The FCC ended the year with a flurry of activity, and though its vote on its net neutrality plans gained most of the headlines, a new project focused on flexible spectrum could have almost as much impact in the future. The agency has issued a notice of inquiry (NOI) about dynamic use of spectrum, which could see licensed bands becoming more flexible, drawing on some of the ideas already adopted for unlicensed white spaces frequencies.
The FCC has made the threatened ‘spectrum famine’ in the US a highlight of its activities since the 2008 election and is investigating a wide range of options to open up new frequencies for wireless and broadband. It is clear that just releasing new spectrum via traditional auctions will not be enough and there will need to be a greater emphasis on using existing bands more efficiently.
Some plans involve harnessing underused areas for unlicensed use (the white spaces in the TV band) or licensed operators (mobile satellite spectrum). Others will look at cramming more service providers into each band, allowing for greater competition and range of services, as well as maximizing the use of the airwaves. The new consultation recalls concepts heavily publicized by Google in 2007, when its lobbying machine adopted the idea of an on-demand approach to spectrum. But new techniques, and lessons learned from programs like the white spaces, may put a stronger dose of reality into the ideas this time around.
In particular, a geolocation database ‘ like that being developed for the white spaces in the US – could be used to allow multiple licensees to run in the same spectrum, without interfering with one another. This database approach could help pave the way for campaigners like Google to get closer to their ideal of all spectrum being available on an on-demand basis to scores of service providers, rather each license being allocated to a single holder (and often underused).
“Even as we look to free up existing spectrum to meet the needs for wireless broadband, we can and should explore ways to make more dynamic and opportunistic use of the spectrum we have,” noted FCC chair Julius Genachowski as the notice of inquiry was released over the holiday period. In particular, the NOI aims to find out “what additional steps the Commission can and should take to encourage, promote, and incentivize their development and use in both unlicensed and licensed spectrum’.
The agency is also seeking comment on issues such as expanding white space services to other bands, or allowing license holders to rent out spectrum to secondary users. It also wants comments on how to lease non-contiguous spectrum, which could help it to allocate spectrum that is currently frozen for lack of a viable business case.
Republican FCC commissioner Robert McDowell commented: “I hope our updated rules will adhere to the Commission’s more recent ‘flexible use’ policy. Old style ‘command and control’ rules not only hamper creative entrepreneurs who are in the best position to understand and satisfy consumer demands, they cause spectral inefficiencies as well.”
It will be interesting to see how the cellcos react to all this. They need more spectrum to support exploding urban usage of mobile data and offer new services, but they want to keep this firmly under their control. In general, they have opposed initiatives that allow for greater license-exempt or shared use of spectrum, because these lower the barriers to entry for smaller players and disruptive new business models. However, as they did with public Wi-Fi, they can also hope to use their market weight and brand ‘ and experience in running networks ‘ to partner with any new entrants, thus harnessing additional spectrum capacity for their own ends (and without the huge cost of bidding for it in auctions). The enthusiasm of cellcos like AT&T for using Wi-Fi hotspots to offer new customer bundles and to offload data from 3G shows the way that new lightly licensed or dynamic systems could go ‘ not necessarily undermining the power of the big operators, even if they are forced to modify their behavior.
This would certainly not be the outcome foreseen by Google during its 2007 campaign, though of course, the search giant is a realist, and works closely with carriers like Verizon even if, in its dream world, the wireless world is dominated by open web players. It knows it cannot actually sideline the big carriers ‘ its famous bid for 700MHz spectrum was geared to manoeuvring the eventual victor (Verizon) into a major open access concession, not actually acquiring airwaves in order to become a carrier itself. But that example showed how it can use its market and lobbying strength to force the incumbents into changes that benefit its open web vision, and the adoption of its adverts and services.
In 2007, Google proposed to the FCC a scheme that would support a common grid of wireless spectrum, that could be used and paid for by various providers on an ‘on-demand’ basis. And many of those ideas are now being revisited as part of the latest debate, before a very different FCC. On-demand is a concept that is familiar in the enterprise technology sector, and in utilities such as power, but in wireless there has been little middle ground between fully owned spectrum (either directly operated or wholesaled) and unlicensed airspace.
Google suggested that the mechanism for buying time on the spectrum should be a real time auction, with pricing therefore fluctuating according to demand rather like a commodities or currency market. This would allow potential service providers to bid in real time in an auction for the rights to use a piece of spectrum for a pre-determined period, which could be very short – “a year, a month or even two seconds”, as the filing put it. It would be similar in concept to AdWords, a service Google offers in which companies bid against each other to have their adverts displayed against key search terms inserted in the Google engine.
Primary licensees would acquire spectrum in the usual way, and then run the auctions for third party bidders. “Licensees in this case can in some sense be an intermediary between the airwaves and people who want to use it,” said Richard Whitt, telecom and media counsel for Google and author of the filing. “What we’re looking for is the FCC to give us the flexibility to explore the business and financial relationships between the licensee, those who build the infrastructure and those who want to be retailers.”
This is blue sky of course, and the centralized auction system would be terrifyingly complex and political to plan (the white spaces database is difficult enough). But the ideas are back on the table, even as the FCC will actually explore more pragmatic approaches to making spectrum dynamic. In particular, it is wondering whether there are the equivalents of white spaces in other bands ‘ and more importantly, how easily these could be identified. That challenge is relatively simple in the TV bands, where the incumbent devices are non-mobile (TV stations and sets, cable head-ends). Pushing the concept out to the satellite spectrum could be “extremely challenging’, the FCC concedes.
There will also need to be a very hard-nosed approach to the proposals on renting spectrum for casual use, relying on cognitive or ‘opportunistic’ techniques to avoid interference. Yet tools such as spectrum sensing or sniffing, which let one device identify and avoid another, are immature and were, in fact, rejected as a way to avoid interference in the white spaces. The FCC remains interested in future potential though, as these technologies evolve.
‘What innovations to sensing are contemplated?” the document asks. “Are there critical component price points that must be reached for advances to take place? Has there been any industry wide consensus regarding methods of implementing sensing?”
The other source of readily available new spectrum may come from techniques to allow operators to lease or buy ‘scraps’ of non-contiguous spectrum, previously regarded as worthless. As Ars Technica points out, the main issue is how close together the slivers of spectrum need to be to make it practical to create devices that traverse them.
Finally, the FCC wants to study how ‘extremely high frequency’ bands ‘ above 38.6GHz ‘ can be used cost effectively for broadband. These higher bands are severely underused, though they are starting to be harnessed for some backhaul and satellite uses ‘ but they have disadvantages such as short range and the need for line of sight.