Monday, January 26, 2009

Some Myths and Misunderstandings about Wireless Safety Broadband

January 25, 2009

Myths and misunderstandings surround the discussion of the “D Block”, the proposed creation of a national wireless broadband network for safety agencies and users. Three in particular appear in almost every story or speech on this topic. These misunderstandings are also present in one way or another in most debates over the broader topic of emergency information and communications technology, of which the D Block is only one limited part.

1. Broadband and Internet Protocol will solve the problem”. “Pipes” alone will not solve the emergency response information and communications technology problems. Translating every communication into Internet Protocol won’t cut it. To hear some proponents of broadband, by itself it will reduce medical costs and make us all safer. The national 9-1-1 association, NENA, has the right perspective. It clearly stated in its January 22 letter to Congress that the new broadband stimulus package should not just be transport, but instead ensure that:

“all 9-1-1 centers and emergency response entities have access to broadband networks and the services and applications enabled by such networks. Deploying broadband networks, establishing emergency service inter-networks that utilize such capacity and developing the software applications, information services, and system interfaces required to take advantage of such infrastructure will truly bring emergency communications into the 21st century.” [Emphasis added]

To be sure, the integrated, interoperable emergency communications system needed for our times requires broadband connectivity for the more than 100,000 independent organizations in the US that respond to emergencies (9-1-1 centers, fire services, law enforcement, emergency medical services, transportation departments, public health offices and numerous other support agencies, plus “N-1-1” groups, NGOs like the Red Cross, and a variety of private organizations). We cannot seriously talk about a modern emergency response system unless all these entities are connected to secure IP backbone broadband networks, and ideally have wireless broadband access to extend those capabilities into the field. The traditional focus of first responder communications leaders on voice grade connections will not suffice.

Unfortunately, in the emergency space there has tended to be almost a total focus on transport, generally ignoring the application layer. But merely getting a “fat pipe”, merely getting access to Google and CNN, will not solve the problems of emergency response. Merely connecting all our doctors and emergency organizations to broadband won’t get us interoperable sharing of emergency medical information, including needed parts of electronic personal health records. It won’t get public warnings distributed, or OnStar data delivered usably to the correct 9-1-1 and trauma centers. That doesn't mean transport -- pipes -- is unimportant, but we already have lots of it, in lots of different flavors. Pipes alone can’t move information properly, interoperably in the emergency eco-system. Try playing on the Internet without a browser or domain name server.

As another example, the P25 first responder radio networks being deployed today incorporate broadband IP-based backbone infrastructures and VoIP technology, yet are not commonly capable of interfacing with the Internet or even other P25 networks because they lack the software applications that would support such interoperability.

This imbalance is a very, very big deal because the only way to make rapid progress on inter-domain, inter-jurisdictional, and inter-everything else safety information sharing is to convert every communication into Internet Protocol and then focus on the application layer; focus on what network-centric software things need to happen "in the middle" and with "interfaces from end points to the middle", shifting from focus on the end points (what happens in and at different agencies, or with the devices being carried by their staff). The “transport” focus has meant little to no federal or state government attention and funding of network-centric, application layer solutions and policies – and thus very little progress on overall emergency data interoperability.

2. "A new safety wireless broadband network will solve the interoperability problem." Lack of voice and data interoperability is a huge problem in emergency response. A prominent myth is that the primary reason for creating a wireless broadband access network (or any transport network) is achieving interoperability with other agencies. It doesn't, anymore than building any new network is about interoperability -- unless everyone will be on that network -- which is never going to happen. Right now billions of Federal and State homeland security dollars are being spent to build statewide P-25 radio networks for police and fire (with a national projected cost of over $50 billion to equip just all first responders). However, P-25 is an entirely different air interface than the LTE or WiMax technology likely to be used for a D Block broadband network (for which an additional $15-30 billion is sought). The key to interoperability is getting new and legacy systems (including but not limited to first responder radios) to communicate with each other, not putting everyone on the same network and buying everyone the same radio.

