Monday, October 3, 2016

In Gigabit Era, Fixed and Mobile Networks Will Reach 10 Gbps Per Device

With gigabit Internet access speeds now shaping the context of the consumer Internet access market, it also should be acknowledged that current development work, historical precedent and competitive markets already suggest that multi-gigabit speeds up to 10 Gbps already are on the standards agenda and product roadmaps for both fixed and wireless networks.

At a time when consumers actually do not have access to applications that require a gigabit, much less 10 Gbps, that might seem an example of pure marketing hyperbole.

What might be shocking is that history suggests 10 Gbps is precisely where we should expect bandwidth to go.

Basically, since the time of dial-up access, we have seen linear increases in bandwidth that very much resemble Moore’s Law. Indeed, some even argue that Internet access bandwidth, in terms of the top marketed speeds for consumers, have increased precisely as Moore’s Law would suggest computing power grows.

What will likely come as a bigger surprise is the improvements we will see with 5G and future mobile generations, where mobile or fixed wireless speeds will reach gigabit ranges (one Gbps up to 10 Gbps) as part of the standards.



Precisely how all that bandwidth can be provided at prices regular consumers are willing to pay is the issue.

And that is why any number of development initiatives, mobile and fixed, are key. Basically, all the efforts aim to supply gigabit bandwidth on networks that are efficient enough to support consumer price points.

In that regard, much attention now is going into fixed wireless.

AT&T, for example,  is working on “AirGig,” a method for combining fixed wireless with power line transmission for communications, without building towers, laying cables or acquiring new spectrum.

All three of those attributes have the potential to dramatically lower the cost of delivering gigabit services in the access network.

AT&T’s Project AirGig has several key advantages:
  • Easier to deploy than fiber
  • Uses license-free spectrum
  • No need to deploy towers, dig trenches or connect cables
At&T expects to conduct field trials in 2017.

Combined with the dominant role of cable TV networks in the access network, and the upgrades to gigabit speeds, serious questions can be asked about whether fiber to the home will continue to be viewed as the “best” way to deliver gigabit Internet access and other services to consumers.

“Project AirGig has tremendous potential to transform internet access globally, well beyond our current broadband footprint and not just in the United States,” said John Donovan, AT&T chief strategy officer.

AT&T says it has more than 100 patents or patent applications supporting this new technology and other access technologies.


“We’re experimenting with multiple ways to send a modulated radio signal around or near medium-voltage power lines,” said Donovan. “There’s no direct electrical connection to the power line required and it has the potential of multi-gigabit speeds in urban, rural and underserved parts of the world.”

Project AirGig is therefore one more potential platform for Internet access and communications that uses fixed wireless.

As part of Project AirGig, AT&T Labs invented low-cost plastic antennas and devices located along the power line to re-generate millimeter wave (mmWave) signals that can be used for 4G LTE and 5G multi-gigabit mobile and fixed deployments.

“These patent-pending devices can mean low hardware and deployment costs while maintaining the highest signal quality,” said Donovan.

Also,  5G standards will include multi-gigabit speeds, and the cable TV industry already envisions 10 Gbps service. Nokia already has demonstrated 10 Gbps symmetrical speeds on hybrid fiber coax, supporting the CableLabs “full duplex” version of DOCSIS 3.1.

All that illustrates a principle: advertised Internet speeds are mostly about marketing, at this point, not “need.” The clearest use case for most accounts is that multiple users in a family or household watch lots of high-definition format streaming video simultaneously.

Generously allocating 10 Mbps per stream would mean a need for 70 Mbps for seven simultaneous HDTV streams.

Some users running servers out of their homes plausibly need similar levels of bandwidth. But the average consumer arguably needs nowhere near 100 Mbps.

A reasonable 2015 analysis of functional need, per user, might have looked something like this:
  • 5 Mbps or less: Basic web surfing and email
  • 5-10 Mbps: Web surfing, email, occasional streaming and online gaming with few connected devices
  • 10-25 Mbps: Moderate HD streaming, online gaming and downloading with a moderate number of connected devices
  • 25-40 Mbps: Heavy HD streaming, online gaming and downloading with a lot of connected devices
  • 40+ Mbps: Hardcore streaming, gaming, and downloading with an extreme number of connected devices.

The fundamental point is that we now are in a phase of development where end user need does not drive bandwidth growth. Suppliers and apps are pushing the trends.

1/3 of Telco Execs Ponder Moves "Up the Stack" into Applications

There are very good reasons why global telecom executives are looking for a range of new revenue generators: the legacy revenue streams are shrinking.

Over the past several years, the telecom business has entered a period of slow decline, with revenue growth down from 4.5 percent to four percent, EBITDA margins down from 25 percent to 17 percent, and cash-flow margins down from 15.6 percent to eight percent, say Paul-Louis Caylar and Alexandre Ménard, McKinsey partners.

Among U.S. telecom companies, landline and mobile voice now account for less than a third of total revenues, down from 55 percent in 2010.

Over the last half decade, mobile data revenue growth has offset the losses in voice and messaging. Mobile data, in fact, now represents 65 percent of total revenues. In 2010, mobile data represented just 25 percent of total revenues.

A third of the 104 respondents to a 2015 McKinsey survey of senior industry leaders said they were preparing to move into adjacent businesses such as financial services, information technology services, media, or utilities in search of new opportunities and revenue streams, McKinsey says.

With the possible exception of media services, most of the opportunities seem to involve Internet of Things to a great extent.


