My prediction is that LTE and WiMAX are toast. The new great thing will be WRANs (wireless regional area networks). WRAN’s will extend and eventually subsume WiFi.

The detailed regulations which implement the FCC decision to free the spectrum formerly known as TV white spaces have now been released. They look pretty good from the point of view of someone who believes the unlicensed use of this spectrum has the potential to make a huge difference in the way the world communicates.

Once radios are built to these regulations and get through FCC testing, wireless broadband at cable speeds—10 Megabits per second (Mbps) downlink initially and increasing—ought to be rapidly available at competitive prices in more and more of the country—especially in rural areas where good broadband is often hard to get but where white spaces abound. Give this just a year to start having an effect.

Slightly longer range but in about two years both mobile voice and data will be widely available at a quarter of the price of today’s prices for the two combined in “unlimited” offers. The download speeds for mobile will be better than 5 Mbps. If I’m right (many knowledgeable people do and will disagree with me), LTE is DOA and WiMAX will be eclipsed before getting to critical mass. That’s bad news for at&t and Verizon who are betting on LTE and for Sprint with its bet on WiMAX.

Here are some of the highlights of the announcement:

Fixed radios in this spectrum are allowed a power output of up to 4 watts EIRP (effective isotropic radiated power, if you really wanted to know). This is the same power output as the most common radios used by wireless ISPs (WISPs) today BUT these are usually operating at much higher frequencies so they have an effective radius of less than 10 miles and don’t go through leaves, trees, and walls very well. Operating in the white space frequencies should give these new radios an effective radius of up to 30 miles. That’s a huge economic difference to a WISP installing radios.

Personal/portable devices are restricted to 100 milliwatts EIRP (a milliwatt is a thousandth of a watt). I would have liked to see that number a little higher but it’ll do for starters. A rough calculation with some help from my friends and a good tutorial by Barry McLarnon indicates that, in good conditions, these should be able to communicate in about a five mile radius of a fixed radio acting as a base station. These personal/portable devices are going to end up in mobile phones, in USB dongles, and built in to laptops, PDAs, and GPSes the way WiFi is today.

For the sake of comparison, 100 milliwatts is about twice the power of most laptop WiFi radios today and the power goes further than WiFi because of the characteristics of the spectrum. On the other hand, this is much less power than mobile phones are allowed to use and mobile phones do operate in similar frequencies. On the third hand—soon I’ll be an economist—mobile phone usually operate at 100 milliwatts or less to preserve their batteries and avoid interfering with each other even though GSM phones are allowed a full 2 watts.

Since these devices are portable, they don’t have to be installed. Going out to houses and installing radios is a huge expense for WISPs. If they can just mail you a cheap radio or count on it already being in your laptop, the price of wireless data goes down and its utility goes up because you can wander around the house, yard, or neighborhood with it. Soon (couple of years+) roaming agreements’ll let you go everywhere with this connection. This is the way of the future but a few more milliwatts allowed power would help.

The fixed devices and some of the mobile devices need to know here they are. This can be accomplished with a GPS chip or an installer of a fixed device telling the device where it is; portable devices which act as clients to smarter devices don’t need to know where they are; otherwise they, too, need GPS capability. GPS chips are cheap so this shouldn’t be much of a problem. The devices communicate with a geo-database (which you can bid on supplying to the FCC if you want) to make sure they are not using frequencies which are in use by licensed TV stations or wireless microphones; the devices also have to register their own locations and characteristics in the database and can be shutdown or ordered to shutdown if they misbehave. All of the devices need to listen for licensed users or wireless microphones which they might be interfere with and leave them alone. This is an intentional belt and suspenders approach.

How “portable” can a fixed device be? That’s an important question since fixed devices are allowed forty times as much power as portable ones. The regulations just define them as “A TVBD [nb. TV Broadband Device] that transmits and/or receives radio communication signals at a specified fixed location. Fixed TVBDs may operate as part of a system, transmitting to one or more fixed TVBDs or to personal/portable TVBDs.” They obviously anticipate they might move or they wouldn’t need to have a GPS in them. The receive antennas for fixed devices, however, must be outdoors and at least 10 meters above the ground while the transmit antennas may not be more than 100 meters above the ground. The latter could be a problem in an area of high trees.

So, you ask, what standard will all these devices follow? The FCC paid a fair amount of attention to the 802.22 project of the IEEE although they did not completely defer to it. NO standard is mandated which is a good thing; there is plenty of room for innovation. More on 802.22 in an upcoming post…