The 700MHz Question: Will the Wireless Spectrum Auction Lead to Innovation or More of the Same?

Date: Sep 28, 2007

Article is provided courtesy of Prentice Hall Professional.

The FCC will soon be auctioning off the rights to use the 700MHz spectrum for wireless communications, with the winner being able to choose the direction of wireless services development in the US. With stakes this high, is the playing field fair, and are business needs trumping consumer and technological interests? David Chisnall examines the players, the playing field, and prize in this complex competition.

Analog television has been around for longer than most InformIT readers. Standards like PAL and NTSC were designed to maintain backwards compatibility with older black and white equipment, and even today it's relatively simple to turn an oscilloscope into a simple monochrome TV receiver.

This is about to end, however. Analog TV has been kept around for a long time based largely on the huge installed base. Gradually, it has begun to be replaced by digital signals, which make more efficient use of the spectrum by heavily compressing the signal.

As with so many other things, one of the primary motivating factors for this migration is money. Since digital TV uses less bandwidth for the same number of channels than analog, this leaves some spare. Since the use of the spectrum in any given country is heavily regulated, this migration gives governments a chance to grab some cash by selling off the rights to use the newly-freed space. In the US, the 700MHz spectrum is beginning this process soon.

Speed and Distance

Picking a good frequency for a particular application is difficult. High frequencies have a greater information carrying capacity. An encoding used to transmit data typically has a fixed number of bits per cycle. On a pure, noiseless, square wave, this would be one; the wave is either up or down, giving a one or zero. More commonly, the amplitude is varied in smaller steps, giving more than two possible symbols per cycle. All other things being equal, however, doubling the frequency (that is, doubling the number of cycles per second), will double the data rate that can be transmitted.

If this were the only factor, higher frequencies would always be better. Unfortunately, there is a conflicting requirement. Higher frequencies mean shorter waves, and shorter waves are much easier to block. During the cold war, submarines typically used 50-100Hz signals to communicate while submerged. These can pass through the Earth without any problems, while the 2.4GHz signal from an 802.11 device has trouble passing from one room to the next.

The 700MHz region was chosen for analog television since it was capable of carrying enough data for analog television while having enough range to cover a country with signal and still be economically feasible. This also makes it attractive for mobile data services, since it requires a fairly low investment to provide good coverage.

License Poker

Spectrum licenses are not new, and have a number of problems associated with them. The winner of an auction typically gains a monopoly in a few bands in a certain geographical region. This is by design, since there needs to be regulation over the devices using a particular portion of the spectrum to ensure that they don't interfere. Since most deployments use a client-server architecture, with consumer devices receiving signals from central towers, ensuring that only one entity will be building the towers makes this relatively easy to do.

The big downside to this is that it turns the auctions into a classical prisoner's dilemma. If everybody bids for just the amount of spectrum they need, at a low price, then they will all get a good price. If, on the other hand, one company can raise the bids enough to exclude one or more of their competitors entirely. If this works, then they have a total monopoly. In practice, it rarely does work, but this doesn't stop companies from trying. In the UK, we saw this with the 3G spectrum auctions, where the winning companies ended up spending so much on the licenses that they couldn't afford to spend much deploying the infrastructure.

Some spectrum auctions try to avoid this. Rather than offering the spectrum to the bidder with the most cash on the table, the regulator sells it for a fixed price to the company willing to invest the most on infrastructure, provide the cheapest service to consumers, etc. This is not the case with the current spectrum auction in the USA.

I Want to Be Free

The current mobile market in the USA is regarded as something of a joke internationally. Features that are standard elsewhere are disabled by networks in order to sell their own proprietary services. The lack of wireless standardization has also lead to much greater difficulty in moving between providers than in the rest of the world (where you just need a new contract, not a new phone).

This is a problem for companies such as Google. Google's business model relies heavily on bandwidth being cheap. Ubiquitous broadband has helped Google a lot, especially since their purchase of YouTube. They depend on their customers having unfettered access to their content. In the wired market, and assuming network neutrality, they have this. The mobile market is much less certain.

Until very recently it was very uncommon to have a mobile device capable of anything close to a desktop browsing experience. The screens were small and the bandwidth was nonexistent. Now, with technologies like UTMS, the bandwidth issue has gone away (although latency remains an issue). The screens are still smaller, but they often have higher resolutions than desktops (225dpi is not unheard of) and more interesting input mechanisms. Mobile devices with 2-400MHz ARM CPUs and 64-128MB of RAM are starting to be relatively common, and these are capable of rendering even quite complicated web pages relatively quickly.

Unlike wired ISPs, however, mobile phone operators are still attached to the business model of selling services rather than access. Things like instant messaging are a direct threat to their business model of selling SMS at exorbitant per-byte rates. Voice over IP upsets their business model of charging according to the distance of the call.

