Providence's Cutting Edge Mesh Network

Posted on March 17th, 2008
By Chewitt

Charlie Hewitt is the Chief Information Officer for the City of Providence.

Many cities and towns world wide are addressing the hot topic of municipal wireless networks. Philadelphia, Boston, San Francisco and others have intentions to provide low cost – even free – public access to the Internet using a WiFi network that blankets the entire city. The city governments promise that the networks will bring affordable service to low income households, foster economic development in the inner city, and increase the efficiency of local government operations.

For the most part, municipal WiFi projects have failed to deliver the expected results. Of the hundreds of cities that set out on the journey several years ago, only a few have networks that are operating today. Many have put their projects on the shelf.

Here in Providence, we considered embarking such a project in the fall of 2004, shortly after Philadelphia announced its plans. We rather quickly concluded that Providence was not ready for this. The risks, both financial and political, were obvious and significant, and we lacked the resources to warrant plunging ahead.

However, we were impressed with the impact that a ubiquitous wireless broadband network might have on municipal government operations, especially public safety. In addition, we needed to find a new way to provide mobile data service to police in cars, because the existing low-speed cellular service, CDPD, was scheduled to terminate sometime in 2006.

We had been researching possible alternatives to CDPD. There were two approaches to consider: 1) leverage the existing network of a wireless service provider like Verizon Wireless; or 2) build and operate a private network. Both approaches had significant disadvantages. As vividly demonstrated during the World Trade Center disaster, the public can easily overwhelm a cellular network just when emergency responders need it the most. On the other hand, a private WiFi network would be expensive to build and operate, and prone to failure due to radio interference.

However, a small, innovative company in Florida, Mesh Networks, suggested a variation on the second approach. Rather than use WiFi, why not use their proprietary protocol which they called MEA (mesh enabled architecture). Originally developed for the US military to support field combat operations, MEA had self-forming, self-healing, robust technology well-suited for public safety applications. The suggestion tempted us, but who would bet the public safety communications of the City of Providence on the strength of an obscure, financially challenged startup? We passed.

Several months later we were still looking for a solution when we learned that Motorola had acquired Mesh Networks. In the public safety communications arena there is no bigger brand than Motorola. We decided to include MEA on the list of viable options.

The whole Mesh Network project, from developing the RFP to initial operational use took about eighteen months. For everyone involved, it was a first-class adventure, much too long to tell here. In the end, in August, 2006, Providence became the first mid-size city in the nation to have a border-to-border, outdoor, wireless broadband data network fully operational and supporting the Providence Police Department. Now, eighteen months later, it is still one of the few successful municipal networks. Providence is regularly cited as a place that got it right.

The Mesh Network has had a major impact on operations within the PPD, enabling the cop in the car to access multi-media information as easily and readily as at a desk in the office. Occasionally the results have been dramatic, as in the quick apprehension of the person who allegedly shot someone near Kennedy Plaza a few days ago. Most of the time, however, the network has simply supported the daily work of the PPD, enabling all kinds of work to be done in the field that used to have to be done in the office.

The greatest value of the network is that it is a force-multiplier. Today it enables cops to spend a lot more time in the field and a lot less time in the office. Over the next several years we hope to extend the use of the network to other city departments that depend on employees working out in the parks, streets and neighborhoods to deliver city services.

So, what about public access? Will the City of Providence follow the lead of other cities and facilitate the development of a ubiquitous, outdoor, low cost WiFi network? At present the City has no plans in this regard, and the experiences of most other cities suggest that this is an area best addressed by private enterprise (either for-profit or non-profit). Such a network probably needs to have a regional footprint because much of its economic value will come from serving mobile workers as they move throughout the local area, not just within the limits of a city or town.

5 comments

Related Items

News
Providence "Innovation and Promise" award finalist for wireless Mesh Network	05.09.07	City of Providence
Providence unveils new wireless safety network	09.09.06	Providence Business News
Providence's First Responders armed with state-of-the-art communications technology	09.05.06	City of Providence

Comments

JackTemplin

Submitted on March 17th, 2008 - 10:23am link

Timely news out of Minneapolis:

New business models for citywide Wi-Fi

Minneapolis is quickly becoming the new poster child for the municipal Wi-Fi movement.

