Once upon a time, Internet access was limited primarily by the "last mile," the copper wires that brought a connection into a home or office. Now, in an era when connectivity is increasingly essential for mobile devices such as laptops and smartphones, wireless Internet will power the next generation of broadband expansion.
Envision this for a second: unlimited wireless broadband, everywhere, for pennies on the dollar of current solutions. Wi-Fi-like service that covers not just your home or office, but virtually anywhere that you care to go.
No, I’m not talking about Verizon’s EV-DO network. While it has its virtues, EV-DO is incredibly restrictive about the devices that you can use, not to mention expensive and limited in footprint. And as long as cellular providers control the network, we can’t expect that kind of service to reach a true saturation point. The use of unlicensed, cheaply operated, and heavily commoditized technology akin to Wi-Fi is the key to providing an inexpensive and robust network with nearly unlimited bandwidth. And make no mistake: we need that bandwidth.
EV-DO cards are nice, but service is incredibly restrictive and expensive. They are not the answer to a wireless future.
More bandwidth please
Most of the exciting mobile technologies of the future need more and more speed to be practical — things like IPTV, on-demand video downloads, VoIP, video conferencing, etcetera. And that bandwidth needs to be wireless. Fiber optics are the ultimate solution for nearly unlimited bandwidth, but they’re never going to reach full deployment. It costs tens of thousands of dollars per mile to deploy fiber, and there are simply too many areas where companies like Verizon would never make enough profit to convince them to install the cables. Even if you had a nationwide mandate for 100% fiber deployment, the way that phone lines were mandated in the last century, those fibers are only as good as the points to which they’re deployed. You can’t drag a fiber cable around hooked up to a mobile device, and cellular alternatives aren’t going to cut it in the long term.
What we need is a truly broadband wireless network to fill the gaps, and to provide enough bandwidth for mobile devices. And just such a thing may be lurking not too far over the horizon.
Goodbye analog TV, hello more bandwidth?
Currently, the FCC is considering the public availability of dozens of radio frequency slots located in between television channels, in the 150-700 MHz range. These new channels would become available after the planned shutdown of analog TV in 2009, and would allow unlicensed radio communication, such as wireless Internet services, to take a seat in one of the more desirable frequency ranges for combining range and bandwidth. You could do this with other frequencies as well — a considerable amount of space in the 2500 MHz range is expected to be used for Sprint’s WiMax network, for instance. But the lower the frequency, the greater the range it has, and the better ability to cut through obstacles, which makes the space between TV stations ideal.
In 2009 your old analog TV will no longer work in the U.S. as the FCC reclaims those airwaves (view large image)
With a relative minimum of hardware, these new channels could deliver wireless Internet to mobile devices at speeds up to or beyond 4 megabits, and ranges up to 25 miles or more. With that kind of range, a single tower could cover most of Rhode Island. And operating in unlicensed radio spectrum, hardware would be quickly commoditized, ala Wi-Fi, keeping setup costs relatively low.
With those channels, and the proper hardware to take advantage of them, you could cover pretty much the entire continental United States with less than 4,000 towers. Even assuming an equipment and setup cost of $100,000 per tower, a high-end estimate, the total nationwide network would cost about $400 million dollars. That’s less than a tenth of what T-Mobile alone paid for radio spectrum licenses to deploy slower 3G solutions, let alone the billions more they’ll have to spend on hardware. A more limited network covering only major metropolitan areas could probably be done for well under $100 million. While unlicensed radio does mean more need for caution in avoiding interference, it also means that virtually anybody with even a few thousand dollars to start one tower can compete in a market. That in turn means more choices, better prices, and fewer limitations.
There would be trade-offs to such a proposed system, of course. As range increased, speed would decrease. Measures would have to be taken to prevent base stations from interfering with each other. New hardware would have to be designed, tested, certified, and commoditized. And of course, this is all operating under the assumption that the analog TV shutoff goes as planned in 2009. But either way, the real future of the wireless Internet lies in an unlicensed, pure network solution, with cheap hardware and cheaper airwaves.