Digital dividend: cognitive access
Statement published 01|07|09
1.1 Since its launch in 2005, our Digital Dividend Review (DDR) has considered how to make the spectrum freed up by digital switchover (DSO) available for new uses.(-1-) This includes the capacity available within the spectrum that will be retained to carry the six digital terrestrial television (DTT) multiplexes after DSO. This is known as interleaved spectrum because not all this spectrum in any particular location will be used for DTT and so is available for other services on a shared (or interleaved) basis.
1.2 In our statement of 13 December 2007 on our approach to awarding the digital dividend,(-2-) we considered the use of interleaved spectrum by licence-exempt applications (i.e. those exempted from the need to be licensed under the Wireless Telegraphy Act 2006(-3-)). We concluded that we should allow cognitive access as long as we were satisfied that it would not cause harmful interference to licensed uses, including DTT and programme-making and special events (PMSE). This could potentially bring substantial benefits to citizens and consumers in the form of new devices and services.
1.3 Cognitive devices should detect spectrum that is otherwise unused and transmit without causing harmful interference. They have the potential to support a wide range of uses, including high-speed always-on broadband.
1.4 In a consultation published on 16 February 2009(-4-), we proposed a number of technical parameters that we suggested would prevent harmful interference while enabling licence-exempt cognitive use of interleaved spectrum. This statement concludes on some of the issues raised but notes that further work is needed on others.
1.5 The cognitive consultation noted that there were three main approaches to determining whether spectrum was unused – detection, geolocation and beacons. It suggested that beacons were the least appropriate and did not merit further investigation and this was widely agreed by respondees. It asked whether detection and geolocation should both be enabled as alternatives. Some respondees agreed to this while others felt that geolocation was essential or that both approaches needed to be used together. Many noted that in any case, detection alone was very difficult to implement and unlikely to be used for the foreseeable future. Overall, existing licence holders of interleaved spectrum generally accepted the level of protection offered by geolocation although some in the PMSE community were concerned about the speed of database update and saw the use of detection in addition to geolocation as a way to resolve this issue.
1.6 If detection alone is used then the likelihood of interference occurring depends on the parameter values selected. The proponents of cognitive access argued that we had been too cautious in our selection of parameters and levels more favourable to cognitive devices should be selected, while licence holders and other affected stakeholders argued that we had been insufficiently cautious and that parameters should be modified to provide greater levels of protection. For the most part, these arguments rested on the identification of certain situations where cognitive devices might be in proximity to licensed devices but unable to detect them.
1.7 We conclude from the responses that the most important mechanism in the short to medium term will be geolocation. While we set out some ideas about how this might operate in the cognitive consultation we did not cover all aspects in detail and hence believe that further work, possibly leading to consultation specifically on geolocation, is appropriate. Hence, this statement does not conclude our work on cognitive access. Instead it sets out those aspects where we consider we can draw a conclusion such that we can concentrate on the remaining issues.
1.8 Our view on detection is that it should be allowed but with sensing levels and transmit levels lower than we had originally proposed. In the cognitive consultation we presented detailed analysis to show how these levels were derived. This was generally supported but some respondees suggested other deployment scenarios and showed how these levels would not afford sufficient protection under these scenarios. We accept that greater levels of protection might be needed in these scenarios but consider them sufficient unlikely that to afford such protection would place an excessive additional burden on cognitive devices. We do not have data to substantiate this view – and indeed until the likely uses of cognitive devices are known it would not be possible to assemble this data. Hence, this decision is based on a subjective assessment of the likely risks. In addition, with the reduction in levels mentioned above, we have gone some way to affording additional protection to all licence holders.
1.9 As a result we propose to allow detection alone as well as geolocation. However, we note that implementation of detection-only devices is likely many years away and hence there is little advantage in rapidly making the necessary regulations to licence-exempt such devices. Given that there may be benefits in European harmonisation and that more evidence may emerge we have elected to set out our chosen parameters in this statement but will not take any further action at least until we have concluded on geolocation parameters. If further evidence comes to light before we make regulations, we would reconsider the parameters set out here as appropriate.
1.10 We welcome the Government’s support for our establishing the parameters for successful cognitive access in the UK and achieving the international harmonisation that is required, as set out in its Digital Britain Final Report(-5-).
Key device parameters
1.11 Table 1 sets out the parameters we have concluded on if cognitive devices are to use detection alone. Transmit powers here and throughout this document are effective isotropic radiated power (EIRP) into an 8 MHz bandwidth.
|Sensitivity assuming a 0 dBi antenna|| -120 dBm in 8 MHz channel (DTT)
-126 dBm in 200 kHz channel (wireless microphones)
|Transmit power||4 dBm (adjacent channels) to 17 dBm|
|Out-of-band performance||< -46 dBm|
|Time between sensing||< 1 second|
1.12 Table 2 sets out the key parameters for geolocation that we are able to conclude upon at this point.
|Locational accuracy||Nominally 100 metres|
|Transmit power||As specified by the database|
|Out-of-band performance||< -46 dBm|
1.13 We will now work with stakeholders to further develop the concepts and algorithms necessary for geolocation and expect to consult further on geolocation later in 2009.
1.- See www.ofcom.org.uk/radiocomms/ddr/ for more information about the DDR, including previous publications.
In this section
Digital dividend: cognitive access (679 kB)
Full Print Version