Interoperability is not going to get solved at the transport layer, other than the conversion of communications into IP at gateways. It will be solved first and foremost at the application layer, which IP makes much easier to do. Use whatever radio access works best for you and your agency, but convert the communications to IP with an inexpensive ($1500-$3000) gateway interfacing to a broadband IP safety infrastructure (which could be a combination of dedicated and commercial facilities). Then use the new sophisticated Radio over IP software packages like Cisco’s IPICS and Twisted Pair Solutions’ WAVE to tie it together with my cell phone, and your office wired phone or laptop, Charlie's old VHF police radio with IP, and Susie’s powerful new radio on the statewide P-25 network. And if the device manufacturers (e.g. Nortel, Motorola) will open their APIs, we can make signaling (device control features) interoperable, not just the communications content. The increasing linkage of varied safety agencies: first responders, 9-1-1, local, state and federal agencies to a common broadband infrastructure will provide a foundation for the growth of critical applications and services that will support interoperability.

There are lots of good operability reasons for having safety broadband networks (and P-25 networks); interoperability is not the driver.

3. "A wireless broadband network for responders is critical". Emergency response organizations don't just need a wireless broadband network. They need to be connected to all the broadband networks that are already there: terrestrial fixed (fiber, microwave), wireless and satellite for backup. Why are we only talking about mobile wireless, and focusing only on some of the incident needs of some responders? The primary reason is the balkanization of our decision making in emergency ICT which I have written about at length elsewhere. But even so, this narrow focus is mistaken. The information to be shared over wireless mobile broadband needs to come from and go to lots of places, not just be shared on the wireless link between a dispatch headquarters and its staff in the field. And fixed broadband networks that reach to those other information locations and agencies are already ubiquitous and cheap. Except in a few cases, we don’t have to build new ones; we just need to connect all emergency agencies to the private and public IP networks that exist and provide the network-centric applications that will allow the sharing of information over them (what we call shared and core services). Why are we reversing the model that is so wildly successful in the commercial and consumer spheres (explosive wired internet demand over time created a rapidly growing market for wireless broadband)? Getting organization to organization interexchange of information going over broadband will help create the demand for wireless broadband services for emergency purposes that is so small today.

Moreover, you can't have a regional or national wireless network without a huge wired network underlying it for backhaul and control purposes. Why aren’t the FCC, DHS, and everyone else looking at the synergies from combining these two closely connected, mutually supportive needs? Why is the emphasis only on the access needs of first responders on the scene, and not the need to get information (e.g. building plans) from third party sources to the headquarters so it can be sent out to the field? And what about the information needs (and inputs) of the myriad other organizations (9-1-1, transportation, hospitals, public health, etc.) that are critical in dealing with emergencies, but the primary communications of which are not wireless?


We are trying to foster the evolution of modern safety information and communications technology -- from tens of thousands of local, regional, state and federal silos that exist today towards a “virtual safety commons” supporting the exchange of critical information between these parties. We need to learn from the wildly success models of the commercial growth of the internet and wireless.

Start with a lean wired and wireless broadband infrastructure and the first generations of the key network-centric applications with limited capabilities. We should adapt as much of existing commercial transport and software solutions as possible to both speed the process and keep the costs down. For the transport layer, we should incorporate quickly what works from broadband wireless commercial service providers, wired broadband commercial service networks, and existing dedicated state and metropolitan broadband safety, transportation and other networks. At the application layer we should focus on developing and deploying the necessary shared and core services applications – in the middle -- supporting the messaging and data standards we already have.

Once emergency response folks see the benefits of such open architecture, and their masters see the cost benefits, I believe there will be an increasingly rapid development and deployment spiral. It won’t be a perfect network to start, but we don’t know now what the “perfect network" will be. Even if we did, we don’t have the current demand to support it financially, as the FCC and PSST leaders discovered last year when the D Block auction failed. Let’s avoid the tendency to declare we will build the perfect solution and bring in the consultants to spec the system. Doing that will produce a unique, overpriced and out of date solution, as we learned from the DOJ IWN (national network for federal first responders) project disaster. To reach a vision, we need a series of well founded steps with a regular re-charting of the progress to the vision, and sometimes of the vision itself.

Responses, critiques and suggestions are welcomed.

David Aylward


COMCARE Emergency Response Alliance

1 comment:

COMCARE said...


Thank you for cutting through the hype on this topic. Government, at all levels, cannot afford to build unique and proprietary public safety communications infrastructure or integrated software packages that prevent sharing of data.

I have often pondered whether public safety would be better off if we started requiring all of our IP-based applications to utilize Linux or some other open source software.

I would encourage you to keep reminding people that IP-based transport is important, but only part of the solution.

Thanks again,

Pete Eggimann
Director of 911 Services
Metropolitan Emergency Services Board
St. Paul, MN