Google: Fiber to the Home?

Google says it planned to build an open-access, fiber-to-the-home service, at speeds up to 1Gbps. Helping to make Internet access better, and faster for everyone.

Google says it planned to build an open-access, fiber-to-the-home service, at speeds up to 1Gbps. Google’s Public Policy Blog says the project will build on their ongoing efforts to expand and improve Internet access for consumers – from their free municipal Wi-Fi network in Mountain View, CA, to their advocacy in the 700 MHz spectrum auction, to their work to open the TV “white spaces” to unlicensed uses.

We’ll deliver Internet speeds more than 100 times faster than what most Americans have access to today with 1 gigabit per second, fiber-to-the-home connections. We’ll offer service at a competitive price to at least 50,000, and potentially up to 500,000 people.

As a first step, we’re putting out a Request for Information (RFI) to help identify interested communities. We welcome responses from local government, as well as members of the public.

Our goal is to experiment with new ways to help make Internet access better, and faster for everyone.

“We’ll operate an ‘open access’ network,” said Googe, “giving users the choice of multiple service providers. And consistent with our past advocacy, we’ll manage our network in an open, nondiscriminatory, and transparent way.”

A source told the Washington Post that Google doesn’t plan to widely offer service beyond the trial.

AT&T and Verizon would be forced to lease fiber to rivals providing Web services to small businesses under a proposal being weighed by U.S. regulators. That may be Google’s real motivation. Google’s partner Clear was forced into building their own wireless backhaul network to feed WiMAX towers, because AT&T and Verizon priced their access high enough to force out nearly all competitors.

iProvo is the name of the Fiber to the Home service owned and operated by the city of Provo, Utah. It is the largest municipally-owned Fiber to the Home network in the United States.

An iProvo RFP [pdf], was issued on April 18, 2007 to reorganize the system which has run into revenue shortfalls and cost increases.

Meanwhile, the Utah Telecommunication Open Infrastructure Agency (UTOPIA), a consortium of 16 Utah cities, deployed a fiber to the premises network to every business and household (about 140,000) within its footprint.

Mayor Sam Adams says Portland is shovel ready for FTTP having completed a business case and feasibility study as well as a Staff White Paper that summarizes why Portland is ideal for FTTP a couple of years ago.

Who’s got an underutilized, ubiquitous fiber network that covers your entire state? That would be your Department of Transportation.

The U.S. government should take a look at South Korea, says Telecom Magazine, where the government subsidizes and encourages broadband growth. The United States lacks such a policy. A new alliance of industry leaders, called the Internet Innovation Alliance, is coming together to try to get a national broadband strategy.

Canada’s Alberta SuperNet is one of the largest private IP networks in the world, a high-speed, high-capacity broadband network linking government offices, schools, health-care facilities and libraries, including approximately 4,200 connections in 429 communities. Prior to the SuperNet, there were only seven service providers operating outside of Alberta’s two largest metropolitan centres. There are now 81.

According to the ITU’s “Digital Opportunity Index: the Top 20 Countries” survey, Korea takes the top spot with household broadband penetration of 89.4 percent. The U.S. is number 20 with broadband penetration at 55%. The U.S. is in 15th place among developed nations, according to the Organization for Economic Cooperation & Development. It was ranked 4th in 2001.

Only 10 percent of Americans still use dial-up Internet service at home, according to the latest findings from the Pew Internet and American Life Project.

Broadband is still too expensive (or scarce) for many Americans. Of the 10% of U.S. households that still use dial-up access, more than a third say they use it because they can’t afford faster service. And another 14% still use dial-up because it’s the only option available to them.

But fiber to the tower would impact more people. Cellular operators are installing their own fiber networks to satisfy demand by broadband wireless users.

New Paradigm says 85% of cell sites in the United States are still being served by copper (T-1) lines while Infonetics Research says backhaul is 30% of a carrier’s expense. But AT&T and Verizon aren’t into sharing their fiber with competitors.

Industry observers estimate about 50 governments across the United States have invested public money to create high-speed fiber connections to homes and businesses. Countries like the UK (through OpenReach) and Australia are working on fiber networks that will be maintained by one entity, but open to all ISPs.

Digging trenches and providing overhead fiber to homes is horrendously expensive and requires a massive corporate infrastructure.

It will cost Verizon $23 billion to run fiber past 18 million homes, roughly half of those served by Verizon, reports the NY Times. Here is how Craig Moffett looks at Verizon’s costs for FiOS.

