The Seattle Times writes about the coming U.S. air-to-ground spectrum auction and its potential outcome: I've been writing about this for several months, and it's just starting to get covered in the mainstream business press as the full impact of a mere 4 megahertz of bandwidth comes clear for domestic flights.
This article notes how Connexion's expensive and not widely deployed service--although it continues to grow at a good clip--will likely have no uptake in the domestic U.S. market because of this spectrum that will become available using cheaper ground stations rather than satellite for the backhaul to the Internet. Boeing estimates it could gross $500,000 per plane per year with its satellite service; no numbers were presented for domestic flights, which will likely cost less for usage but have a higher uptake.
This article doesn't mention two interesting factors in this auction, scheduled for May of this year which I documented in an article I wrote Dec. 9. First, AirCell already operates a national ground station network for general (non-commercial) aviation for data and voice, and they will almost certainly bid on this spectrum. They qualify for bidding credits under the auction rules, too.
Second, Verizon states in this article and has stated elsewhere that they could have their data service running within a year of winning the auction, but the FCC order that settled auction terms several weeks ago state that Verizon has two years from the end of auction to decrease its use of current spectrum for AirFone. The order states, "Airfone must transition its incumbent narrowband operations from four to one megahertz of spectrum in the band within two years of the initial grant date of a new license in the band."
If confirmed, it means that Verizon has a full year financial advantage for starting up revenue over any competitor, even if a competitor wins one of the two overlapping 3 MHz chunks up for auction in one bidding configuration. I'm not sure how this is the case because the company made a compelling set of arguments accepted by and documented by the FCC in its order that it needed a full two years to upgrade AirFone equipment on thousands of planes, some government owned and operated.
(There are three possibilities: 1 + 3 MHz, 3 +1 MHz, and overlapping 3 MHz; the latter is the only one that makes sense for bidders, but Verizon could game the auctions by bidding exceptionally high for a 1+3 or 3+1 configuration, which would exclude bids for the overlapping spectrum. The 1 MHz chunks aren't wide enough to provide the bandwidth necessary for profitable in-air broadband.)