Tuesday, January 31, 2017

AT&T to Test AirGig on Real-World Electrical Utility Plant

AT&T says it is in advanced discussions with power companies and others to trial Project AirGig in at least two locations by this fall. One location will be in the United States with others to be determined in the coming months.

AirGig is different from prior “broadband over powerline” technologies as it does not transmit data on the electrical media, but instead moves external to the physical media, along or near to the actual wire.

The attractions are obvious. Power lines go everywhere businesses work, and people live. If AirGig works as well as AT&T hopes, AirGig might allow new options for data service access plant, and also potentially serve as backhaul for small cells.

Monday, January 30, 2017

Millimeter Wave Bugs Can be Features

Oddly enough, some might argue, there are times when a “bug becomes a feature.” Some argue millimeter wave signal propagation is a “bug” that actually provides a valuable feature.

Millimeter waves (3 GHz to 300 GHz radio waves) have propagation issues that limit the distance such signals can travel, for example. That is why, in the analog era, before low-cost signal processing, such frequencies were not so extensively used to support communications networks, and when used, were restricted to point-to-point backhaul applications, where high-cost equipment could support trunking functions (per-instance, per-user, per-customer or per-passing economics always are much better wherever such functions are shared in the network).

But Siklu argues that propagation limitations actually are a feature, not a bug. Such propagation limits also mean that millimeter wave signals are less prone to interference than signals that propagate much farther.

Less interference means base stations and radios can be deployed close together, according to Yigal Leiba, Siklu CTO.

In terms of implications for bandwidth, huge new amounts of millimeter wave spectrum are expected, as part of the 5G evolution. The U.S. Federal Communications Commission, for example, has decided to release about 11 GHz for flexible, mobile and fixed use wireless broadband, including 3.85 GHz of licensed spectrum and 7 GHz of unlicensed spectrum.

The rules create a new Upper Microwave Flexible Use service in the 28 GHz (27.5-28.35 GHz), 37 GHz (37-38.6 GHz), and 39 GHz (38.6-40 GHz) bands, and a new unlicensed band at 64- 71 GHz.

The U.S. Federal Communications Commission, for example, already is on a path to open up 14 GHz of additional spectrum (an order of magnitude more access spectrum than currently is available for all mobile and Wi-Fi use.

The Commission also adopted a Further Notice of Proposed Rulemaking to apply the flexible use service and technical rules to another 18 GHz of spectrum encompassing eight additional high-frequency bands.

Just in the range from 57 GHz to 71 GHz, that millimeter spectrum represents 100 times more spectrum than typically is available to support mobile access networks, says Leiba.

That new spectrum, with other shared and unlicensed new spectrum, will allow internet access providers to provide the 1000-fold increase in internet access capacity envisioned for 5G. Small cells, better radios and modulation methods also will help.

Will India Mobile Operator Market Share be Rearranged?

If Vodafone successfully completes the proposed merger of its India mobile business with Idea Cellular, that new firm would vault to the top of the market share rankings, displacing Airtel, the current leader.



India "Free Data" Rules Illustrate How Hard it is to "Solve" Internet Access Problem

Of the several clear problems confronting nations, governments and service providers, lack of facilities and the price of service are clear issues. Frustratingly, those are, to some extent, mutually exclusive goals. Low prices stimulate usage, but low prices also discourage investment to provide service.

Tactically, internet service provider ability to offer low cost or free service can be challenging, as when it is unlawful to do so, or lawful under only some conditions.

Though the Telecommunications Regulatory Authority of India (TRAI) bars zero rating (free service), it apparently does not believe that the principle of offering consumers free data should be completely barred, only “who” should be allowed to do so.

Subsidizing rural users from universal service funds, for example, is permissible. Subsidized free access by mobile operators or app providers (toll-free internet access or free access to bundles of apps) is not permissible, under network neutrality rules.  

If it seems confusing, that is because the policies on how consumers receive subsidies is confusing. Promotional campaigns have not yet been seen as impermissible, for example.

But free” is a powerful price point, as shown by mobile account additions in India in October 2016, where Reliance Jio is grabbling most of the account growth in the whole market. Nor, in principle, does TRAI object to free services paid for by universal service funds.


