UK white space could fill up fast

Ofcom demands new regulatory powers

The UK's white space spectrum could start filling up before the Digital Dividend comes around, but only if we're prepared to give Ofcom a lot more power, and only if we can all agree on what white space actually is.

While the FCC is still dithering about who's going to run the American white space directory, UK players have been gathering in Reading to talk about how best to exploit the UK's TV transmission gaps. Ofcom says white space devices could be legal next year if it's given the power to regulate them properly.

White space spectrum is found between the TV stations, and exists because of the huge scale of television transmitter networks. The frequency used to broadcast BBC1 in London is empty in Oxford and can't be used for TV transmission without interfering with the London broadcasts, so it lies empty as "white space". But there's no guarantee that it will be empty, or remain so.

The original idea was that white space devices would detect which frequencies were in use and avoid them, but that's proved technically impossible (Ofcom did calculations, and in the US the FCC invited demonstrations). So now the plan is for a database of locations and frequencies requiring devices to check in before settling on a frequency - perhaps using detection to avoid interfering with each other. Such a database, updated in real time, presents a new model of spectrum management which could ultimately be spread far beyond the gaps between the TV channels.

But that's all theoretical for the moment. Right now white space is a chunk of spectrum that Ofcom would like to hand over for licence-free use, if it only had the power to do so. Aside from the technical problems two major logistical issues remain: Ofcom's reluctance to run a database while still seeking to control it, and the fact that no one can agree on how much white space is available.

Over in the USA the FCC is still consulting on who should run the database: Google has offered to run a master database, with everyone else drawing from it, but others see too much power going to the Chocolate Factory and would prefer to see peer-to-peer synchronisation.

The US has a lot more white space to play with. In cities the ubiquity of cable TV reduces the need for broadcast systems and rural areas don't get a lot of channels, so they stand to gain from white space applications. But the UK has a homogeneous broadcast TV network, with frequency reuse dependent on terrain as much as on distance, which limits the gaps available.

Not that everyone gathered in Reading agrees there's any white space available all. Sennheiser, talking on behalf of the British Entertainment Industry Radio Group (BEIRG), pointed out that the PMSE luvvies have been hanging around in white spaces for decades, and are already getting squeezed by digital TV (which uses fewer frequencies, and thus provides fewer gaps). The BEIRG reckons that Digital Television hasn't left a lot of white space anyway - as little as 8MHz in Edinburgh, for example - which contrasts with other estimates of 100MHz of available spectrum around the country.

Ofcom's white-space maps aren't up to date enough to resolve that question, which also depends on how white one expects the space to be. BT reckons that if the technology developed allows the use of adjacent bands then 90 per cent of the UK will get more than 40MHz. Otherwise the majority will only get 15MHz, while Londoners get nothing at all.

The BBC meanwhile isn't convinced that any use of white space is possible without interfering with digital television. Being British the corporation didn't have evil animated telephones, but rather dry technical facts and experimentally-proven evidence used to point out that location-based services won't work indoors and that sensing technologies don't work. But brushing those under the carpet ("technology" will make it work, apparently) the presentations moved on to applications for white space.

Which was a shame, as there aren't any applications, as such. Even Microsoft couldn't be drawn on what this unlicensed spectrum would be used for, saying only that it expected to be surprised and that more bandwidth is always a good thing. Cloud-based computing was mentioned several times, with Google's Chrome OS brought up as an example of the kind of device that requires ubiquitous connectivity - despite the fact that Chrome devices only exist as a couple of prototypes and in dozens of press releases.

The two applications everyone agreed on were in-home distribution of HDTV and fixed broadband internet access. The former could even transmit a digital TV signal, thus not requiring a receiving box and giving an advantage over the competition, though as the chap from Cambridge Consultants pointed out that's not going to go down well with Sky. As for fixed broadband, the problem comes when BT undercuts by upgrading the exchange, unless the kit can be deployed really cheaply.

Also present was a delegation from the University of Strathclyde, who talked about point-to-point connections between disparate islands, providing backhaul for Wi-Fi-based community access networks as a fill-in technology until BT gets around to upgrading the local exchanges. This needs no sensing of TV transmissions and no live database requiring a secondary connection for updating. Just tune both ends to an unused TV channel and change the frequency if anyone complains.

But that's not going to make billions in revenue for anyone, which is what the white space crowd would like to do. But reality is rarely as exciting as the PowerPoint presentation would have it, so the meetings continue and the details are debated while the Scottish get on with connecting people. ®

Sponsored: Minds Mastering Machines - Call for papers now open

Biting the hand that feeds IT © 1998–2018