Mobile Drives Subscriptions, Fixed Networks Drive Data Volume

The proportion of Internet access instances originating from mobile devices will vastly outnumber fixed network access, including both fixed networks and Wi-Fi, by 2018, Cisco now forecasts.

It is likely that fixed networks still will represent an order of magnitude more total data throughput than mobile networks do, by 2018, according to Ericsson.

source: Informa Telecoms and Media

Wider adoption of both smartphones and faster networks will help propel the trend, as will consumption of video on phones. Already, real-time video entertainment drives 63 percent of fixed network data usage and 43 percent of mobile data usage.

Globally, smart devices represented 21 percent of the total mobile devices and connections in 2013, but accounted for 88 percent of the mobile data traffic.

As more smartphones are added to the installed base, bandwidth demand is going to grow in almost linear fashion. In 2013, on an average, a smart device generated 29 times more traffic than a non-smart device.

By 2018, there will be more than 10 billion mobile-ready devices and connections, about three billion more than there were in 2013, and about five billion global mobile users, up from more than four billion in 2013.

Global mobile data traffic grew 81 percent in 2013, nearly 18 times the size of the entire global Internet in 2000.

One exabyte of traffic traversed the global Internet in 2000, and in 2013 mobile networks carried nearly 18 exabytes of traffic, while mobile video traffic exceeded 50 percent for the first time in 2012.

By 2018, global mobile IP traffic will reach an annual run rate of 190 exabytes, up from less than 18 exabytes in 2013

And one might reasonably expect that ultra-low-cost smartphones, costing perhaps $25 each, instead of $200 to $600, will help drive the trend, as will growing instances of Internet access by sensors (machine-to-machine apps).

Globally, residential Internet users with fixed Internet access will grow from 1.9 billion in 2013 to 2.5 billion by 2018, according to Cisco's Visual Networking Index.

The total number of global consumer mobile devices and connections will grow from 6.1 billion in 2013 to 8.9 billion by 2018, the perhaps-significant development being that there will be 3.5 billion consumer smartphones in use by 2018, up from 1.5 billion in 2013.

That means Internet access will happen from 3.5 billion smartphones (and some billions of feature phones, to an extent), and 2.5 billion fixed locations, by 2018.

Ultimately, one might argue, mobile access could reach beyond 3.5 billion up to almost nine billion, assuming smart phones eventually become “phones,” and are used by nearly all mobile subscribers.

That will have “speed” implications, as the typical fixed connection will operate at 42 Mbps in 2018, according to the Cisco Visual Networking Index.  

Western Europe will experience the fastest average fixed network speed of 49 Mbps by 2018, and Asia Pacific will have the greatest average speed increase, 2.7-fold growth by 2018 with fixed speeds reaching 48 Mbps by 2018.

Globally, Wi-Fi connection speeds available to mobile devices will more than double by 2018, exceeding 21 Mbps by 2018, yet still representing access at speeds less than half that of the fixed network.

Globally, mobile network connection speeds will nearly double by 2018, yet increase only to 2.5 Mbps by 2018. So fixed networks will continue to operate at an order of magnitude faster speeds than mobile networks, in aggregate.

North America will have the highest average mobile speed (4.5 Mbps) by 2018 along with the fastest growth (2.6-fold) from 2013 to 2018, according to Cisco.

The point is that, as Internet access shifts to mobile methods, the notion of “average” access speed will become more complex. And “average” speeds could decline.

A crucial factor promoting the increase in mobile speeds over the forecast period is the increasing proportion of fourth-generation (4G)mobile connections. The impact of 4G connections on traffic is significant, because 4G connections, which include mobile WiMAX and Long-Term Evolution (LTE), generate a disproportionate amount of mobile data traffic.

Global Mobile Speed Growth

Visual Networking Index

The composition of IP traffic will shift dramatically in the coming few years, Cisco predicts.

During the forecast period, the majority of traffic will originate from devices other than personal computers (PCs) for the first time in the history of the Internet.

