Pick any band you want, and let the analyzer run over a period of time. Compute the average amount of energy over the sampled period of time. What do you find? That the airwaves are mostly empty most of the time. We don't, then, have a spectrum shortage. Rather, we're just not making very good use of this precious natural resource.
Why is that? Well, the answer is rooted in the history of wireless. In the early days, radio was commonly used for broadcasting, and, after quite a while, for mobile communications of the push-to-talk variety. Thus it appeared that we had tons of spectrum, and the government allocated spectrum in a land-rush kind of fashion -- big chunks went to those who could make a case for having these.
Today, the situation looks quite different. We have more demand for applications than we do spectrum, and it now behooves us to think more intelligently about how that spectrum is allocated and used. It makes little sense to allocate spectrum to applications that leave that spectrum fallow much of the time. Sharing is the name of the game, just as it is on the highway.
But there’s another problem. Historically, specific radios were of necessity designed only for particular frequencies and particular applications. Modern technology removes these limitations. One of the key steps in this direction was the development of software-defined radio, or SDR, in which many physical-layer functions are configurable in software. Imagine a phone that can work on both CDMA and GSM networks (or receives TV broadcasts) and does so by loading new software.
As it turns out, this kind of flexibility is unlikely to be as important as we originally thought five or ten years ago. We seem to be converging on LTE and Wi-Fi, plus some legacy Bluetooth, limiting the utility of SDR in handsets. Of course, essentially all radios today have a significant software component, but SDR-class flexibility isn't really a requirement. Except in one dimension -- namely, the frequency upon which the radio operates. Being able to support a broad range here, say from a few hundred megahertz (TV channels) to a few gigahertz, would be ideal. Also, wouldn’t it be nice if that radio could automatically (algorithmically) select the right frequency for any given moment in time, thus intelligently sharing the airwaves? If we thus throw a little intelligence at the problem, we can in fact make better use of that otherwise scare resource, radio spectrum.
The first big application for these cognitive radios is in the so-called TV White Spaces, unused television broadcast channels recently made available by the FCC. A variety of mechanisms, including spectrum sensing and the use of a database of occupied spectrum, are required to make the cognitive radios truly smart.
While controversy continues to swirl about the White Spaces effort, as is always the case when spectrum is re-assigned, this spectrum is going to become available and thus will serve as the proving ground for cognitive radio technologies. My hope is that we won't stop there. Cognitive radio technologies are the key to handling the enormous increase in traffic and demand which will unquestionably materialize as communities of users -- personal, corporate, and government -- all around the world continue to rely on wireless communications as, quite literally, a key to prosperity.
Craig Mathias is a Principal with Farpoint Group, a wireless and mobile advisory firm based in Ashland, MA. Craig is an internationally recognized expert on wireless communications and mobile computing technologies. He is a well-known industry analyst and frequent speaker at industry conferences and trade shows.