Services like YouTube, similarly, are a direct threat to mobile operators' ability to charge a premium for access to video. When it comes to mobile Internet, a lot of Google products compete directly with services that the network operators wish to provide themselves. It's very difficult to compete for online services when the other player controls both the network and the client device. When Google competes with a company like Yahoo!, the playing field is level. When they compete with MSN, they have to fight against Microsoft's ability to automatically install their software on the majority of desktops. When they compete against a mobile operator, they have to fight against their competitor's ability to throttle or block their service, or simply charge their customers more for using it.

At the moment, mobile Internet users comprise a tiny fraction of the total, but this is likely to change. While devices like the iPhone or Nokia N800 are currently very expensive, they show what's possible with current technology — and in the computing industry it doesn't take long for high-end equipment to become a cheap commodity toy.

In Google's ideal world, mobile network access would have the same commodity status that wired connections currently enjoy. To this end, they issued a public statement that they would bid at the FCC's reserve price if four conditions were met:

Open applications
Open devices
Open services
Open networks

The first two are aspects of the same requirement; that client devices should not be limited by the service provider, in terms of software or hardware. An end user should be able to use any device they want and install any software on this device. This is the situation that currently exists with wired broadband, where users can connect any computer to the Internet, running any applications.

These first two requirements are often met by existing services. GSM providers often offer SIM-only plans, where the user buys their own device and inserts a SIM card provided by the network. Phones that are not supplied by a network typically allow the user to run a wide variety of software.

Open services and networks are a little more interesting. This would enforce the same requirements on wireless providers that currently exist for a lot of wired telecoms companies; that they make their network available to third parties at a reasonable rate.

In the UK, the telephone network was built by the Post Office and later spun off into a separate company, BT, which was later privatized. To foster competition between ISPs, BT's wholesale group was required to offer other ISPs access to their network at the same rate they offered their own retail department. In a separate program, local loop unbundling, they were required to allow other ISPs to install their own equipment at exchanges and use the "last mile" connections.

Google's last two conditions would place similar restrictions on the wireless networks. There are logical arguments on both sides as to whether this is a good idea. BT's wired network was built either with public money or with government subsidy (the same is true of AT&T's network in the US, which has had similar constraints placed on some of the Baby Bells). Granting a private company exclusive use of public infrastructure doesn't make a lot of sense, and so it's fairly easy to defend the requirement that they share. This is not, however, the case for a new wireless network. No money, public or private, has yet been spent deploying 700MHz towers in the USA, because the spectrum is not yet available.

The counter argument is that the spectrum itself is a public resource. In principle, a competing telecom company could deploy their own wired network. It would be expensive, but the fact that the cable TV companies and some use municipal cooperatives have done so indicates the possibility. A startup could, with some initial investment, deploy a fiber network in a town and then lease external bandwidth. The only time a new wireless ISP can start is immediately after a spectrum auction like the current one, which is not a particularly common occurrence.

The Final Conditions

The FCC accepted the first two of Google's conditions, but not the second two. This means that whoever wins the spectrum auction will be required to allow any (FCC approved) devices, running any software to connect. They will not, however, have to resell the bandwidth wholesale, nor allow other companies to connect to their network at arbitrary points. Apparently even these two constraints were too strict for some. Verizon is now suing the FCC over these conditions, claiming that they will "limit the introduction of new and innovative wireless services." It seems innovative services are only possible when sanctioned by the network.

If the existing mobile telecoms companies win the auctions, then there will likely not be much change to the status quo. The current oligopoly will keep setting high prices. Consumers will be able to connect any device they want to the network, but the price plans will be structured in such a way that it will be (or, at least, appear) much cheaper to get one from the operator at a subsidized price which is locked down.

Will Google buy the spectrum? They certainly have enough spare cash to do so. If they do, it seems unlikely that they would operate the network themselves since it's a long way away from their core business. Instead, they would be likely to sublicense it to other players with the four conditions they originally hoped the FCC would impose.

Doing this would likely be a net win for Google; they would derive some income from the licensing and get the commodity network they want. Since this would be quite bad for most of the existing wireless companies, it seems likely that they will try hard to avoid it happening, even if it means bidding more than they can really afford.

The most interesting outcome would be for a coalition of companies with a vested interest in an open Internet to follow this plan. While Google has the highest profile, they are by no means the only company which has a business model dependent on the existence of such a network. A small group of some of the big players in this area could easily outbid the telcos.

The auction is likely to be a very interesting process. Google and Verizon are already maneuvering in the courts of public opinion and law, respectively. The prize to whomever wins is the ability to choose the direction the development of wireless services will take in the USA. With stakes this high, expect an exciting show.