Wi-Fi traffic will exceed wired traffic for the first time and high-definition (HD) video will generate more traffic than standard definition (SD) video.

The Internet of Everything is also gaining momentum and by 2018 there will be nearly as many machine-to-machine (M2M) connections as there are people on earth. Smart cars will have nearly four M2M modules per car.

$25 Smartphones Will Disrupt Forecasts of Smartphone Use in Asia

As typically is the case, the emergence of low-cost smartphones is going to disrupt all existing forecasts of smartphone adoption in Asia, as those forecasts rationally have been based on devices costing more than $200, not the projected $25 of the new Mozilla-using devices.

Low-cost smartphones are expected to have a huge role in extending mobile Internet access in Asian markets. Mozilla, for example, is introducing “ultra-low-cost” devices to India in the next few months, said to cost about $25, an order of magnitude less than existing high-end smartphones.

Mozilla has partnered with Intex and Spice, two of India’s leading mobile device brands, to bring the first Firefox OS devices to India in the next few months.

The same trend is expected to happen elsewhere in Asia. low-cost smartphones are a key strategy in the China mobile market as well.

In 2014, for example, low-cost phones will drive smartphone sales in both India and China, but sales have been robust in Southeast Asia as well, over the past several years.

Wi-Fi Offload Will Represent 52% of Mobile Internet Traffic in 2018

Percent of Total Mobile Data Traffic Will Be Offloaded by 2018

Mobile Internet access traffic offloaded from mobile networks to Wi-Fi represented 45 percent of total mobile Internet traffic in 2013, and will grow to 52 percent of total mobile Internet access traffic by 2018, according to Cisco’s Visual Networking Index.

Likewise, the amount of traffic offloaded from smartphones will be 51 percent by 2018, and the amount of traffic offloaded from tablets will be 69 percent by 2018.

That is a good indicator of how foundational and strategic unlicensed spectrum and fixed network Internet access has become for mobile service providers and their customers, even when the fixed network access is not owned by the mobile service provider.

That, in turn, illustrates a fundamental principle of modern computing and communications, namely that applications and access are fundamentally separated.

The implications for communications service providers are decidedly mixed, as a result. On one hand, the ability to offload traffic to third party networks creates new end user value for suppliers of fixed access.

Consumers understand how offloaded mobile Internet access saves them money on their mobile service and also improves experience, since fixed networks can run an order of magnitude faster than mobile connections, and typically between twice and five times faster than mobile connections.

Mobile service providers also are able to avoid making capital investments in their networks as often as would otherwise be the case.

42 Percent of Global Tablets Will Be Cellular Connected by 2018

The ability to offload mobile phone data demand to Wi-Fi networks reduces demand growth on mobile networks by about four percent a year, according to Cisco’s Visual Networking Index.

Global mobile data traffic would grow at a compound annual growth rate of 65 percent instead of 61 percent, were offload to Wi-Fi not possible.

At the same time, more consumers are choosing to connect their tablets using the mobile networks, a trend that fuels net subscriber growth for the leading four national mobile service providers in the United States.

Cisco estimates that by 2018, 42 percent of all tablets will have a mobile connection, up from 34 percent in 2013.

Cisco’s estimates of mobile data traffic offload includes traffic from both public hotspots as well as residential Wi-Fi networks.

Some suspect Wi-Fi offload will be less relevant once Long Term Evolution 4G networks are widely used. That argument assumes consumers default to Wi-Fi primarily for reasons of better user experience.

To the extent that consumers also default ot Wi-Fi to save money by limiting mobile data plan usage, that might not be the case.

In fact, 4G networks lead to much-higher data consumption, and that usage plans are volume sensitive, consumers will have high incentives to offload traffic to Wi-Fi, when possible.

In 2018, for example, Cisco estimates that 4G connections will be 15 percent of the global total. At the same time, in 2018, 4G networks also will represent 51 percent of total global mobile data traffic.

