The story of the DVB Project began in 1990. Experimental projects such a SPECTRE showed that the digital video compression system known as "Motion Compensated Hybrid Discrete Cosine Transform Coding" was highly effective in reducing the transmission capacity required for digital television. Up until then, digital television broadcasting was thought to be impractical to implement commercially. In the United States, the first proposals for digital terrestrial HDTV were made. In Europe, the Swedish Television suggested that fellow broadcasters should form a pan-European platform to develop digital terrestrial HDTV. During 1991, broadcasters and consumer equipment manufacturers discussed how this could be done. Broadcasters, consumer electronics manufacturers and radio-regulatory bodies agreed to come together to discuss the formation of a pan-European group that would oversee the development of digital television in Europe - the European Launching Group (ELG). Over the course of about a year, the ELG expanded to include the major European media interest groups, both public and private, the consumer electronics manufacturers and common carriers. It drafted a Memorandum of Understanding (MoU) establishing the rules by which this new and exciting game of collective action would be played. The MoU was signed by all ELG participants in September 1993, and the Launching Group became the DVB (Digital Video Broadcasting) Project. DVB provided the forum for gathering all the major European television interests into one group. It promised to develop a complete digital television system based on a unified approach. It was now clear that digital satellite and cable television would provide the first broadcast digital television devices. Fewer technical problems, and a simpler regulatory climate meant that it could develop more rapidly than terrestrial systems. Market priorities meant that digital satellite and cable broadcasting systems would have to be developed rapidly.

Because DVB System provides a complete solution for digital television and data broadcasting across a range of delivery media. Careful planning and common sense have been the keys to success. From the outset, it was clear that the sound and picture-coding systems of ISO/IEC MPEG-2 should form the audio and image-coding. DVB would need to add, to the MPEG transport stream multiplex, the necessary elements to bring digital television to the home through cable, satellite and terrestrial broadcast systems. Interactive Television is also an attractive service and DVB might help with the framework for the interactive television services of the future.

The digital vision and sound-coding systems adopted for DVB use sophisticated compression techniques. The MPEG standard specifies a data-stream "syntax". The system designer is given a "tool-box" from which to make up systems incorporating greater or lesser degrees of sophistication. In this way, services avoid being over-engineered, yet are able to respond fully to market requirements and are capable of evolution.

The sound-coding system specified for all DVB systems is the MPEG audio standard. The DVB Project suggests, in its current MPEG-2 Guidelines document, that MPEG Layer II (MUSICAM) should be used. This is also used for many other audio products and services throughout the world. MPEG Layer II is a digital compression system which takes advantage of the fact that a sound element will have a masking effect on lower-level sounds (or on noise) at nearby frequencies. This is used to facilitate the coding of the audio with low data rates [3]. Sound elements which are present, but would not be heard even if reproduced faithfully, are not coded. The MPEG Layer II system can achieve a sound quality which is very close to Compact Disc. The system can be used for mono, stereo, or multi-lingual sound, and in future MPEG-2 audio can be used for surround sound [3].

DVB standards are open and inter-operable. Following approval by the Steering Board, DVB specifications are offered for standardization to the relevant standards body (ETSI or CENELEC), through the ETSI/EBU/CENELEC JTC (Joint Technical Committee) and the ITU-R, ITU-T and DAVIC. For each delivery system, a set of User Requirements is drawn up by the relevant Commercial Module. These are used as constraints on the specification. For example, in the case of DVB-T, the Terrestrial Commercial Module proposed user requirements that outline broad market parameters for a DVB-T system (price-band, user functions, etc.). The Technical Module then developed the specification, following the user requirements. The approval process within DVB requires that the relevant Commercial Module support the specification before it is finally approved by the Steering Board.

In addition DVB has developed DVB Receiver Guidelines, suggestions for Interfaces for the domestic receiver, and a Common Interface specification intended for conditional access applications. Both interface documents were agreed by the DVB Steering Board in March 1995 and are now with CENELEC for standardization [5].

DVB-S has been acknowledged by the ITU in a Recommendation for the broadcast transmission of digital television for 11/12 GHz satellites. DVB-C is also included in an ITU Recommendation. Technical specifications in the areas of subtitling, MVDS (Multipoint Video Distribution System) and interactive television are being submitted to the appropriate standards body.

Copies of the Standards are available from ETSI and the EBU (see Table 1):

Baseline System for modulation and channel coding for satellite digital multi-programme TV/HDTV services to be used for satellite digital TV for primary and secondary distribution in FSS and BSS bands[9] The Baseline System is intended to provide Direct-to-Home (DTH) services for consumer Integrated Receiver Decoders (IRDs) as well as collective antenna systems (SMATV) and cable television head-end stations. The Baseline System uses QPSK modulation and concatenated error protection strategy based on a convolutional code and a shortened Reed-Solomon code. The Baseline System is suitable for use on different satellite transponder bandwidths. Compatibility with MPEG-2 coded TV services, with a transmission structure synchronous with the packet multiplex, is provided. The multiplex flexibility allows the use of transmission capacity for a variety of TV service configurations, including sound and data services. All service components are time multiplexed (TDM) on a single digital carrier.

