Nortel's CTO weighs WiMax plans

13.10.2006
John Roese, who became chief technology officer at Nortel Networks Inc. in June, discussed WiMax wireless technology -- and Nortel's plans for it with both businesses and carriers -- this week after delivering a keynote address at WiMax World in Boston. Excerpts from that interview follow:

How is Nortel fitting WiMax into its technology strategy? The transition from 3G [third-generation wireless] to 4G is interesting to us, because the 4G technologies, including WiMax, are based fundamentally on MIMO/OFDM [Multiple Input Multiple Output/Orthogonal Frequency Division Multiplexing] technology. Nortel spent the last seven years pretty much inventing that technology. It's a logical fit.

Nortel knows how to operate in a carrier ecosystem, so WiMax is the way to go. We have trials with 802.16e equipment [using the mobile WiMax standard] this fall, and Sprint Nextel expects to have networks up and running by this time next year.

Sprint is using Nortel gear? Sprint will use a lot of different gear. Our trials are with network carrier infrastructure gear in Asia and North America.

So how will Nortel's WiMax affect the large business user? Let's say our vision of the 4G world pans out. We see 4G accelerated, and its cost per bit is one-tenth of what 3G can provide. It's very economically feasible. The base station design, the spectral efficiency -- all of those are very superior to 3G from a data services perspective. That's pretty exciting.

So suddenly there's this concept of high-performance, pervasive broadband. It changes the enterprise's view of the broadband wireless network. Today, they view it as the cell phone with some added features on it. Tomorrow, they start to think about it as, maybe, a logical extension of [the] enterprise infrastructure. So applications that heretofore have lived principally inside the enterprise in the LAN or have leaked out through a VPN suddenly become mobile applications.

In order to make that happen, there are challenges. Clearly, carriers have to operate on enterprise sensitivities. Nortel can drive 4G because Nortel already has a multibillion [dollar] carrier business and a multibillion-dollar enterprise business, and we understand the two of them. Clearly, for enterprises to extend their applications out into the carrier network cloud, the cost has to be right -- such as the cost for outfitting everybody in your mobile organization with data services. So $100 a month per mobile worker isn't going to work.

We also understand that you can connect machines to the network, with a machine-to-machine interface, because it will be more cost-effective with 4G. So, the idea of telematics coming from your truck fleet works. Today over cellular, it's very expensive. Imagine instead you have WiMax connectivity in every vehicle and that vehicle provides high-speed connections over this broadband wireless network. So you can have real-time telemetry from your trucks. If you are an insurance adjuster, you can stream video back to somebody to look at a car that's been in an accident.

Well the insurance adjuster sending back wireless data has been around a while, so what's new here? It's been out there as an idea, until you start to calculate today's cost with EV-DO wireless. People rarely ask what the cost is, and it's probably $100 a whack (per user per month) to send the data wirelessly. That's pretty expensive per insurance adjuster. And ask yourself, If this idea is so great, why aren't all the insurance companies doing it? What we've found is that the other ones are facing an economic barrier.

And maybe wireless network coverage is poor? The cost is too high, but more importantly, the coverage of these wireless networks is not comprehensive. 4G is cheaper access, but also cheaper infrastructure, which can give you a broader coverage area. It's Metcalfe's Law, which says the value of a network is equal to the square of the connected users. I happen to believe in that. I'm excited about 4G and WiMax because it might drive more subscribers to a bigger network, bringing in more nonsubscribers.

To clarify, WiMax is licensed spectrum that you get through a carrier? You couldn't do this as an insurance company or other enterprise on your own. It is licensed spectrum you get through a carrier. Sprint says it will have 100 million people covered in a year or so with WiMax, which is pretty dramatic. 4G to Nortel is not about any one wireless technology, because there will be wireless mesh, Wi-Fi, outdoor Wi-Fi and WiMax. They will interact through gateways.

So, with WiMax and 4G wireless, Nortel will be sort of a Switzerland between enterprise users and carriers? We'd like to be that Switzerland with technology that bridges the gaps, as an enabler of the enterprise to consume 4G and also to work with carriers to present a set of services they sell to the enterprise. It means that enterprises should plan for more than just one new IP phone line per user because of all the other data that will be coming from machines, such as with a sensor network, and the Internet over wireless broadband.

I'm not sure the average enterprise understands why WiMax matters. It matters because it cost effectively extends broadband connectivity. We're excited about it.