While 3G and 3.5G account for 60 percent of mobile data traffic today, a 4G connection generates nearly 15 times more traffic than a non-4G connection, according to Cisco.

In large part, that difference in consumption is driven by both higher usage and higher-bandwidth apps on 4G devices, especially video. Mobile video traffic exceeded 50 percent for the first time in 2012 and was 53 percent of traffic by the end of 2013.

Mobile Data Traffic and Offload Traffic, 2018

Also, as smartphones become the devices of choice, consumption will grow. In 2013, on average, a smart device generated 29 times more traffic than a non-smart device, Cisco estimates.

In 2013, a 4G connection generated 14.5 times more traffic on average than a non‑4G connection. Although 4G connections represent only 2.9 percent of mobile connections today, they already account for 30 percent of mobile data traffic.

The point is that so long as mobile data plans are usage-based, there will continue to be high demand to offload traffic to Wi-Fi networks.

Wearables Will Drive Half of All App Interactions by 2017?

Gartner predicts that by 2017 wearable devices will drive 50 percent of total app interactions, part of a trend that also will have mobile-originated apps by 2015, most mobile apps will be syncing, collecting and analyzing data about users and their social graphs.

Should that trend emerge, it will be one more inication that the lines that separate “machine-to-machine” or “Internet of Things” applications and devices from “connected devices” such as tablets, watches, game players and TVs are blurring.

Machine to machine apps, typically involving the collection of sensor data, would in such cases also be instances of sensor apps. In a similar way, sensor data from smartphones likewise might be considered part of the Internet of Things trend, even if the device itself is a "traditional" appliance, not an industrial sensor.

That is one reason why forecasts of IoT revenues are so disparate at the moment. Aside from the difficulty of forecasting the emergence of what most expect will be a big industry, delineating revenues is a problem.

Is sensor data collected from a mobile app on a wearable device an instance of IoT or an instance of traditional Internet use?

International Data Corporation has predicted that IoT technology and services spending would generate global revenues of $4.8 trillion in 2012 and $8.9 trillion by 2020, growing at a compound annual rate (CAGR) of 7.9 percent.

By 2020, the installed base Internet of Things devices will be approximately 212 billion globally, including 30.1 billion installed "connected (autonomous) things" in 2020, arguably mostly

sensors.

Other forecasts are more conservative. Machina Research expects an installed base of about a billion devices by about 2022, with revenue of about a trillion dollars.

Google Buys Skybox Imaging: Maps Will Benefit, Maybe Internet Access Later?

source: Business Insider

Google is buying Skybox Imaging, a firm that supplies high-quality satellite imaging services, but also manufactures low-cost satellites. Skybox satellites are about the size of a phone book.

The immediate application likely will be to improve the granularity of Google Maps images. But some speculate the phone-book-sized satellites might eventually play some role in satellite-delivered Internet access as well.

The $500 million acquisition might also eventually help Google supply low-cost internet access in underserved regions

The expansion into satellites comes two months after Google bought drone maker Titan Aerospace, a move that might also ultimately have some application for Internet access, but immediately should help Google improve the images presented as part of Google Maps.

NFC Adoption Forecast Slow in North America, Western Europe, Next 2 Years

At least so far, near field communications (NFC) has not propelled the mobile payments business as fast as some had predicted. And Juniper Research has now scaled back its forecasts for NFC adoption in the North American and Western European markets, though not the South Korean and Japan forecasts.

To be sure, skepticism about NFC has been growing for years, despite some thinking that NFC and mobile payments would hit an inflection point in 2013. In fact, some might argue Juniper Research made the NFC transaction revenues reduction in 2013.

On the other hand, neither should near term sluggishness in adoption come as a surprise.

Even popular consumer innovations can take a decade or more to reach 10 percent adoption, when the underlying ecosystem is complex, and has to be built almost from scratch. Use of automated teller machine cards and debit cards provide an example.

