Business

Build With LOS Today,
Transition To NLOS Tomorrow

Spike Broadband says FUD factor impeding non-line of sight technological advance is no reason to dally. Stick with LOS technology and get building today.

by Gerry Blackwell [October 16, 2001]

Should fixed wireless carriers wait for highly-touted non-line of sight (NLOS) equipment to come to market and mature? Or make the most of the line of sight (LOS) gear available today?

Most carriers have already made their decision, says fixed wireless equipment vendor Spike Broadband Systems of Nashua, NH. They understand they must continue to deploy LOS equipment in the short term.

But Spike is concerned that the hype around NLOS technology—which is not expected to be available for several months yet—is creating "fear, uncertainty and doubt" (FUD) around fixed wireless.

And that could harm the entire industry at a higher level.

"Fixing the wireless business model does not require waiting for a pure non-line of sight solution," argues Cristian Parrino, Spike's vice president of marketing. "The fix is in technology evolution, not a revolution."

Waiting game
Spike develops and markets network equipment for carriers using sub-10GHz licensed spectrum. Its largest customer, announced earlier this year, is Sonofon, a Danish joint venture between Telenor ASA, the former PTT in Norway, and BellSouth International.

Sonofon is deploying a hybrid fixed and mobile wireless voice/data network, using Spike infrastructure equipment for the fixed wireless portions. According to Spike, it's the largest fixed wireless deployment in the world. The company has already delivered 180 base stations.

Parrino insists Spike is not making its seemingly anti-NLOS pitch because sales of its LOS equipment are slumping. Most carriers, he says, are moving ahead with LOS technology—witness Sonofon. They have to, he says. It's all that's available. "And time to market is a hard requirement."

But the idea persists that fixed wireless needs NLOS technology to be viable—perhaps because MMDS carrier Sprint Broadband Direct is widely reported to be delaying deployment of new markets while it waits for the NLOS equipment it hopes will improve the economics of its business.

Sprint is about the only carrier that can "afford to wait," though, Parrino says.

So if sales are unaffected, what's Spike's problem?

"It's more a financial markets and public relations concern," Parrino says.

"The primary issue is that the financial anlayst community's confidence in the fixed wireless space is a lot lower if they believe that a technology breakthrough is required for the market to take off."

And no, it's not that Spike is frustrated at being unable to find venture capital, Parrino hastens to add. The company is well funded, he says. It's more the market-chilling effects of the FUD created by NLOS technology.

Unspecified migration
The two-fold case for NLOS is compelling to be sure. First, it will increase penetration to include customer premises that don't have line of sight to base stations—thus making the market universe bigger.

Second, because NLOS equipment doesn't require careful aiming of antennas, customers can install it themselves. This will eliminate carrier installation costs, which according to some estimates mount to a crippling $100 to $150 per customer.

It will also make it possible for fixed wireless carriers to move to a retail model, further reducing distribution costs and broadening the market.

It's not that Spike doesn't accept these arguments.

"What we believe—what customers are teaching us—is that for the industry to come to volume, we have to rely on current line of sight equipment," Parrino says. "But to stay at volume, non-line of sight will have to come into play."

Spike is proposing an evolutionary strategy that involves first installing LOS networks, then integrating NLOS pico cells. It's an architecture borrowed from the cellular world, Parrino says, but unique in fixed wireless so far.

Spike's argument for a hybrid LOS/NLOS approach makes sense to us.

First, as Parrino points out, the penetration possible with LOS equipment typically runs between 60 and 75 per cent of the within-range population. It will take two or three years for most carriers to saturate the market created by establishing a LOS network.

In the case of Sonofon, because of favorable propagation conditions in Denmark, the penetration with LOS equipment is about 70 per cent, and it is estimating it will take three years to exhaust the market.

By then, Spike expects to have NLOS equipment it can deploy in pico cells. Each access point will be connected to a LOS customer premises unit and provide very short-range NLOS access to perhaps 15 or 20 additional customers.

The LOS network in effect will provide low-cost backhaul for the NLOS cells.

Not only does the Spike approach allow full-speed-ahead deployment of current, proven LOS equipment—thus satisfying most carriers' "time to market" requirements—it also avoids a flaw in the pure-NLOS model.

Building markets from the ground up with NLOS equipment will mean using much higher-capacity and higher-powered base stations than are currently used in LOS networks, Parrino points out.

As a result, they will cost an estimated four times as much as LOS base stations. And that clearly will negatively offset the savings on installation costs.

The much lower-powered pico cell NLOS base stations Spike is proposing will bring the cost per customer into the $300 range—down from $1,100 to $1,200 under a pure NLOS architecture, Parrino says.

We obviously have no way of corroborating the Spike math on this, but the logic seems reasonable.

Big chill for Wi-Fi ?
Can Spike deliver the NLOS pico cell technology it's talking about? "This idea of a transitional architecture is what is driving our R&D," Parrino says.

He says he expects Spike to partner with one or more of the "next-generation" equipment vendors currently making so much noise about the NLOS technology they will eventually bring to market.

What does all this signify for the average regional ISP or CLEC operating in license-free fixed wireless spectrum? It's not clear.

Equipment vendors with products that work in unlicensed spectrum appear to have a significant headstart with NLOS.

As we've reported here, vendors such as Toronto-based WaveRider with its 900 MHz LMS3000 product line already have ISP customers deploying first-generation production NLOS technology in the field.

But if Parrino is right about a funding chill around the whole idea of fixed wireless, there could be a trickle-down effect.

Besides, in the U.S., there are lots of regional players with MMDS spectrum just starting to come out of the woodwork. These issues will affect them directly.

Even if Spike isn't ultimately the one—or the only one—to provide the solution, the basic approach it's proposing sounds right.

—End