Forecasting wireless broadband is a complex proposition

I've had a few conversations now that center around the difficulties in projecting market growth for OFDMA-based mobile broadband systems, namely WiMAX and LTE (Long Term Evolution). It's no cut-and-dry prospect.

Ron Resnick, president of the WiMAX Forum, recently voiced his concern to me over analyst projections that forecast the market for one flavor of OFDMA as being bigger than another. "How are they getting there, because there are so many variables?" he lamented. That's why the WiMAX Forum plans to release a commissioned report soon, taking into account a number of variables.

What are they? In the mobile voice world, projections are relatively easy to make as most operators have deployed similar voice networks with similar coverage. But the mobile broadband world will be defined by the fact that not every operator will deploy their networks in the same way. Some will deploy fixed networks, others nomadic and still others fully mobile. What type of deployment an operator chooses has a bearing on capex, how many subscribers it will attract and what its return on investment could be. Frequency bands certainly play an important role given the fact that an OFDMA network would require more cell sites at 3.5 GHz than 700 MHz. Moreover, what sort of business plan are mobile broadband operators embarking on? A simple access model? VoIP services, mobile TV? How much spectrum does an operator have? Then there are technical considerations such as MIMO and smart antennas. The list goes on.

Michael Thelander, head of Signals Research, and a crew of experts have spent 15 months tackling these variables. His firm has released a new economic study evaluating mobile broadband [1]. Rather than predicting any technology winners or subscribers, the firm plugged these different variables I mentioned to come up with economic feasibility of deploying OFDMA-based systems. Signals Research studied wireless broadband across 65 countries, eight frequency bands and the various flavors of OFDMA. The firm then plugged in different variables, such as nomadic vs. fully mobile deployments, to determine an operator's network economics.

For example, if an operator deployed a truly mobile broadband wireless network that is capable of deep in-building penetration to a handheld device at 2.5 GHz, covering the top 70 percent of the population would require about 11 times more coverage of cell sites than a network that was designed to provide semi-fixed coverage and nearly 20 times more coverage cell sites than a network that was designed to provide fixed wireless services, Thelander said.

Of course, just because it's cheaper to deploy a fixed network, doesn't mean every market can support one. The North American market has a limited market for fixed or semi-fixed networks, while a market like India, with low broadband penetration, could easily find success from such deployment scenarios.

"Given our analysis of existing telecom demand, combined with these metrics and their underlying economic implications, operators in developing markets may be better off initially designing their networks for nomadic or semi-fixed services and then grow into higher degrees of mobility as subscriber uptake increases and they begin to add new cell sites for capacity purposes," Thelander said.

The bottom line is the greater the risk, the greater the return. Operators taking the most risk are those deploying a fully mobile WiMAX network, such as Sprint, because the industry has yet to see a dramatic swing in mobile broadband subscriber uptake and usage patterns. "There has to be something that drives a lot of data," Thelander said.

"One could argue that with the introduction of a new broadband wireless network that all of a sudden there will be a dramatic swing in subscriber uptake and usage patterns," Thelander said. "We don't disagree, but if we use the demand that exists today in these countries, or even multiples of what exists today, as the frame of reference for predicting future adoption on these new networks, then fully mobile networks would be largely under-utilized and, as a result, network economics would suffer."

Clearly, the mobile broadband world has become inherently more complex.--Lynnette [2]