Reading through the very positive story in Telephony today about how AT&T is “doubling” its network capacity, and wondering where reporter Kevin Fitchard — usually pretty solid in things telecom — gets the numbers to back up his assertation that AT&T “has reams of cellular and PCS spectrum” that it could throw at digital data should its 3G networks clog up more than they already are now.
For one, I don’t believe that AT&T has “reams” of spectrum, at any bandwidth that can quickly be reassigned (or tailored for new uses with equipment to match). What I wish Fitchard had asked AT&T is how big are AT&T’s channel sizes per cell site, and how many cell sites does it really have in its 3G/4G network?
(Without getting too technical, the “channel” we are talking about is the size of the wireless “pipe” being sent from tower to user; the bigger the channel, the more data and capacity that can be sent per site. Current cellular implementations typically use 1.25 MHz channels, which accounts in part for their limited capacity. So-called “4G” technologies like Clearwire’s WiMax implementations use 5 MHz and 10 MHz channels, for example, to help support faster download speeds. But even AT&T’s “reams” of 700 MHz spectrum only have around 20 MHz of depth in any market — limiting the channel size to 5 MHz in most implementations, which means even LTE deployments won’t be as fast as they are often advertised.)
It’s all well and good to talk about increasing the download-speed capacity on AT&T’s high-speed packet access (HSPA) network to a theoretical number of 7.2 Mbps, but that number is useless if it’s not backed up by spectrum depth, which is the thing that matters when it comes to whether or not a network delivers as promised. Unless AT&T comes through with some real numbers, reporters like Fitchard and his readers would do well to question why those “reams” of cellular spectrum aren’t already being used to ease the pain of frustrated iPhone users.