The city is expected to have the majority of its 59-square-mile network finished by the end of this month, and already experts are pointing to the nearly completed network as a model other cities should follow.

ajhunt

Submitted on March 17th, 2008 - 8:10pm link

Would it be possible for organizations to beam their wireless networks and open them up to the public at different spots in providence? For example, the city could subsidize Brown to strengthen its wireless signal so that you can get online close to campus, and Brown would, in return, keep its network open to the public.

(Just a thought).

bjepson

Submitted on March 18th, 2008 - 9:33am link

This is really cool, Charlie. What kind of bandwidth do you get with this set up? I've always wondered whether mesh networks suffer the same problem as Wi-Fi repeaters (halving the bandwidth for each hop). Still, you're replacing CDPD (19.2 kbps), so this is going to work great for your purposes. But if bandwidth does decay as you get out to the edges, I think that's another reason to not open this to the public, as business and home users continue to use bandwidth-hungry apps more and more.

sfreiman

Submitted on March 18th, 2008 - 11:11am link

Nice summary and great project. Because of my involvement in RI-WINs for three years, ubiquitous, border-to-border wireless broadband is a topic that is near and dear to my heart.

I've been watching this project since its inception and it is a great example of "Innovation at Scale" -- using the compact nature of Rhode Island to demonstrate how new ideas can be successfully implemented -- and become a showcase for the rest of the world.

Charlie and the City of Providence have been getting national and international attention and have been been invited to present to conferences and meetings around the world..

As I learned during the RI-WINs project -- the value of wireless broadband capability can be leveraged across ALL sectors -- public, private & education and I hope this project serves as an example for others in the state to consider these kinds of projects in their communities and across the state.

Chewitt

Submitted on March 20th, 2008 - 1:51pm link

Great comments! Thanks, Stuart, for the kind words.

Regarding Minneapolis: that is an important project to watch. It appears to be well-planned and well-executed. However, there are still a lot of open questions (see the comments attached to the article at New business models for citywide Wi-Fi). Their famous bridge collapse demonstrated the value of the network for public safety, for sure. However, that was an outdoor incident that occurred while the network was still under construction and not available to the general public. It remains to be seen whether the network will be suitable for indoor residential use and whether it will respond well for public safety when the network is fully shared with the public. Here in Providence, the Mesh Network works great outdoors, but not-so-great indoors; and just a few concurrent video sessions saturate the capacity of the backhaul. In other words, our network works well for its intended use, but is definitely not suitable for public access.

Regarding technical performance:

1) the network has a two-hop design. In other words, a laptop can usually reach the backhaul in two hops. With one hop we get about 1 Mbps bandwidth, with two hops, about 750 Kbps and three hops, about 500 Kbps.

2) The network operates in the 2.4 Ghz unlicensed spectrum. I was concerned that this would be a problem (we would have specified 4.9 GHz, but products weren't available when we went to bid), but in practice, this hasn't been a problem. The mesh network is quite resistant to interference. Every device in the network is a switch and every data packet is dynamically routed over the best available path.

Regarding the possibility of leveraging private networks to facilitate public access: this has actually been done in some places (I believe Champaign-Urbana, Illinois, is one). However, given the current fiscal situation at both the City and State level, I would expect no help in the form of subsidies. In addition, as a matter of policy, I think the City needs to concentrate its resources on its primary mission areas (e.g., public safety) and rely on private enterprise (both for-profit and non-profit) to provide public access to the Internet.

Regarding RI-WINs: I think as communities get more experience with wireless networks, they will realize that border-to-border within the city limits is only a partial solution. Even within the realm of public safety, we run ambulances beyond the city limits. The solution today is awkward, expensive and complex. We put two modems in the computer: the mesh network card for when we are in Providence; and another card (e.g., an EVDO modem) for when we are somewhere else. A regional network, like RI-WINs, would be far better. Such networks are already being planned in other parts of the US and Canada where the need for a regional solution is more compelling than in Rhode Isalnd.