I think I've mentioned before that our goal for wireless is to minimize, if not eliminate, the behavioral and performance differences that separate wireless and wireline. In other words, we want to offer the same services on wireless that we do on wire -- voice, broadband, video -- everything. We've made great progress in all of this over the past 15 years, and things continue to improve.
Still, there is one fundamental difference between wireless and wireline that is very, very difficult to address indeed. And that is spectrum.
I'll define spectrumhere as that portion, or band, of the electromagnetic spectrum reserved for a given activity, such as cellular, RFID (radio frequency identification), GPS (Global Positioning System), television or wireless LANs. One should think of the airwaves as a collection of bands whose specific behavior is a function of frequency rather than as one big wire in the sky. The entire electromagnetic spectrum (that one big wire) itself is a property of the known universe that allows electromagnetic waves to be generated at one location, and then, given the right conditions, to propagate from transmitter to receiver with enough signal strength and quality that communication can take place.
The physical process we use for radio-based communications isn't all that different from what we do on wire. Those are indeed electromagnetic waves propagating through wire, cable or fiber. But wire, cable or fiber encapsulate a nice, neat little world for the waves to travel in. Put a signal on one of these, and in most cases, it goes to the other end without difficulty. Outside interference usually isn't a problem. Breaks in the cable itself are much more of a concern.
But while the air doesn't break, the airwaves are subject to all manner of considerations that aren't issues for the wired world. First of all, there's no guarantee that a given signal will in fact make it from transmitter to receiver. The geometric relationship or antenna orientation between the two might be inappropriate (especially when mobility is a factor), and there could be interference, fading of various forms and assorted other artifacts of radio-based communications that come into play.
One of the key issues is the inherent propagation characteristics of a given band of frequencies. Some frequencies, like shortwave radio (3 to 30 MHz) bounce off the surface of the Earth and the ionosphere in the atmosphere, and thus can propagate quite literally around the world. The microwaves (1 GHz and above) used in most telecommunications applications don't -- they are very line-of-sight in nature. Thus, regulators, such as the Federal Communications Commission, assign various services and applications to particular frequencies as best they can; some parts of the spectrum, depending upon application, really are better than others.
This led to the observation, about 15 years ago, that certain bands of the spectrum could be quite valuable to certain users. There is always demand for more spectrum, of course, but certain bands could indeed prove to be very popular. A great example is the PCS bands at 1,850 to 1,990 GHz, which are excellent for cellular-type communications. These bands are also the first big chunk of spectrum to be allocated to specific users, not by the lottery process previously used, but rather via an auction process. In other words, under the auction doctrine, the really good parts of the spectrum would no longer be free; a big chunk of cash would be required upfront from those intending to offer services in certain bands.
This philosophy was mostly driven by a U.S. Congress hungry for revenue and averse to directly raising taxes. The strategy fits in with the general idea of using fees and other nontax means to raise government funds. Now, how one feels about this idea is, of course, a function of one's personal politics, but two arguments can be easily made regardless. First, auctioning spectrum provides a powerful incentive for the government to make more spectrum available. More spectrum means more users, higher data throughput and greater reliability. On the other hand, someone has to pay for spectrum auctioned through a process that is designed to maximize revenue. And ultimately, that someone is you and me, the subscribers.
The concept of spectrum auctions is global, with hundreds of billions of dollars raised for government coffers around the world. We have another really big auction coming up shortly. It's called the AWS-1 bands. Don't get too excited about the name "AWS" -- it's just a name, and almost anything is possible in these bands. But it is a very nice chunk (actually chunks) of spectrum that will prove to be quite popular, perhaps even with bidders beyond the traditional cellular carriers. We'll look at the auction process in more detail and at AWS in particular next week. The outcome of these auctions could have a profound effect on the future of wireless services available in this country.