Municipal Wi-Fi 2.0

As large-scale, for-profit projects falter, innovative new models emerge

Municipal Wi-Fi is dead. At least in terms of the model whereby cities award an exclusive contract to a service provider, who builds a citywide mesh network and hopes to make money selling broadband to consumers.

Municipal Wi-Fi is dead. At least in terms of the model whereby cities award an exclusive contract to a service provider, who builds a citywide mesh network and hopes to make money selling broadband to consumers.

But municipal Wi-Fi is also alive and well in a variety of other incarnations, including municipally owned networks deployed for specific purposes such as public safety, viral nets put up by non-government entities, and imaginative efforts that combine wireline and wireless technologies. (Compare Enterprise Wireless LAN products.)


See an interactive map that pinpoints grassroots projects around the country.


The initial promise from both wireless advocates and Wi-Fi equipment vendors was based on the fact that inexpensive Wi-Fi radios were relatively easy to deploy. The technology seemed to offer a way to bring mobility to city employees, create an inexpensive alternative to wired broadband Internet access (and to cellular data services), and extend high-speed access to people who weren't offered it by incumbent providers or who couldn't afford it.

In the 2004-2006 time period, scores of municipalities of every size were writing up requests for proposals for something that had never been deployed before: large-scale, Wi-Fi mesh networks.

The results have been disappointing. Many large-scale schemes have collapsed, stalled, or entered a Twilight Zone of uncertainty.

"It's pretty clear that 'free Wi-Fi' was an unrealistic expectation," says Stan Schatt, vice president and research director at ABI Research. "What's happened is that the early business models didn't work. They weren't realistic."

Unproven business plans, technical limitations and occasional political conflict created a mutually reinforcing downward spiral. And many of these issues remain to be resolved in any future wireless deployment.

Municipal Wi-Fi 1.0: What went wrong?

Following the lead of Philadelphia, the preferred model for many was a public-private partnership that was familiar, since it was a variant of cable TV licensing. Glenn Fleishman, journalist and editor of Wi-Fi Network News, calls this "municipally-authorized Wi-Fi."

In exchange for an exclusive corporate franchise, and access to municipal buildings and utility poles for the radios, the network provider promised to foot the entire bill to blanket the city, offer a low-cost or free tier of service, undercut broadband offerings from the cable vendors and telcos, and reap wireless advertising revenues.

Cities typically refused to invest any capital funds, or any funds for that matter. "The 'public-private' label in practice meant 'soak the vendor,'" says Craig Settles, principal with Successful.com, an Oakland industry analyst and author, who focuses on municipal broadband issues. "The idea was 'it's got to be free, and this technology will do everything close to parting the Red Sea.'"

In some cases, cities agreed to become an "anchor tenant" – a customer on the network, using it for mobile city workers and public safety staff instead of cellular data plans.

EarthLink pulls the plug

This whole concept suffered a body blow a year ago when EarthLink put the brakes on, and eventually scrapped, its ambitious plan to deploy Wi-Fi mesh networks in big cities as an alternative to wired DSL services. That leaves projects like Philadelphia trying to sort out what do to next. Corpus Christi, which sold to EarthLink the wireless network it developed on its own, has now signed a deal to buy it back. The city will own and operate it as an extension of the city's wired enterprise network, for municipal services (see story). 

Portland, Oregon's wireless network is stalled because the start-up vendor which runs it, MetroFi, says it needs an infusion of cash, and now wants the city to renegotiate the contract and become an anchor tenant, assuring a minimal yearly revenue annuity. The city so far has said no, insisting that MetroFi has a contractual obligation to try to complete the network, now used by about 15,000 users per month.

The 900-node wireless mesh in Tempe, Ariz., has passed through at least three owners in less than three years. The network effectively shut down for residential users at the end of 2007 when the current owner, Globility, came to an impasse with its suppliers, and unplugged the network's authentication servers. The city declared Globility in default of the contract and may declare the network abandoned, says Dave Heck, Tempe's CIO.

Technical troubles

Technical complications of all kinds made, and make, network performance widely uneven in many cases, though it improves somewhat with more radio nodes.

"They were so oversold, there was no way to match the hype," Fleishman says. "I don't think the hardware vendors were deliberately misleading their customers. But it turns out that doing radio frequency networks in urban environments is more complicated than anyone thought."

Minneapolis ran into some of those complications early. As in most such projects, the city planned to let the network operator, USI Wireless, hang mesh radio nodes from city-owned utility poles. "We found not all poles can sustain the weight of those radios," says Lynn Willenbring, the city's CIO. "That was a disappointing discovery." So was the realization that electricity to run the pole-mounted streetlights, and the Wi-Fi radios, was only turned on at dusk by an automated system of photoelectric cells. The costly solution: run electricity to the poles for the radios. 

It was an issue that came up repeatedly around the country, Fleishman says.

With no experience in building large-scale networks, wireless mesh vendors at first recommended 20 to 25 wireless nodes per square mile. But many urban projects, such as Philadelphia's, discovered they had to double that number, and double the capital outlay, to get adequate coverage. And then they discovered that even with the expanded coverage, radios signal penetration into buildings was poor, often requiring an access point or repeater in the subscribers home or business. It was an especially difficult problem where stucco was a favored exterior material: the wire mesh underneath the plaster stubbornly blocked signals.

