Christoph Dosch, Institut für Rundfunktechnik (IRT) and David Wood, European Broadcasting Union (EBU)
Administrations throughout the world need to decide how to use radio spectrum so that the public will be best served. In particular, they need to consider the future use of the frequency bands currently used for analogue television broadcasting, after television broadcasting goes digital.
Should we seek the best solutions for the short term? Or should we look to the long term? How important is “green” efficiency? How important are interoperability and flexibility? These and other dimensions need to be carefully weighed.
Broadcasts, the Internet, and telephony are all important to people. National broadcasts in particular help define national identity, enhance social cohesion, and are a gateway to knowledge, information and entertainment. Our lives today are unimaginable without both broadcasting and the Internet. And in future we may see hybrid broadcasting, bringing together broadcast and Internet services on the same television set. What is the most cost-effective way to provide these services, given the evolution of technology and the spectrum available?
Digital terrestrial broadcasting versus mobile broadband
The top candidate for the spectrum used today for analogue terrestrial broadcasting is digital terrestrial broadcasting. In 2006, ITU developed the Geneva 2006 (GE-06) Agreement for planning the digital terrestrial broadcasting service in parts of Regions 1 (Africa and Europe) and 3 (Asia and Australasia), in the frequency bands 174-230 MHz and 470-862 MHz. Under this agreement, all of the frequency bands used for analogue television broadcasting would be used for digital television broadcasting containers (multiplexes).
The 2007 World Radiocommunication Conference (WRC-07) made an allocation in the upper part of the UHF broadcast band for the mobile service, to accommodate international mobile telecommunications (IMT) in Region 1 in the range 790 to 862 MHz, to take effect at the time of the transition to digital. Little information was available on how to share IMT with the broadcasting service, and the ITU Radiocommunication Sector (ITU-R) was charged with studying this. These studies were carried out by the Joint Task Group 5-6, whose report will be considered by WRC-12 under agenda item 1.17.
Under discussion now is whether IMT services should extend further into the former analogue broadcast bands. As explained, the broadcast band 790-862 MHz is already planned for IMT in some countries of Europe. This so-called 800 MHz band is earmarked in Europe for the mobile service, specifically for IMT, after the digital switchover. Should we use more of our broadcast spectrum for IMT in future?
A case being made for increasing the bands for wireless broadband is that it would bring Internet services to rural areas, where the cost of wired Internet services per user would be high. Technology that could help achieve this efficiently is being developed and tested. It might be possible to introduce such services in the current analogue broadcast bands without disturbing digital broadcasting by using the so-called white spaces. This would be a win-win situation both for broadcasting and for the additional services, provided agreed compatibility rules are observed.
In some countries, for example Germany, where the switch-over from analogue to digital television had already taken place, planning for IMT in the form of the “older” long-term evolution (LTE) system in the band 790-862 MHz has started.
This allocation is set for use by IMT, but the Joint Task Group 5-6 has not yet considered all cases of interference, so some problems remain unresolved, in particular:
The LTE uplink affects the broadcasting service below 790 MHz; LTE handsets operating near a television receiver can block television reception;
Because of the small guard band (of 1 MHz) between the broadcasting and the mobile services (from 790 MHz to 791 MHz), there are no adequate, economically viable filters to provide sufficient attenuation of out-of-band emissions of mobile service IMT signals, so channel 60 cannot be used for mobile television services, if at all.
Strong LTE downlink signals in the vicinity of a mobile service IMT base station may cause a severe degradation of the television picture quality in areas where the DVB-T signal is relatively weak in comparison to the LTE signal.
There are thousands of licence-exempt wireless microphones in the hands of the public that operate in the range 790-862 MHz on a secondary (that is, non-interfering) basis, and that now need to be relocated in other frequency bands. The proposed relocation of these wireless microphones to frequencies below 790 MHz is problematic as this band is already used for professional applications ancillary to broadcast production – the so-called programme-making special event systems – that also operate on a secondary basis.
Criteria for a complex choice
Radio spectrum is a resource that belongs to nations, so what is in the interests of the public at large? Should we try to provide the greatest choice for the public, the highest transmission efficiency, the highest quality, or the lowest cost? Should we provide the most valued services to the largest number of people?