Friday, January 27, 2017

In 2020, 8 Billion IoT-Capable Devices Could be Shipped

By 2020, nearly two billion devices with mobile connections capability, 2.7 billion devices that can connect to one or another low-power communications network and 3.3 billion devices with Wi-Fi capability will be shipped, or more than eight billion total devices in one year, according to IHS Markit.

source: IHS Markit

TV White Space Spectrum Market is Too Small to Bother With: Just Go Unlicensed

A study by Marketsandmarkets predicts the global TV white space spectrum market was worth at US$1.2 Million in 2015 and is expected to reach US$53.1 Million by 2022. Two observations: it is hard to value a commodity that is not sold commercially, as it is hard to value Wi-Fi spectrum that likewise is not sold.

In fact, one might argue that the small global market for TV white space spectrum is so small it should not be bothered with: regulators should just let people use it, as the costs of selling it will exceed the possible proceeds.

End of Traditional Spectrum Economics?

“Are we finally nearing the end of spectrum allocation as we know it?” Rethink Wireless rhetorically asks. “One of the most-anticipated U.S. auctions ever has flopped, while the guardian of licensed spectrum technology, Qualcomm, has spawned a technology, MulteFire, which could allow organizations with no spectrum of their own at all to build cellular networks.”

Some with long experience in communications continue to insist that, without exclusive access or ability to maintain quality of service, large networks will not be built.

“In a research note, MoffettNathanson analysts posed the question which is concerning many investors in mobile operators around the globe: is spectrum an asset of real value any more?”

That obviously overplays the present trend. Spectrum auction prices have been lower for all of the recent big auctions (India, United States). So one existing trend is a lower valuation of new spectrum, by network operators.

One might reasonably ask what that could mean for public company valuation and capital spending patterns. How extensive the trend might become is unclear, but it seems already the case that potential bidders are aware they have other options to acquire new capacity, ranging from use of small cells to better radios, and with huge allocations of new spectrum--licensed and unlicensed--on the horizon.

“It is increasingly possible for service providers to build robust public wireless networks without licensed spectrum,” Rethink Wireless argues. Again, though possible, it remains unclear how prevalent that trend might ultimately become, the more-likely outcome being hybrid networks that blend licensed and unlicensed spectrum.

Also, Qualcomm’s MulteFire, allows service providers to run LTE in unlicensed spectrum, mainly 5 GHz, without the need for an anchor network in a licensed band (as other 5 GHz LTE approaches, like LTE-Unlicensed and LTE-LAA, do).

Big changes are coming.

Thailand to Release 380 MHz of New Spectrum for Mobile

The Thai telecom regulator--National Broadcasting and Telecommunications Commission--plans to auction 380 megaHertz of new spectrum for mobile use by 2020.

The resources include 180 MHz of downlink and uplink bandwidth in the 2600 MHz spectrum, to be released by 2017. An additional 90 MHz in the 1800 MHz band and 20 MHz in the 850 MHz region will be released in 2018.

Finally, 90 MHz in the 700 MHz band will be opened up for use in 2020, according to plan.

Thursday, January 26, 2017

Comcast to Launch Mobile Service this Year

“Another priority in 2017 is to launch our wireless product,” said Comcast CEO Brian Roberts. “We plan to include wireless in our multi-product bundles in a way that is designed to add value to our customers (and) improve retention.”

That clearly means a quadruple-play or other bundles that provide value by providing cost savings. Comcast is expected to rely on a Wi-Fi-first model, which is why it intends to heavily concentrate marketing on its own customer base, especially those customers who buy internet access service from Comcast, as that will allow Comcast to leverage its homespot capabilities.

Separately, Neil Smit, Comcast SVP, said that Comcast has a role in 5G as well. “We think that it's going to need economical space, power, and backhaul,” he said. “We have, call it, 150,000 miles of fiber across 650,000 miles of total plant and we think that we are well-positioned to participate in the 5G rollout, no matter how it happens.”

Smit noted that internet account usage showed a median of 88 gigabits per month in the first quarter, up 55 percent from 57 gigabits during the same period of 2015.

Would Verizon Contemplate Selling All its Fixed Network Assets?

Many have speculated over the years that Verizon would--if it could--exit the fixed network business entirely. In recent years it has sold many of its rural and non-contiguous assets. On the other hand, Verizon recently has argued that coming 5G networks will require a much-denser network of backhaul facilities that are best exemplified by an extensive fiber-deep network.

So it seems unlikely Verizon would contemplate a complete exit from the fixed network business. But which assets might it prefer to use? That is among the questions raised by the rumors about a purchase of Charter Communications.

Some of us think merger approval would be difficult, at best. The rumored Time Warner merger with Charter Communications likely would be viewed as a horizontal merger, not a vertical merger, as AT&T proposes with its acquisition of Time Warner. That is going to be a tougher sell, to regulatory and antitrust authorities, without huge asset dispositions.

