Spectrum efficiency is a key benefit of the transition to digital broadcasting.  In many European countries, digital terrestrial TV transmissions are being squeezed in alongside the analogue TV transmissions.  After the analogue TV services have been closed down, the released spectrum could be used for a range of purposes such as:

• increasing the number of digital TV services (e.g. designed for reception on roof-top antennas or set-top antennas);
• improving the coverage of digital TV transmissions (e.g. increasing the population coverage and/or providing higher field strengths so as to deliver more reliable services to mobile receivers or set-top antennas);
• digital TV services designed for reception on hand-held receivers (e.g. DVB-H);
• high-definition TV (as in the USA, Japan, Korea, Australia)
• non-broadcasting services, such as 3G mobile phones, WiMax, etc.
• other services yet to be defined . . . . .

This release of spectrum is widely referred to as “the digital dividend”.  Many people seemed to think that the ITU Regional Radiocommunication Conference recently held in Geneva (RRC-06) – described in one of the articles in this issue – would identify the digital dividend on a country-by-country basis (e.g. x MHz in country A, y MHz in country B, etc.).  However, the main objective of RRC-06 was to produce a Plan for digital terrestrial broadcasting in VHF Band III and UHF Bands IV/V. In practice, the resulting GE-06 Plan is an all-digital plan for use after the analogue TV services have been closed.  Indeed, the GE-06 Plan makes it clear that analogue TV services in UHF Bands IV & V will not be protected against interference after 17 June 2015.

Considering that many Administrations had argued during RRC-04 in favour of delaying the transition period to 2028 or even 2038, agreement at RRC-06 on such a rapid timetable for the withdrawal of analogue TV services is a remarkable achievement. Nevertheless, there is considerable pressure to accomplish an even quicker withdrawal of analogue TV services: the European Commission has proposed that analogue switch-off should be completed across the EU by the beginning of 2012.  Some countries with a high penetration of cable and/or satellite TV expect to withdraw all of their analogue terrestrial TV services fairly rapidly (e.g. by 2008), whilst other countries may not meet the EU 2012 deadline (especially those that have so far made no firm plans to introduce digital terrestrial TV services). 

Although RRC-06 did not define the use of the digital dividend, regulators in each country now know more about the size of their digital dividend because they probably asked for more broadcasting services than they needed.  These “extra” frequencies have been coordinated with neighbouring countries and are included in the GE-06 Plan.  Assignments and allocations in the Plan can be easily modified as long they do not cause additional interference or demand protection greater than that defined in the Plan. 

Most countries have not yet revealed how they intend to use the frequencies obtained at RRC-06.  One notable exception is the UK which has declared 14 UHF TV channels (channels 31, 32, 33, 34, 35, 37, 39, 40, 63, 64, 65, 66, 67 and 68) as their digital dividend.  The UK regulator Ofcom is holding a public consultation (the Digital Dividend Review) to help them determine how this spectrum should be used in the future.  It is important to emphasise that the size of the digital dividend varies dramatically from one country to another: small landlocked countries surrounded by several other countries cannot hope to match the digital dividend achieved in the UK – which benefits from its geographical position at the edge of Europe, as well as having only one land border (with Ireland). 

One of the little-noticed provisions of the EU’s common regulatory framework for electronic communications networks and services (Directive 2002/21/EC) concerns management of the radio spectrum: in particular, Article 9 says “ Member States  …… shall ensure that the allocation and assignment of such radio frequencies by national regulatory authorities are based on objective, transparent, non-discriminatory and proportionate criteria.”  This statement is obviously applicable to decisions to be taken by individual Member States on how the digital dividend is to be used – but it does not mean that all Member States will make identical decisions.  As regulation of the radio spectrum is clearly the responsibility of national regulatory agencies, different countries are certain to use the digital dividend in different ways.  Such differences across the 25 EU countries are not unexpected, given the diversity of spectrum requirements – as well as cultural, economic and geographic differences.  Although an EU-wide harmonised frequency plan would certainly simplify the lives of spectrum managers, it would not necessarily deliver optimal results in terms of matching the requirements with spectrum availability.