Should we consider the long-term interests of the industry and the nation? If so, we will need to understand the likely growth in demand and to anticipate the evolution of technology.
Another criterion could be to minimize the environmental load associated with making and delivering services. This would involve assessing the total end-to-end energy consumption associated with various options, and choosing systems with the highest green efficiency.
The need for terrestrial television broadcasting
In Europe, people on average watch standard-definition television broadcasts for more than four hours per day, and viewing time is increasing annually.
Although television broadcasting can be provided by satellites or cable, and they serve the public well, terrestrial broadcasting is important for most countries of the world. In the European Union, 50 per cent of households rely on terrestrial broadcasting. Among the many reasons for this are:
convenience for viewers, with reception possible on fixed, portable or handheld receivers;
facility for adjusting services to local needs and coverage;
ease of rectifying breakdowns (the repair team cannot go to a satellite);
national control of distribution infrastructure;
often the only accessible medium in national emergencies;
for many consumers, the most cost-effective way of receiving television.
Factors affecting spectrum needs
One of the factors affecting the amount of spectrum a service needs is the technical quality it offers. Digital technology allows for a range of television picture qualities, so programme providers can choose how technically attractive to make their programmes. Other things being equal, the higher the sound or picture quality, the longer the time the viewer will watch (or listen to) a given programme. The choice of picture quality also affects the cost of providing the service, and the number of different programmes that can be accommodated in a given spectrum space.
Another of the major factors affecting the amount of spectrum a service needs is the quality efficiency of the transmission system. This is the technical quality delivered relative to the bit rate needed to do so. Quality efficiency is affected by the combination of modulation and compression technology used. It is improving with time, and this can be taken advantage of when new systems are introduced.
Quality efficiency is an important factor in understanding the difference between delivering media by broadcasting, by Internet or by combined systems. Multi-purpose systems with modes allowing for broadcasting or wireless broadband could be developed but – like a vehicle designed for both land and sea – would be more complex technically and less efficient than systems developed for a single mode of use.
Television broadcasting today uses the radio spectrum in three bands. Digital mobile phones have traditionally used radio frequency bands above those used for broadcasting. The advanced forms of digital mobile phone networks – the 3G systems that can provide Internet – have also used frequency bands above the broadcast bands.
Different frequency bands have different advantages and disadvantages. The lower the frequency band, the larger the receiving antenna needed, and the lower the capacity available for transmission. But the lower the frequency band, the fewer the number of transmitters needed to cover a given area and the more easily the signals will pass through the walls of buildings.
Higher bands (above those used today for broadcasting) offer higher capacity and smaller antenna, and are thus very suitable for services via handheld devices with high-capacity broadband Internet. Higher bands offer greater isolation to other broadband Internet cells and therefore increase transmission efficiency. Lower bands, such as those used formerly for analogue broadcasting, are particularly suitable for digital broadcasting or for low-capacity (“second class”) Internet.
In broadcasting, the service area is independent of the number of receivers that are switched on, and so is the service quality. Television service areas can thus be large (up to 100 km in diameter), and are ideal for services to large populations. In contrast, mobile Internet services can only accommodate a certain number of users simultaneously. Service areas are small – often limited in cities to a few hundred metres, to cope with high traffic demand. Quality of service is not guaranteed but is provided on the basis of “best effort”.
Consumer demand for digital television
Digital technology allows for greater flexibility and larger numbers of television channels in the same spectrum space. It also provides a route to services which offer a better and richer viewing experience.
Broadcasting needs growth potential. Over the next 40 years, a series of quality steps will take television from standard definition to high definition, then on to three-dimensional, ultra high definition and object wave television. If television broadcasting does not follow the available technology it will eventually be spurned by the public – as the public would today spurn the dim black and white television pictures of the past. A service that is not allowed to evolve is sentenced to die.
The higher spectrum efficiency of digital transmissions provides the basis for this growth potential – provided there is adequate spectrum. Digital modulation and compression technology are evolving over time. Coupled with the evolution in chip density (Moore’s law) in consumer electronics, this makes possible systems that have greater capacity and can offer a better experience to the viewer.
