Chris Weck, Institut für Rundfunktechnik GmbH Andrew Oliphant, Head of Transmission Systems Group, BBC R&D Department Raul Schramm, Institut für Rundfunktechnik GmbH

Digital terrestrial television broadcasting is already a reality in many European countries and around the world. The terrestrial digital video broadcasting (DVB–T) standard is gaining ground as an approved solution for digital terrestrial television broadcasting. In countries where terrestrial television is still dominant, application of DVB-T is expected to provide more channels and increase choice for the consumers. In other countries where there is a much higher penetration of cable and satellite and less reliance on terrestrial broadcasting, DVB-T is expected to offer mobile reception of video, Internet and multimedia data. This could bring new features to broadcast networks, making applications and services accessible and usable by anyone, anywhere, anytime, for business or personal use.

The DVB–T specification was completed in 1996, following an in-depth system evaluation in the Task Force on System Comparison — an entity of the DVB Technical Module. It was based on research and development within the European "RACE project 2082 dTTb" (digital terrestrial television broadcasting) and other national and international projects that were associated with dTTb. Before DVB-T could be turned from a paper specification into a system, broadcasters, network operators and manufacturers had to evaluate its performance. This triggered a long process of validation carried out through field trials and tests in pilot networks and, later, in regular networks.

The first important step to validate DVB–T was made in the framework of a project known as VALIDATE (Verification And Launch of Integrated Digital Advanced Television in Europe). VALIDATE was part of a programme called ACTS (Advanced Communications Technologies and Services) supported by the European Commission through its Fourth Framework Programme (1994–1998) agreed under the Maastricht Treaty. The VALIDATE project started work in November 1995 when the DVB-T specification was close to being finalized. Its twenty partners formed a "virtual laboratory" extending across nine European countries, with close links to the DVB project and to broadcasters and manufacturers around the world.

Mobility issues of DVB–T were taken up in a follow-on ACTS project known as "MOTIVATE" (1998– 2000). This work continued in the Multimedia Car Platform (MCP) project with a focus on mobile services and, later, in the "CONFLUENT" project, where again a variety of reception tests were performed using new diversity DVB–T receivers. The results of these tests were impressive; with very high performance for portable indoor and mobile reception.

The goal of the VALIDATE project was to prepare the launch of digital terrestrial television services and to verify the DVB–T specification. This verification required three elements:

• Checking that the specification was clear and unambiguous by demonstrating interworking between simulations and then between real hardware produced by separate laboratories.
• Checking that the system performed as expected in the repeatable conditions of the laboratory.
• Checking that it met broadcasters’ expectations in field trials.

This vital task of verification had been completed by the end of 1996; with the result that the DVB–T specification was rapidly approved as a standard of the European Telecommunications Standards Institute (ETSI). The work was described by Nokes et al [1]. Since then, further field trials of different network configurations have been carried out in many European countries — a compendium of the results has also been published [2]. These field trials have shown not only that the DVB–T specification meets broadcasters’ current expectations, but they have also opened the possibility of mobile reception.

The laboratory tests and field trials reported by VALIDATE formed the basis of international agreements on coordination of digital TV transmitters agreed by 32 countries at a meeting held in Chester (United Kingdom) in July 1997 under the auspices of the European Conference of Postal and Telecommunications Administrations (CEPT) [3]. Detailed results of the VALIDATE tests are not given here because they have been superseded by more recent work.

VALIDATE has also carried out a wide range of other work on distribution networks, transmitters, service planning parameters, single frequency networks (SFN) and gap-fillers both for professional and domestic use. Much of this work has been documented in "Implementation Guidelines", prepared for the DVB Project and now published as an ETSI Technical Report [4].

These guidelines draw attention to the technical questions that need to be answered in setting up a DVB–T network and offer some ways forward. They explain the DVB–T specification and the basic characteristics of transmission networks; they then cover transmitters and issues of sharing with existing services, distribution networks, SFN operation and network planning.

In June 1998, VALIDATE performed the first interworking test of a wide range of DVB–T equipment from different manufacturers within and outside the project. Sixty-one different DVB–T modes were tested. These included examples of all the possibilities and options offered by the DVB–T specification. For the first time, the interoperation of hierarchical modes and remotely synchronized SFN operation [5] were demonstrated successfully using modulators from different manufacturers.

The successful results of all of these tests prove the interoperability of DVB–T equipment from different manufacturers.

Network operators can safely mix equipment from different manufacturers in their networks. These results provide a sound basis for the launch of commercial services.

VALIDATE partners also tested professional gap-fillers that could be installed by a network operator to fill gaps in the coverage area of a main transmitter caused by shadowing from terrain or large buildings. Furthermore, domestic gap-fillers were tested that could be installed within a house to improve portable reception.

So in just less than three years, VALIDATE verified the DVB–T specification and provided test results for reliable service planning and international coordination. VALIDATE completed its work at the end of June 1998. More details are given in the project’s final report at http://www.bbc.co.uk/rd/projects/validate/.

