which carriers must obtain FCC authorization to discontinue legacy services.

source: McLean Engineering

source: FCC

source: FCC