Going forward in neutral: the technology paradox

The crock of gold at the end of the rainbow is the optimised use of the radio spectrum. ATDI managing director, John Berry (pictured) examines the options.

Attempts to find this prize have been going on for as long as radio has had a commercial use.

First there was harmonisation and with it came a degree of certainty for all network operators. Ofcom (for intra-UK) and CEPT (for the international picture) developed band plans and inter-network coordination methods. This defined the degree of protection a user would have from the transmissions of another. The methods recognised the imperfections of transmitting and receiving devices. They established the concept that some interference was inevitable but aimed to keep the amount at realistic levels. In extreme cases such as broadcast, pre-coordination even set spectrum use in stone for years to come. The concept was simple. By invoking an a priori regime, the market was assured of its future, albeit perhaps not the future it wanted.

Current trends mean the future will see market measures and technology neutrality. The use of market measures to make spectrum decisions assumes a fundamental belief in the capitalist system as the best show in town - certainly better than any other yet tried. If pre-coordination is completed one would define the band plans, the protection and the methods for interference avoidance after the market event. As a gift to the legal profession, this means that the ultimate remedy in the case of interference after an auction would be a cooperative agreement between victims and culprits (assuming that with aggregated interference they can uniquely identify one another).

For the market to be able to place a value on the spectrum (and hence to know how high to bid) it must be free to determine the technology it might use and for the spectrum market system to function well, usage of tradable parts of the spectrum must be technology neutral. The question is how neutral is neutral?

To look at technology neutrality one needs to determine the parameters that are important thereby making a technology unique in spectrum terms. Each radiocommunications technology has a different waveform that it presents to the air. The term waveform describes the way a signal looks in both frequency and time. A signal described by a waveform will present an amount of energy to other operators (who make use of the same or adjacent spectrum) for a period of time. This energy contributes to noise which in turn may result in harmful interference over a percentage of the victim operator's service area. Harmful interference is that which denies the operator of another spectrum dependent system due benefit from his spectrum investment.

In the harmonised world, the maths of protection was done, a priori, assuming the same waveform for all co-band sharers. A similar state existed for inter-band sharing. In the technology neutral world every co-band sharer or potential adjacent-band interferer uses a different waveform. In the case of differing waveforms the energy into any victim receiver will be different for each technology option. But one needs this energy level before knowing the risk in sharing. One needs the energy level before placing a bid. After the event is no good for making business plans; somehow prior knowledge of all parties' intentions and interference potential is needed.

The argument so far is that a priori determination of protection (and risk) is essential to the market value and tradability of spectrum. Without a return to harmonisation, how might this be done?

Co-band and adjacent band sharing both between regions and within regions under the harmonisation regime simply required the maths to generate limit nuisance fields at test points representing the sharers' interests. Sharers using the same technology maintained their signal below the limits and everyone was happy. The term spectrum usage rights has now been coined to describe sharing in a technology-neutral regime. Test points are again used to delineate the sharers' interests so how does this differ from the harmonised case?

The answer is that the maths needs to be done so that the limit of energy at the test points is computed for all possible or permissible technologies to allow subsequent technology choice. Then any technology working within the limit (pre-computed considering all possible or permissible technologies) can be used. There lies the paradox. Before the limit can be determined, the waveform(s) must be known. But we don't know, a priori, that waveform. We must therefore assume a waveform for the purposes of computing a limit and in so doing we determine the technology.

Like most paradoxes the solution lies in philosophical debate. Potential bidders will need to model their proposed valuation and its intended technology against the spectrum usage rights or spectrum masks proposed by the regulator. Whether they find the proposals of the market creator's invisible hand acceptable will depend on the degree of freedom available to them and how those map to their business plan.