Although DVB–T was not developed for mobile reception, it appears to work also in mobile environments. Mobile reception opens up new possibilities for digital terrestrial broadcasting, offering value-added services that could make terrestrial broadcasting an attractive proposition; even in countries where there is substantial penetration of cable and satellite. MOTIVATE, a collaborative project of the European Union and the successor to VALIDATE, placed special emphasis on mobility [6]. MOTIVATE has investigated the practical and theoretical performance limits of DVB–T for mobile reception. For this purpose also, MOTIVATE defined new criteria for evaluating the "quality of service"— the so-called "subjective failure point" (SFP) based on one erroneous second in a twenty second period (this object is also currently discussed in ITU Working Party 6E of the Radiocommunication Sector (ITU–R)). MOTIVATE did perform a lot of laboratory tests and field trials at different test sites and with different receivers. Doppler frequency, deep fading and shadowing — these are the dominant factors that decrease the system performance in mobile environments. Therefore, the optimization of receiver algorithms for channel estimation, channel correction and time synchronization are the key issues for mobile receivers. MOTIVATE studied network topologies and service planning constraints and also prepared guidelines for broadcasters and network operators on how to implement DVB–T networks for mobile receivers [7].

The CONFLUENT project focused on improving receivers specially optimized for mobile reception in order to enhance the quality of mobile reception (see http://www.brunel.ac.uk/project/confluent/). CONFLUENT has succeeded in linking up to the results already achieved by the MCP project and in realizing further enhancements in diversity receivers. By way of illustration, it has been possible to reduce significantly the necessary minimum carrier-to-noise (C/N) ratio and to maintain this at a constantly low level over a high-speed range. Furthermore, these receivers are no longer pure prototype developments but pre-production series devices, which will be mass-produced in the foreseeable future.

One of these diversity receivers was demonstrated by the Institut für Rundfunktechnik (IRT) at the IFA 2003 in Berlin, by means of a direct comparison with a standard DVB–T receiver. Both receivers were fitted into a car and showed the same television programme, broadcast from the same transmitter tower in Berlin. During that demonstration, absolutely no distortions occurred using the diversity receiver; even if the reception of the single antenna system failed completely. This proves the superiority of the diversity receiver over the single antenna system in a convincing manner.

The improved performance of DVB-T receivers using diversity reception was measured at the IRT, in terms of C/N values too. For mobile reception, the maximum speed at which there is still good reception with a diversity receiver, increases too, compared to a single antenna receiver.

Some more work has still to be done in order to find more reliable planning parameter values for single and diversity reception, especially towards a revision of Recommendation ITU–R BT. 1368-3 on the "Planning criteria for digital terrestrial television services in the VHF/UHF bands" approved at an ITU meeting in Geneva in March 2003.

In November 2002, the gradual conversion from analogue television to DVB–T in Germany was started in Berlin. The introduction of DVB–T in Germany is based on a concept which can be called "start-up-islands" (Startinseln). This means that DVB–T will first be introduced in selected areas, for example, large cities or highly populated areas to perform a technically perfect and controlled switchover from analogue television. After this process, the coverage of larger areas should follow later. The technology of single frequency networks is used in order to guarantee a reliable reception for all viewers in the selected islands.

With the availability of high-power transmitters for portable indoor reception in Berlin, a new basis was given for measuring and validating the planning parameters of DVB–T. Using the transmission channel parameters, which are defined in "The Chester 1997 Multilateral Coordination Agreement", the coverage prediction for fixed reception (roof-top antenna) was fairly accurate. However, measurements of the IRT in Munich and the Rundfunk-Betriebstechnik (RBT) in Nuremberg in collaboration with three programme providers have shown that the predicted coverage for portable indoor reception based on Chester 97 has been much too optimistic. On average, ten decibels more power would be needed to fulfil the requirements of indoor reception in very hostile dense urban conditions. This is partly applicable for portable outdoor reception.

A lot of indoor reception measurements made in 90 different apartments in Berlin confirmed that the reality meets the predicted coverage when a ten-decibel higher transmitting power is considered compared to Chester 97.

Field trials were also carried out to compare the performance of the VHF single frequency network (16QAM 3/4) with the UHF single frequency network (16QAM 2/3). The main parameters having a big impact on portable indoor reception are the height loss in dense urban areas, the building penetration loss and the carrier-to-noise ratio in a transmission channel for portable indoor reception.

The IRT analysed transmission measurements made in a large number of apartments in order to find more accurate values for the planning parameters [8]. One outcome of this analysis was that, even for multi-path propagation, the receiving antenna polarization has to be the same as the transmitter antenna polarization.

Laboratory simulations of portable DVB-T reception in realistic, time-varying transmission channels made in the IRT [9], show that carrier-to-noise values for 16QAM 2/3 are about four decibels higher than previously considered, but also that new generation receivers have a much lower implementation loss of only one decibel, compared to three decibels for receivers of the first generation.

As a result of a long validation process performed by many international partners such as broadcasters, network operators and manufacturers in several European projects, DVB–T is a widely approved solution for digital terrestrial television. Furthermore, there are various DVB–T networks in operation in numerous countries confirming the high performance and attraction of DVB–T for digital terrestrial television. The performance of DVB–T receivers has also been improved significantly for single antenna as well as for two-antenna diversity receivers for mobile and portable indoor reception. Diversity receivers will help to compensate for the optimistic planning parameters of Chester 97 for portable indoor reception.