As mentioned already, digital broadcasting uses the spectrum much more efficiently than analogue broadcasting – but some people seem to think that broadcasters have deliberately under-used the spectrum allocated to broadcasting.  For example, I occasionally meet people (usually lawyers or economists, rather than engineers) who note that 4 – 6 analogue TV services are available in a given geographical area and conclude that the remaining 40 or so channels in the UHF bands are unused because of the incompetence of the frequency planners.  Such accusations demonstrate the old maxim that “a little knowledge is a dangerous thing” – but the accusers rarely want to be bothered by the boring details or facts, such as the supposedly “unused” channels being heavily used in adjacent geographical areas!

In practice, the poor radio-frequency performance of analogue TV sets has been a severe constraint on frequency planning: for example, if you operate an analogue TV transmitter on channel n, it is not possible to transmit services on the adjacent channels (n + 1 and n - 1) [1].  Fortunately, digital TV set-top boxes have much better RF performance and, consequently, adjacent-channel interference is much less of a problem.  In fact, as shown in the spectrum analyzer screen-shots in Editorial 302, co-sited digital TV transmitters can operate on adjacent channels.  Although this is a significant improvement over the analogue situation, it important to understand that interference from adjacent-channels cannot be ignored – even for digital TV.

Imagine that you currently receive a good signal from a digital terrestrial TV transmitter on channel n and that another broadcaster decides to add an extra service on channel n + 1 from a different transmitter site – and that this site is close to your home.  Interference to channel n will occur if the signal on channel n + 1 is, for example, 20 or 30 dB stronger than the signal on channel n. Note that, in the world of digital broadcasting, such interference does not result in patterning on the picture: your digital service simply disappears!  Such concerns are not hypothetical: practical experience with DAB services shows that the addition of adjacent-channel transmitters punches “holes” in the service areas of existing transmitters.  Such problems can be solved by adding low-power gap-filler transmitters on channel n whenever you build a transmitter on channel n + 1 or n – 1.  This solution is not cheap – but it is often necessary even where all of the transmitters have been carefully planned (as in a broadcasting network).

Imagine now that the adjacent channels are used by low-power mobile transmitters (e.g. mobile phones) that might be used by your neighbours – or even by members of your own family in other parts of your home.  If your reception of a digital TV programme suffers occasional failures or perhaps frequent failures, would you be able to diagnose the cause and correct it?  More generally, how could reception of digital TV be protected in such circumstances?  

It is not appropriate for mobile transmitters to operate in or near the frequency bands used by broadcast services (which deliver relatively weak signals to millions of receivers).  In essence, it is difficult to mix downlinks and uplinks without adequate frequency separation. The UMTS Forum is demanding that mobile services should be permitted to operate within UHF broadcasting bands, but even they have acknowledged the need for a guard band of 16 MHz between broadcast and UMTS services.  As we are constantly being reminded that spectrum is a very valuable resource, it is interesting that the UMTS Forum has actually proposed that 16 MHz of prime spectrum should be wasted as a guard band!

If the broadcasting bands are to be used for various applications, some degree of harmonisation could be very valuable – for example, a block of channels in UHF band IV would allow manufacturers to produce hand-held devices (e.g. DVB-H receivers involving frequency-sensitive physical parts, such as antennas and filters) that could operate on a pan-European basis. Looking at the GE-06 Plan, there is no prospect of the digital dividend releasing harmonised channels across Europe. To rectify this problem, it has been suggested that another planning conference should be held, perhaps in 2010 or 2012, to replan UHF Bands IV and V.  It seems rather surreal that this proposal surfaced less than 3 months after the delegates to RRC-06 rightly congratulated themselves on a very successful outcome!  Given that preparations for RRC-04 & RRC-06 involved huge amounts of effort, it seems unlikely that the ITU would be keen to hold another Conference in the near future. But who can predict the future?