Through 2010 the company will pay an average of $817 to run the fiber past the 19 million homes, on poles or under the ground. It will also incur $172 per home passed in other costs related to the video infrastructure. He assumes that 40 percent of the customers passed will buy at least one FiOS service.

If you allocate the cost of running the fiber past the homes that don’t buy FiOS to those that do, that makes the cost of building the network $2,473 per home. (That cost would be less if more than 40 percent of the potential customers sign up. Or it could be higher, if sales don’t achieve the 40 percent level.)

But the costs don’t end there. Once someone decides to buy FiOS, Verizon will spend another $718, as Mr. Moffett calculates it, for equipment in the home and the labor to connect it to the network. There is another $130 in marketing expenses, and $576 for interest on money spent to build the network before it is completed.

All this adds up to $3,897 in capital cost for every FiOS customer.

According to Wikipedia, as of 2009, the number of homes with FiOS availability was 12.7 million, of which 3.1 million subscribe to the Internet service, and 2.04 million to FiOS TV.

But fiber to WiMAX…well that would be something else.

While Verizon pays $23 billion to run fiber past 18 million homes, Google might pay something like $2.3 billion to offer 20Mbps service to 180 million homes. That’s 1/100th the cost. FiOS charges $50/month for 15 Mbps. Google could offer it free and make money with advertising. Their cost per home is closer to $40 than $4,000.

Samsung is demonstrating Mobile WiMax 2.0 (IEEE 802.16m) technology. Samsung says they will provide the first trial solutions of Mobile WiMAX Release 2 for Yota in Russia, Clearwire in the United States and UQ Communications in Japan by late 2010, and work closely with other major operators. Significant features of 802.16m (that aren’t available on FDD LTE) include; multi-hop relay architecture, multi-carrier operation, beamforming and interference mitigation.

WiMAX 2.0 is 4 times faster than Mobile WiMax (IEEE 802.16e). The WiMAX Forum expects to see WiMAX Release 2 available commercially in the 2011-2012 timeframe.

By using 4X2 MIMO in an urban microcell, and 20 MHz TDD channel (double the usual 10 MHz), the 802.16m system can support both a 120 Mbit/s downlink and 60 Mbit/s uplink per site simultaneously, says the WiMAX ForumTriple play with a Boxee Box. Using fixed, rather than mobile, WiMAX 2.0, like other “true” 4G standards, is expected to deliver up to 1 Gbps to homes — that might be right around the 100 Mbps target.

Buying a 60 MHz chunk on the Clear/Sprint network for a 120Mbps nationwide WiMAX 2.0 service would be cheap by comparison, and should typically deliver 20 Mbps (mobile) or 50-100 Mbps (fixed) per home. That’s comparable to AT&T’s U-Verse or Verizon’s FiOS and enough for multiple channels of HDTV. Sharing backbone fiber would make good economic (and political) sense for Google.

Another (fantasy) option would be the “D Block” at 700 MHz. The FCC paired the upper band D block (a single 10 MHz nationwide license) with 10MHz of public safety spectrum located next to the D block, and conditioned the D block license on an obligation to negotiate with public safety representatives towards the construction by the D block licensee of a nationwide public safety network. The Inspector General’s Report (pdf) reviews the failure of the D Block to find an investor in the 2008 auction.

Verizon paid $4.6 billion for 22Mhz on the 700MHz band and plans LTE service. Cellular infrastructure is more expensive than WiMAX. The ongoing network investment for Verizon Wireless now totals nearly $5.7 billion in California and more than $55 billion nationally since the company was formed in 2000.

But neither public service users nor the FCC seem likely to acknowledge the merits of data-centric WiMAX. They’ll stick to LTE’s symmetrical FDD and blow away half the bandwidth. Still, that could deliver 3-5 Mbps to everyone in the country for a few billion dollars.

I just tested my WiMax connection in Portland, Oregon, for which I pay $40/mo:
Download Speed: 6528 kbps (816 KB/sec transfer rate)
Upload Speed: 554 kbps (69.3 KB/sec transfer rate)

WiMax 2.0 should be 2-4 times faster because it:
1. Doubles the channel space to 20MHz
2. Utilizes 4×2 MIMO and other techniques.

4 X 6 = 24.

Luke Pensworth Written by: Luke Pensworth

Luke is the managing editor and site manager of Dailywireless. As a wireless enthusiast/consumer, he reviews a lot of services based on his own experience. Disgruntled as he may be, he tries to keep his articles as honest as possible.

Leave a Comment

This site uses Akismet to reduce spam. Learn how your comment data is processed.