This article originally appeared in the March 2011 issue of NGN.

Everyone is hyping 4G, but what’s really happening?

Originally, the ITU defined 4G as a network delivering 100mbps to mobile users and 1 gbps to fixed users. This means significant new technology compared with 3G. As these 4G technologies were developed, anything using 4G technology came to be called 4G, regardless of its current performance; thus, today’s WiMAX and LTE are called 4G. Recently, AT&T’s (News - Alert) marketing department went even further, calling its advanced 3G technology a 4G service. It’s not clear that one will stick.

The next point about 4G is it’s all IP. That means traditional voice services don’t work. New specs for voice over LTE (News - Alert) are in development, but widespread deployment is at least three to five years off. Until then, a 4G handset isn’t really a 4G telephone. It will fall back to 3G to make voice calls or send an SMS.

The big issue is data performance. Verizon (News - Alert) won Block C in the 700mHz auctions giving them 22mHz of nationwide bandwidth (746-757 and 776-787mHz).  In LTE terms, this is 10+10mHz of spectrum. With 10mHz of spectrum for downstream (to the mobile device), today’s LTE gear can reach a theoretical peak bandwidth in one sector of 59mbps. However, this would be to a single ideal device located very near the cell site. The real world is much more complex.  For a reasonable distribution of customers, the average capacity per sector, shared among those customers, is likely to be 16 to 18mbps.

Typically, an individual’s demand for data is highly intermittent. Thus, one Verizon LTE sector easily can support 40 to 50 people browsing and doing e-mail, with each person getting Verizon’s advertised 5 to 12mbps of apparent performance. Of course, if too many people start streaming Netflix movies to their handsets, performance will collapse.  

As a platform for streaming video, 4G will approach the performance of home DSL service, if there are only five to eight Netflix users per sector. Indeed, when Netflix published actual sustained streaming data rate averages for 16 different national ISPs, the only 4G service with a statistically significant number of subscribers (Clearwire (News - Alert)) came in just below the poorest DSL provider. Verizon’s LTE gear is a couple of years more recent than Clearwire’s WiMAX gear, so Verizon has a chance of being competitive with consumer DSL, but no more.

In summary, emerging 4G services provide a significant performance increment while 4G infrastructure costs significantly less than 3G for equivalent capacity, so 4G deployments are a no brainer. But 4G can’t approach the capacity of fixed Internet access networks. It’s only lame competition for the duopoly. So let’s look forward to widespread 4G mobile services, but go easy on the hype.