As useful as those innovations are, it took a decade for each to reach 10 percent adoption by the public. NFC adoption arguably will be tougher as debit and ATM cards required consumers only to change behavior and use new plastic cards. NFC requires replacing merchant cash registers and requires users to buy NFC-capable smartphones.

Adoption of ATM and debit cards cost consumers virtually nothing, or very little, in terms of transaction costs.

Also, one might argue the value proposition for use of ATM cards and debit cards was easier to understand. Debit cards replaced checks and check charges. ATM cards allowed users to get cash, or make deposits, outside of normal banking hours.

Paying by phone might not offer so much value over using a debit card.

While the Juniper Research report finds that by 2017 the proportion of NFC-enabled smartphones will be only marginally below previous estimates, global NFC retail transaction values are now expected to reach $110 billion in 2017, significantly below the $180 billion previously forecast.

Apple’s decision to omit an NFC chipset from the iPhone 5 has reduced retailer and brand confidence in the technology, leading to reduced POS (Point of Sale) rollouts and fewer NFC campaigns.

In turn, this will lead to lower NFC visibility and adoption in the near term, at least in the North American and Western European markets, creating a “two year lag” on previous forecasts.

“While many vendors have introduced NFC-enabled smartphones, Apple’s decision is a significant blow for the technology, particularly given its previous successes in educating the wider public about new mobile services” said Dr. Windsor Holden, Juniper Research analyst. “Without their support, it will be even more difficult to persuade consumers – and retailers – to embrace what amounts to a wholly new means of payment.”

Conversely, retail transactions in NFC’s heartland in Japan and Korea are likely to experience little or no impact from the decision, Juniper Research forecasts.

Google Wallet and ISIS likewise have struggled to gain traction in the U.S. market, and both rely substantially on NFC.

Is There a "Silver Bullet" for Reducing Fiber to Home Costs?

Many would like to discover some silver bullet capable of dramatically reducing the cost of fiber to the home networks.

And there have been improvements.

A municipal fiber to the home network in Loma Linda, Calif. has been able to reduce access costs from $50 a foot to $12-$18 per foot, a saving of between 64 percent to 76 percent in the most labor intensive and expensive part of the access network, according to m2fx.

Use of micro-trenching and use of m2fx cabling avoided traditional trenching for underground construction. Instead, micro trenching required only cutting a one-inch wide trench, installing micro duct and rapidly restoring roadways.

Once the micro duct is installed, m2fx Miniflex cable can simply be pushed or pulled to its

destination from the manhole, without the need for expensive blowing equipment

or specialized skills.

That is important, as it obviates the need for use of special installation crews. The majority of deployments are completed by municipal staff, with the City electrician deploying fiber to the premises, m2fx says.

Whether such techniques can scale in larger metro areas requiring retrofits and rebuilds is a question some might have. Loma Linda has only 1,600 houses to connect. And some might question the economics.

Since 2004, all new Loma Linda homes have been installed with structured wiring.

The structured wiring and fiber connection added $3,000 to the cost of a new home, according to Konrad Bolowich, Loma Linda’s IT director.

KB Homes built two essentially identical new developments, and pre-wired Loma Linda homes sold for $10,000 to $12,000 more than the unwired, but otherwise identical, models nearby.

By ordinance, new homes built in Loma Linda must have structured access cabling costing a developer about $3,500.  

Bolowich said in 2013 that the city is still trying to come up with a cost effective solution for hooking up pre–2004 homes.

A pilot project cost the city between $1,200 and $1,500 each for the 36 retrofitted homes connected, too much to provide a payback, even assuming about a 50-percent take rate.

The municipal-provided Internet access service costs $30 per month for 5 Mbps service, going up to $100 a month for 15 Mbps service.

The point is that Loma Linda, even using the new trenching method, might not have found a repeatable and sustainable payback model for most housing, which is not greenfield construction, in new developments.

Already, the city offloads the access cost for newbuilds to developers. But Loma Linda cannot do so for most houses.

So there appears to be no “silver bullet” solution for fiber to home construction costs, despite the improvements in trenching and cabling costs.