We Should Not Regulate for Cognitive Radio
I was asked some questions by a group carrying out some kind of analysis of the state-of-play of cognitive radio about regulating for cognitive radio and which bands should be considered if this were to happen. It got me thinking about the issues from a very general perspective.
If I really believe in technology neutrality – which I so very much do – then as a logical extension there should be no regulations made to specifically support cognitive radio. There should NOT be a cognitive radio pilot channel. There should NOT be a cognitive radio band or sandbox. There should NOT be a set of spectrum access rules based on the ability to sense spectrum or to garner an understanding of the operating environment. You get my drift …
However this does not mean that cognitive radio should not be considered when regulating.
Our mantra here in CTVR is – imagine where technology can lead … let your imagination go wild … take the technology out of the result … regulate. So we need to imagine where cognitive radio can lead us in all its glory and then we need to take the technological specifics out of the equation.
How can we extract or abstract what cognitive radio can offer? Some of it is pretty obvious in terms of the technical capabilities of a cognitive radio but perhaps less obvious in terms of the general. For me three key points are:
(1) Cognitive radio can facilitate new forms of spectrum access – whether that be dynamic access to exclusive swathes of spectrum or non-exclusive sharing of spectrum or some other scheme. More importantly and more generically cognitive radio can support access to spectrum either as a rival or nonrival good as appropriate. In fact in principle the cognitive radio of the future should be capable of playing a role in deciding whether maximum benefit is obtained from the resource should it be made available as a rival or nonrival good. Now that I think of it, spectrum is so interesting in that it can be either.
(2) Cognitive radio can facilitate the self-correction of negative externalities and remove the onus on the regulator to foresee in advance where these issues might arise. In spectrum management negative externalities are nearly always related to interference costs though of course there are others. For example a cognitive radio can enable the dynamic introduction of self-induced guard bands, support the shaping of out-of-band emissions and otherwise react to the environment in which it finds itself. It can facilitate bargaining and collaboration with neighbours to adjust negative externalities or indeed increase positive externalities.
(3) Cognitive radio can support all sorts of self-organising autonomous behaviour. This can be at the node or network level or at the spectrum access level (i.e. through automated spectrum auctions) or at the network planning and rollout level or at the compliance and monitoring level etc., etc, etc.
I have not explained these points in as much detail as they warrant and indeed not everyone may agree but the point I want to make in this post, whether you might agree or disagree with the above statements, is that the regulations should follow from the general . So based on the above comments my suggestions for regulating would include the following:
A. Rules governing spectrum should be technology neutral.
B. They should be constructed in a manner that allows bands of spectrum to fluidly change from a rival to a nonrival good rather than rely on the current classifications of licensed/unlicensed or exclusive usage/commons or whatever the labelling. In fact any labelling of bands in this fashion or in accordance with a technology or service (TV bands or LTE bands) should be banned.
C. No guard bands or any other such instruments used to control interference should ever be defined again.
D. We should move towards the complete automation of all regulatory processes which means ceding more authority to the communication systems that emerge both in terms of operation and compliance.
E. It might be useful to accept the general point that wider collaboration and cooperation between different communication systems is inevitable and to allow for a refining of any base or default rules and positions through subsequent negotiations between spectral and spatial neighbours.