Broadband services

Google Fiber May Go Wireless in the Valley – Report

Google Fiber is said to have shelved plans to roll out fiber networks in San Jose and other Silicon Valley cities while it considers the use of cheaper wireless alternatives.

The search giant's broadband subsidiary has told officials in Palo Alto and Mountain View, the site of Google's headquarters, that fiber rollouts have been delayed, according to a report from the San Jose Mercury News. Google has also told workers in San Jose that its project is on hold while it looks into "going aerial," according to the same report.

Intriguingly, the report comes just a few weeks after Google (Nasdaq: GOOG) announced plans to buy a San Francisco-based company called Webpass, which uses "point-to-point" fixed wireless technology to support gigabit-speed broadband services. (See Google Fiber Buys Webpass in Wireless Play.)

Eric Schmidt, the executive chairman of Google parent Alphabet Inc. , has also indicated that fixed wireless technology could play a major role in Google Fiber Inc. 's business.

"There appear to be wireless solutions that are point to point that are inexpensive now because of the improvements in semiconductors; that these point to point solutions are cheaper than digging up your garden… and can carry the gigabit performance," he said during a meeting of Alphabet shareholders in June. (See Alphabet Wants to Network the Nation's Cities.)

Google is not the only broadband player in North America that has shown interest in using wireless technologies instead of digging up streets to lay fiber, often at considerable expense.

As noted in a recent blog by Heavy Reading senior analyst Gabriel Brown, the first applications of 5G mmWave technology are targeting fixed access, with both AT&T Inc. (NYSE: T) and Verizon Wireless -- the two biggest telcos in the US -- announcing plans for field trials this year. (See The Return of Fixed Wireless Access.)

This is not the first time that fixed wireless has attracted interest. Several years ago, technologies including LMDS (local multipoint distribution service) and fixed WiMax were grabbing headlines, but neither of those has lasted the distance.

Brown thinks this time could be different, noting that wireless technologies have improved as costs have fallen.

The rollout of gigabit broadband access networks is spreading. Find out what's happening where in our dedicated Gigabit Cities content channel here on Light Reading.

The involvement of Google and other technology giants in this area could obviously help to bring costs down even more. Through its Telecom Infra Project, Google rival Facebook has been developing new wireless technologies as part of its overarching goal of bringing Internet services to underserved communities and improving the economics of Internet access.

Facebook's Terragraph system, for instance, uses unlicensed spectrum in the 60GHz band to provide high-speed connectivity in densely populated areas.

During Light Reading's Big Communications Event in Austin in May, Hans-Juergen Schmidtke, the director of engineering for Facebook's Infrastructure Foundation team, told attendees that Facebook is targeting a cost point for Terragraph that is "significantly" less than that of rival connectivity solutions. (See Facebook Lauds Terragraph Cost Savings.)

Interestingly, Facebook has previously revealed plans to conduct trials of Terragraph in the downtown part of San Jose, which means the city might soon become a testbed for a number of new wireless technologies.

Using more conventional fixed-line infrastructure, Google Fiber has now advanced into the cities of Atlanta, Austin, Kansas City, Provo and Nashville.

While its progress has been relatively slow, market observers believe the company's main objective is not to become a major service provider in its own right but to spur investment in gigabit-speed networks that will support more advanced Internet services.

— Iain Morris, Circle me on Google+ Follow me on TwitterVisit my LinkedIn profile, News Editor, Light Reading

DanJones 8/22/2016 | 12:53:31 PM
Re: Long Time Coming We Will see... I know what you mean though, simply the backhaul challenge of 5G fixed Wireless seems enormous.
brooks7 8/22/2016 | 11:30:52 AM
Re: Long Time Coming Well, it hasn't worked in the past.  No reason to believe it will work now.  People act like these ideas a new.  Clearwire was fixed wireless.  BellSouth did FTTC.  AT&T has done FTTN.  Verizon did FTTH.  We have existence proofs of all of these deployments.  

The reality is that it is all dictated by a construction cost versus bandwidth tradeoff.  The more bandwidth you want, the more construction you have to have.  Lots of small sites = lots of places to just get and then get power to and then have some resiliency for.  

Shannon and Nyquist are still large and in charge.

kq4ym 8/22/2016 | 10:54:20 AM
Re: Long Time Coming It would seem prudent to condider the hybrid approach and I would guess that's just what's going to happen. Getting to the ideas that "these point to point solutions are cheaper than digging up your garden," is probably a too simplistic look at this point even though in some cases that's going to be true, but not everywhere and for every use.
KBode 8/16/2016 | 5:38:02 PM
Kudos! Kudos in getting this story right. I've seen too many reports trying to suggest this somehow means that Google Fiber has hit a "snag" or is suddenly overwhelmed by the high cost of fiber deployment. Webpass is a supplemental point to point solution, and Google Fiber has long eyed a variety of different experimental wireless solutions to help complement existing fiber builds. 
DanJones 8/10/2016 | 4:26:02 PM
Re: Long Time Coming They continue to chip away at a hybrid/HetNet approach. Could take years to come to final fruition but Google has some time on its side.

See: http://www.lightreading.com/mobile/5g/google-could-stitch-5g-into-hetnet-future/d/d-id/725352?
mendyk 8/10/2016 | 4:18:51 PM
Re: Long Time Coming The key point here is the hybrid approach. Using Rumsfeldian logic, fiber is a known known. Fixed broadband wireless is an unknown known (the unknown part being exactly how it can scale, and then there's the whole landscape/geography issue). Yes, fiber is more expensive to deploy (and probably maintain) than wireless, but there's also the opportunity cost of waiting for broadband wireless to catch up (if it ever can). Hybrid is a hedge, which is why it looks like the preferable option.
DanJones 8/10/2016 | 3:37:08 PM
Long Time Coming They've been looking to combine wireless (licensed/unlicensed) and fiber networks for a long time. I reported in January 2015 that Google was looking for engineer that could stitch together cell, WiFi etc for Google Fiber.

See: http://www.lightreading.com/google-searching-for-5g-wireless-engineer/d/d-id/713377
msilbey 8/10/2016 | 1:40:46 PM
Delay in Portland too Google Fiber postponed its rollout in Portland too, and the speculation is that it could look at fixed wireless tech there too -

Sign In