Consumer demand for Internet capacity
In the developed world there has been spectacular growth in consumer demand for Internet capacity (bit rate). The capacity delivered to Internet users has risen more than 20-fold in 15 years. Domestic demand in the developed world is predicted to rise to 100 Mbit/s per user in the next 20 years. Demand in the developing world will follow suit.
Over the past ten years, the trend in the developed world has been to ever higher quality multimedia on Internet as a source of entertainment. This has meant not just the demand for higher bit rates, but also for prolonged usage times. Demand for many hours of use per day, with peak usage times, looks likely to emerge as the pattern in countries where it can be achieved.
As an example, if a future Internet service carries an entertainment service, it will need to be able to deliver bit rates of 8-12 Mbit/s simultaneously and independently to the national population between 7 p.m. and 10 p.m. each day. Technically, this kind of service cannot be provided by wireless broadband in the broadcasting bands, however much of the current analogue television broadcast bands are given over to carrying that amount of data. Delivery by means such as optical fibre or higher bands will be required. Wireless broadband in the current broadcast bands will be limited to providing second class Internet services that will probably not have a long-term future.
Performance of broadcasting versus Internet
Internet has the advantage of direct interaction with the user and the potential for a greater degree of personalization. But delivery capacity is limited by the available infrastructure, and there is a financial cost for each additional user. Also, Internet reception is not anonymous, in that the State or service provider is able to track its use.
Broadcasting has inexhaustible capacity and zero marginal cost with respect to the number of simultaneous users. The number of users who take up a broadcast can be infinite in a given service area, and there is no cost for an additional user. And broadcast reception is anonymous (at least for free-to-air emissions).
Multi-purpose systems that offer both broadcast and Internet
Wireless Internet delivery systems can have broadcast modes, but those developed to date are less efficient than systems developed specifically for digital broadcasting. No mobile network operator has yet introduced such a system.
Possibly for commercial reasons, the mobile telephone community has been unwilling to work with digital broadcasters to develop an efficient single receiver system that allows reception of both wireless broadband and digital broadcasting.
Environmental load of broadcast and Internet
Studies by the British Broadcasting Corporation confirm that the environmental load associated with broadcast and wireless broadband depends on the number of users. For services where there are a large number of users for the same content, broadcasting is the more environmentally friendly mode of delivery. Only for services with a very small number of users is wireless broadband more environmentally friendly.
Options for wireless broadband delivery
Wireless broadband services could be provided by selling allocations of spectrum space to wireless broadband operators to use “older” systems such as LTE, which may be inefficient.
With a newer approach – cognitive radio or white space devices – the broadcast allocations could remain. A network operator providing a wireless broadband service would calculate on which frequencies such services could be transmitted, at any given time, without interfering with broadcast services.
The old and new approaches can be compared in terms of a car park. The old approach – where network operators are given continuous allocations of channels for wireless broadband – is to pre-assign all the parking spaces, whether there are any cars there or not. The new approach – where network operators apply cognitive radio for wireless broadband – is to look around the car park to see where the spaces are and take them up as needed. Such a system, if proven, could make much more efficient use of the public’s radio spectrum.
The national and public interests
Broadcasters hope that all necessary studies will be completed before another part of the UHF broadcasting spectrum is considered for allocation to the mobile service. If WRC-12 does include, under its agenda item 8.2, potential allocations for consideration by a future WRC, ITU-R must be tasked to carry out and complete these studies, which should include capacity considerations, service concepts and, especially, the complete compatibility considerations with the existing broadcasting service (and the secondary usage of the band).
People will certainly need all three kinds of service in the years ahead: broadcast, broadband, and digital telephony. The national interest is to arrange spectrum use so that it is proportional to the benefits and needs of each kind of service. The public interest is to achieve Internet for everyone – with a capacity that will take account of future demands – without jeopardizing the future of terrestrial broadcasting. To meet future capacity demands in a sustainable way, the provision of fixed fibre-lines to off-load data traffic from mobile networks seems inevitable. The most efficient technology for wireless broadband, where needed, is likely to be cognitive radio.
The case for extending wireless broadband further into the current broadcast bands appears to be based more on short-term profit than on long-term public interest. It does not take into account the quality evolution of television, the increasing public demand for Internet data capacity, or the latest thinking in wireless broadband delivery technology.