Charter passes 48 million homes. Verizon passes some 27 million homes, with significant overlap in New York, Maine, Massachusetts, New Jersey, Virginia, Pennsylvania, Rhode Island, Connecticut and Delaware. In other words, asset dispositions across most of the Verizon footprint would be expected.

Assume there are a total of about 135.7 million total U.S. homes. Charter passes about 35 percent of U.S. homes, arguably at the limit of traditional antitrust thinking (roughly 30 percent has been a rule of thumb for any single provider). Assuming only half the Verizon homes overlap, that still leads to a Verizon-Charter homes passed count somewhere in the neighborhood of 61.5 million homes, or about 45 percent of total.

It is hard to see antitrust authorities approving that large a footprint. Or, of course, Verizon might contemplate spinning off its entire telco footprint, relying on the hybrid fiber coax network for fixed network access. That would be a huge change for Verizon, but is possible, in principle.

Among other practical issues, Charter arguably has customer satisfaction performance far below what Verizon touts. AT&T, when it purchased Tele-Communications Inc., at that time the largest U.S. cable TV operator, found the facilities far less robust than it anticipated. Verizon likely would encounter some of those same issues.

On the other hand, Verizon could, if it were to sell its fixed networks, also be able to deal with its union labor issues, as Charter is non-union.

So does Verizon sell its telco assets, and keep the Charter assets, or sell the cable assets, which are said to be the reason Verizon wants Charter in the first place?

In other words, does Verizon contemplate (and who would buy) divestitures of most of its texlo fixed network, and replace it with the Charter hybrid fiber coax network? And will telcos once again arguably be shown to be the “greater fools,” at the hands of cable companies?

Wednesday, January 25, 2017

Will Fixed Wireless Allow Internet Access Cord Cutting?

“Cord cutting” has affected--and is affecting--several fixed network communications products, including voice (fixed to mobile), messaging (email to text messaging) and video entertainment (linear subscription TV to on-demand TV to mobile TV).

Some believe the next big wave of cord cutting will involve internet access. To be sure, mobile internet access, in most markets, already is the most-common form of internet access. But in developed markets, it is hard to beat the economics (value for price) of fixed internet access.

That all could change as the 5G era begins, primarily because, for the first time, mobile operators will have the ability to substantially match fixed network speeds and prices. Today, in the U.S. market, mobile bandwidth can cost up to $10 per gigabyte consumed. Fixed network costs are at least an order of magnitude lower ($1 per GB or less).

Major U.S. telcos (not cable TV companies) will be the primary potential beneficiaries. Today, cable TV companies can upgrade to gigabit speeds much more affordably than telcos can, as gigabit speeds normally require upgrades to full fiber to the home networks (at least in the single-family residence portion of the market).

That is not a good business model in many, if not most cases, given the expected adoption rates. Verizon, for example, where it has operated Fios networks for some years, has gotten about 40 percent penetration. Cable TV companies tend to get 60 percent adoption rates.

That suggests an upward share limit for a well-run, well-capitalized telco competitor in a fixed network market. Stranded assets as high as 60 percent are among the chief problems as the “cost per customer” is more than double the “cost per passing.” The business model is even more challenging for overbuilders who are the number-three providers in a market.  

Many believe that the ability to deploy 5G in a fixed wireless mode, with speeds between 1 Gbps and 10 Gbps, could be a game-changing development. Use of new millimeter wave spectrum, unlicensed spectrum, better radios, plus small cell architectures, could radically change bandwidth economics in the internet access business.

Among the issues are the deployment models. Millimeter wave approaches might work best for urban areas, and possibly some multi-dwelling units. But some believe rural areas are where business models might be best.


source: Arris

Tuesday, January 24, 2017

Verizon Seems to Target U.S. for IoT

Sometimes a service provider’s choice of an air interface can indicate which markets it intends to target. Verizon, for example, is touting its  Cat Mi LTE platform for internet of things connectivity. In Europe, narrowband LTE is the preferred mobile IoT platform, where it or Sigfox and LoRa are not seen as the solution.