One completely unexpected complication was the response of incumbent service providers, Fleishman says. Their first reflex was to lobby state legislatures to block or frustrate municipal wireless infrastructure, an effort that had mixed success. But it was quickly replaced by a far more effective ploy: expanded wired broadband service and lower prices.

"EarthLink, MetroFi, [and cities like] Lompoc, [Calif.] didn't think the incumbent broadband providers would respond as fast as they did," Fleishman says. Incumbents slashed prices for DSL and other services, and started digging more fiber, "dramatically increasing access to wired broadband" in the case of Lompoc, he says.

Lessons learned

One way to look at the struggles of the past three years is to think of it as an immense, multimillion dollar, public laboratory for high-speed network access. As with any lab project, if one approach fails, you regroup and try again.

Many municipalities see the unlicensed spectrum and the licensed 4.9GHz public safety band simply as a way to extend the city's enterprise network. These networks are limited to or mainly for municipal employees -- building inspectors, public works teams, automated meter reading, police and fire -- sometimes combined with Wi-Fi or other wireless bridges to eliminate leased T-1 lines. Some also support Wi-Fi connectivity in hotspots or hotzones -- specific locations or areas, such as the core downtown business district.

Malden, Mass., a Boston suburb covering less than 5 square miles, exemplifies this approach. The project initially started with federal homeland security funds to create a limited wireless video surveillance network, using fixed and portable cameras. "We had great line-of-sight in our topography, so rolling this out was less problematic for us than for some others," says Anthony Rodrigues, the city's director of IT.

For about $100,000, mostly in federal homeland security funds, Malden has created what Rodrigues calls a lily-pad model of connectivity: creating a backbone mesh based on Proxim's "pre-WiMAX" 60Mbps Tsunami radio base stations, in the unlicensed 5GHz band, along with associated subscriber units and wired-in cameras, and Wi-Fi access points in some areas, to interconnect buildings and cover selected areas such as city squares and main road corridors.

Fire stations, schools, municipal buildings are now wirelessly linked, eliminating leased T-1 lines, and saving the city about $20,000 a year. Police need full mobility, and Malden relies on cellular voice and data services for that.

Residents can access the network, via what is in effect a Wi-Fi guest account with an associated grant of bandwidth, but can only access a local Web portal of city information and services. But ISP-like service was never part of the plan. "We have a great deal of [wired] broadband here," Rodrigues says. "I'm in the business of running city operations. An ISP is a totally different endeavor."

Hot zones are hot

"Hot zones are back in favor," Fleishman says. "There's a resurgence of interest, because of increasing numbers of users, in covering 20 city blocks, to get [limited] free Wi-Fi, and some quality of life and promotional benefits."

Some projects combine mobile city services with the goal of also being a wireless broadband access alternative for community residents, with "digital inclusion" programs (usually funded with grants and contributions) to train, equip, and support low-income families with computers and applications.

Some of these, in turn, stipulated a wholesale network operator, who then would lease the network out to various wireless service providers who, it was predicted, would flock to the network because they didn't have to spend their own money to build it.

The skinny on Minny

Minneapolis, for example, managed to avoid the political battles that have crippled if not killed some wireless projects. Some observers say that's because the wireless network was originally a project of the city's IT department, sidestepping the political process.

Surprisingly, the Minneapolis network is not yet fully deployed to residents or even to police and firefighters. About 85% of the city is covered, Willenbring says, and until at least the 4.9GHz public safety band is deployed border-to-border, the city's emergency staff won't be using it. It is being used now by some inspectors and public works employees, who don't need that kind coverage.

The network's sole owner and operator, USI Wireless, has signed up about 8,000 users, from roughly 175,000 households (total population is 385,000), according to Willenbring. The vendor agreed to set up a number of free-access sites and zones around the city, at libraries for example, and donates a portion of its profits to a digital inclusion fund, managed by the non-profit Minneapolis Foundation, which oversees that effort.

USI Wireless is the Internet provider to residential and business users, with the city as the anchor tenant for the network, providing several million dollars in upfront investment. "That was how it was going to be financially viable for a vendor," Willenbring says. "We didn't want to own and operate this network. We don't have the competency or the financial wherewithal to do that."

Willenbring's comments run counter to a strong and vocal activist consensus that argues that municipalities "have to take more, not less, responsibility and control over the broadband networks they're involved with," writes Sascha Meinrath, research director, Wireless Future Program, at the New America Foundation, a Washington, D.C. public policy group.

Meinrath argues that 2007 showed the failure of the free market approach to critical infrastructure. Government has a traditional role, he writes, as the builders and maintainers of critical infrastructure. Rather than focusing narrowly on wireless networks, he says, we should be focusing on the larger issue of municipally controlled or owned broadband networks, which may make use of a variety of wired and wireless network technologies.

"We need to get back to the original rationales [of] why we should be building these networks in the first place," he says. "Personally, I'm business model agnostic. I'm far more focused on how these models meet the social and economic justice

needs of the communities they serve."

Settles agrees on the need for a broader perspective. "In 2005 and 2006, it was all about Wi-Fi," he says. "They had self-imposed tunnel vision that limited opportunities, and it created problems because Wi-Fi isn't the best solution in many cases." The key is to focus on the issue of high-speed network access, which can then make use of many different technologies, both wired and wireless, and many often unexpected opportunities.

1 2 Page
Insider Tip: 12 easy ways to tune your Wi-Fi network
Join the discussion
Be the first to comment on this article. Our Commenting Policies