By 2025, as Few as 110 Telecom Service Providers Could Still be in Business

Eventually, the world will have about 10 global-scale telecom providers, with about 100 local network service providers, where today there are perhaps 800 service providers, according to Bell Labs.

Scale always has mattered, but hyperscale arguably matters even more as the world shifts to cloud-based, ultra-high-bandwidth, ultra-low-latency networking literally connecting almost everything.

One reason scale might matter is that investment in state-of-the-art networks will produce capacity 66 times greater than today, by about 2025. The amount of data each person has stored in the cloud will grow by about 20 times.

Local access speeds on fixed networks will grow by perhaps 100 times by 2025, up from perhaps 100 Mbps today to 10 Gbps by 2025.

Radio improvements (multiple input, multiple output) will be something on the order of 250 times today’s performance, by 2025.

source: Bell Labs

India Mobile Market Share Rearranged with Reliance Communications-Aircel Merger

The consolidation of India’s mobile market has taken a big step as Reliance Communications and Aircel are merging their businesses.

The combined entity will have 187.64 million subscribers, creating a new number-three mobile business behind Bharti Airtel Ltd (255.7 million) and Vodafone India Ltd (199.4 million). Until the merger, Idea had been the number-three provider.

The merger vaults the new company into the “top-four” ranks, creating significant separation between the fifth and all other smaller providers.

Nobody yet knows what further changes will occur as Reliance Jio enters the market, either.



Can Regulators Boost Internet Access Speeds by Enabling Cable TV Competition?

source: Ofcom
With the caveat that there are many institutional and historical processes at work, one supply-side policy communications regulators can implement--or should have implemented--is legalizing and aggressively supporting deployment of cable TV networks.

That appears to lead not only to more competition in a market, but also seems to lead to higher overall Internet access speeds.

Likewise, one demand-side development that has helped spur speed upgrades, as well as deployment footprint, is the triple-play bundle pioneered by cable TV companies.

Such are the conclusions one might draw from recent Internet speed tests in the United Kingdom, where Virgin Media, if not ubiquitous, passes about 44 percent of U.K. homes.

One data point: a recent speed test shows Virgin Media providing the fastest downstream speeds, compared to Internet service providers using the BT wholesale network.

In those tests, Virgin Media speeds in the 7 a.m. to 3 p.m. portion of the day were 79 percent faster than BT’s speeds in the same time period, using one test methodology.

In another test, Virgin Media was more than twice as fast as BT in the 7 a.m. time period.

In the 6 p.m. to 3 p.m. period, Virgin Media was either 46 percent faster or twice as fast, depending on test method.

Peak and Off-Peak Download Speed Tests Results September 2016
Provider
tbbx1 Test
(1 download)
httpx6 Test
(6 downloads)
7am-3pm
6pm-midnight
% difference
7am-3pm
6pm-midnight
% difference
BT
24.8 Mbps
23.9 Mbps
-3.8%
21.8 Mbps
21.2 Mbps
-2.8%
EE
14.2 Mbps
13.2 Mbps
-7.6%
14.4 Mbps
13.6 Mbps
-5.9%
Plusnet
23 Mbps
21.6 Mbps
-6.5%
18.7 Mbps
18.5 Mbps
-1.1%
Sky
15.3 Mbps
14 Mbps
-9.3%
13.9 Mbps
12.5 Mbps
-11.2%
TalkTalk
14.7 Mbps
13.1 Mbps
-12.2%
13.8 Mbps
13.3 Mbps
-3.8%
Virgin Media
44.6 Mbps
34.9 Mbps
-27.8%
52.6 Mbps
44.5 Mbps
-18.2%


Sunday, October 2, 2016

Software Will be Biggest Revenue Driver in 2030 Autonomous Vehicle Ecosystem

Software will be the biggest autonomous vehicle value chain winner, with $25 billion in revenues in 2030, a 28 percent compound annual growth rate, according to Lux Research. By that point it is possible autonomous vehicle ecosystem revenues could reach $87 billion.

Optical cameras and radar sensors will amount to $8.7 billion and $5.9 billion opportunities in 2020, respectively. By perhaps  2030 computers will be biggest hardware opportunity on-board autonomous cars, amounting to a $13 billion opportunity.

Ironically, some industries will be severely disrupted. Uber drivers, for example, will be replaced by autonomous vehicles. Auto sales should decline, as more people forego car ownership.

And insurance sales volume could drop as much as 80 percent, according to KPMG and Deloitte.  
GOTW_5_11_14

source: Business Insider

DSRC or 5G?

It is not yet clear whether autonomous car communications (vehicle-to-vehicle and vehicle-to-infrastructure) will use 5G networks or dedicated short range communications (DSRC).

Indeed, it is likely some applications will make more sense for one or other other of the two methods, while yet other connection platforms also are used, as well.

But it is clear that the stakes are quite high for mobile operators, as revenue growth premised on ever-growing subscriptions for mobile data, plus rising recurring payment streams, are close to saturated in some markets.

That means the industry already is searching for the next big driver of incremental revenues.



According to analysts at Deloitte, we already are at the peak of the smartphone era. In fact, 2016 “will likely mark the end of the smartphone growth era, and the start of its consolidation,” Deloitte argues. “A mere nine years after the launch of the first full touchscreen smartphone, adoption is nearing a plateau, at 81 per cent of UK adults, and 91 per cent of 18–44 year olds.”

Generative AI Will NOT have the Impact Many Expect

Generative artificial intelligence, to say nothing of machine learning or neural networks (and eventually general AI), might collectively re...