Specs
LTE Cat 1
LTE Cat 0
LTE Cat M1
(eMTC)
LTE Cat NB1
(NB-IoT)
3GPP Release
Release 8
Release 12
Release 13
Release 13
Downlink Peak Rate
10 Mbps
1 Mbps
1 Mbps
250 kbps
Uplink Peak Rate
5 Mbps
1 Mbps
1 Mbps
250 kbps (multi-tone)
20 kbps (single-tone)
Number of Antennas
2
1
1
1
Duplex Mode
Full Duplex
Full or Half Duplex
Full or Half Duplex
Half Duplex
Device Receive Bandwidth
1.08 - 18 MHz
1.08 - 18 MHz
1.08 MHz
180 kHz
Receiver Chains
2 (MIMO)
1 (SISO)
1 (SISO)
1 (SISO)
Device Transmit Power
23 dBm
23 dBm
20 / 23 dBm
20 / 23 dBm

So one might suggest that Verizon plans to concentrate its IoT efforts in the United States, not international markets. That would be consistent with Verizon’s existing strategy, which is U.S.-centric, compared to AT&T, which already has entered the Mexico mobile market, and many South American video markets.

In U.S. Mobile Market, Time Spent with Wi-Fi Might Actually be Decreasing

It is a common understanding that most of the time mobile users spend connecting to the internet uses a Wi-Fi connection, rather than the mobile network. That is only partly correct.

Mobile offload to Wi-Fi exceeded mobile network traffic for the first time in 2015, according to Cisco, when 51 percent of total mobile data traffic was offloaded onto the fixed network through Wi-Fi or femtocell. But that is a far cry from the claim--often made--that Wi-Fi represents about 80 percent of mobile connections. What is true is that perhaps 80 percent of data is consumed that way, but not sessions.

In fact, as mobile data charges get cheaper, the amount of time users spend connecting over the mobile network tends to increase, while connections (sessions) using Wi-Fi tend to decrease. In the U.S. mobile market, “time spent” connecting to Wi-Fi generally is decreasing, compared to mobile usage.

Up to this point, Wi-Fi likely has been used for heavier sessions involving consumption of entertainment video. While mobile sessions were used for other purposes, such as social interactions. With the advent of new tariffs that do not levy usage for video entertainment consumption, even that use of Wi-Fi for video consumption might change.



Monday, January 23, 2017

Video Shift to Mobile: Other Key Apps Already Have Done So

At least some mobile service providers hope subscription-based entertainment video will be a big deal, representing the latest evolution of entertainment video from movie theater to television, to cable TV (plus satellite and telco delivery) to Netflix and other over the top services, and then possibly to mobile consumption of OTT subscription video.

There are good reasons for believing that migration will happen. The shift from desktop to mobile happened in search, internet browsing, social networking and now video.  Smartphones and other mobile devices are--or are becoming--the preferred devices for many of our day-to-day activities.

Chart 5


The most significant change in the status of competition in the market for the delivery of video services has been the introduction of Sling TV by Dish Network and DIrecTV Now by AT&T, the U.S. Federal Communications Commission says.

Others might also note the importance of the AT&T acquisition of DirecTV, which vastly expanded the addressable universe of homes AT&T could reach from less than 21 percent to virtually 100 percent of U.S. homes.

One sign of the robustness of competition is that profit margins, which once were as high as 40 percent, had fallen to about 10 percent in 2015, and likely are a bit lower in early 2017.

The FCC says profit margins were 15 percent in 2014 and 20 percent in 2013. That might be unappetizing in one sense, but arguably is meaningful in a context where other legacy services, whatever their profit margins, do not contribute much revenue. Video services represent a huge percentage of potential access provider revenue, if margins are not so high.

By some estimates, internet access gross margins (at least for cable operators) might be as high as 60 percent, while voice services might have profit margins near 20 percent. Telco margins likely are not that robust.

If cash flow matters--and it does--then the revenue video entertainment represents is hard to match, averaging between $80 and $110 a month, per account.

According to SNL Kagan, there were 134.2 million housing units in 2014 and 135 million housing units in 2015. The FCC therefore assumes that cable and satellite companies cover nearly every household in the country.


At the end of 2015, cable suppliers accounted for 53 percent of all subscribers, down from 53.4 percent at the end of 2014. Direct broadcast satellite (DBS) providers accounted for 33.2 percent of subscribers at the end of 2015, down slightly from 33.3 percent at the end of 2014.

Telephone companies accounted for 13.4 percent of MVPD subscribers at the end of 2015, up from 12.9 percent at the end of 2014.

Total subscribers declined in 2013, 2014 and 2015, with the suppliers losing about 1.1 million video subscribers in 2015.

But total video revenue increased from $112.7 billion in 2014 to $115.6 billion in 2015, partly because of rate increases and partly because of subscribers upgrading to higher levels of service.

Is Sora an "iPhone Moment?"

Sora is OpenAI’s new cutting-edge and possibly disruptive AI model that can generate realistic videos based on textual descriptions.  Perhap...