Baseline System for a Digital multi-programme Television by Cable (DTVC) [10]. Harmonized transmission standard for cable and satellite, based on the MPEG-2 System Layer, with the addition of Forward Error Correction (FEC) technique. It is compatible with the modulation/ channel coding system used for digital multi-programme television by satellite (DE/JTC-DVB-6). The System allows for further evolution as technology advances and is capable of starting a reliable service as of now. The Baseline System is based on QAM modulation scheme. It allows for 16, 32 or 64-QAM constellations and permits future extension to higher constellation such as 128 and 256-QAM.

The provision of a solution for the "Seamless Transition Networks" with respect to the "Network Identification Table" in the case of SMATV and some CATV networks was not covered by the ETS 300 468 (SI) Ed 1. As a consequence HISPASAT, RAI, DTI, SAT/Sagem, Sharp and the Spanish Administration proposed revision EWP items on ETS 300 468 and ETR 211. Specification of Service Information (SI) data which forms part of DVB/MPEG-2 bitstreams is defined, in order that the user can be provided with information to assist in the selection of services and/or events within the bitstream, so that the Integrated Receiver Decoder (IRD) can automatically configure itself for the selected service. SI is specified in ISO/IEC 13818-1 as Programme Specific Information (PSI)[6]. This specification complements the PSI by providing data to aid automatic tuning of IRDs, and additional information intended for display to the user, mainly in text form. The form of presentation of the information is not specified, and IRD manufacturers have freedom to choose appropriate presentation methods. It is expected that Electronic Programme Guides (EPG) will be a feature of Digital TV transmissions. The definition of EPG is outside the scope of the SI specification, but the data contained within the SI specified here could be used as basis for an EPG. The present specification describes Service Information (SI) for use in broadcast MPEG-2 bitstreams. The MPEG-2 System layer specifies SI which is referred to as Programme Specific Information (PSI). The PSI data provides information to enable automatic configuration of the receiver to demultiplex and decode the various streams of programme within the multiplex.

The basic building blocks which are part of the DVB System are the MPEG-2 data packets. These are fixed-length containers with 188 bytes of data. MPEG includes Program Specific Information (PSI) so that the MPEG-2 decoder can capture and decode the packet structure. This data, transmitted with the pictures and sound, automatically configures the decoder and provides the synchronization information necessary for the decoder to produce a complete video signal at its output [6]. MPEG-2 also allows a separate Service Information system to be used to complement the PSI. DVB has prepared an open Service Information system to accompany DVB signals. It can be used by the decoder and the user, to navigate through the array of services offered.

Key data, necessary for the DVB IRD (Integrated Receiver Decoder) to automatically configure itself, is available in the MPEG-2 PSI. DVB-SI adds information that enables DVB IRDs to automatically tune to particular services and allows services to be grouped into categories with relevant schedule information.

In a DVB environment, the viewer of tomorrow will be receiving a multitude (perhaps hundreds) of channels with his IRD. These services could range from interactive television, to near video-on-demand, to specialized programming. The viewer will need help. DVB-SI provides the elements necessary for the development of the Electronic Programme Guides (EPG) that are likely to become a feature of the new digital television services. More elaborate EPGs may also be provided, perhaps as additional elements via a receiver interface.

DVB-SI is based on four tables, plus a series of optional tables. Each table contains descriptors outlining the characteristics of the services/event being described. The four tables are:

In addition there are three optional SI tables:

With these tools, DVB-SI covers the range of practical scenarios. Seamless transition between satellite and cable networks, near video-on-demand, and all operational configurations, are possible.

Digital terrestrial broadcasting is one of the cornerstones of DVB. The DVB-T system specification, for the terrestrial broadcasting of digital television signal, was approved by ETSI in February 1997 [4].

The Coded Orthogonal Frequency Division Multiplexing (COFDM)-based system, proposed in DVB-T, allows the use of either 1705 carriers or 6817 carriers. Reed-Solomon outer coding and outer convolutional interleaving are used, in common with the other DVB standards. The inner coding is the same used for DVB-S. The data carriers in the COFDM frame can use QPSK and different levels of QAM modulation. The modulation system combines OFDM with QPSK/QAM.

Modulation and channel coding system for the distribution of digital multi-programme Television (TV)/High Definition Television (HDTV) by Multipoint Video Distribution Systems (MVDS) in the 40 GHz band. The System described in this ETS is based on that described in ETS 300 421 for 11/12 GHz satellite services. It allows the same consumer Integrated Receiver Decoder (IRD) to be used for either service, when used with a Low Noise Block (LNB) down-converter for the appropriate frequency band. The frequency band 40,5 to 42,5 GHz has been harmonized within CEPT under Recommendation T/R 52-01. The System however, is applicable to other frequency bands above 10 GHz. The System uses Quaternary Phase Shift Keying (QPSK) modulation and concatenated error protection strategy based on a convolutional code and shortened Reed-Solomon (RS) code. The System is suitable for use on different MVDS transmitter bandwidths. Compatibility with Moving Pictures Experts Group-2 (MPEG-2) coded TV services (see ISO/IEC IS 13818-1), with a transmission structure synchronous with the packet multiplex, is provided. Exploitation of the multiplex flexibility allows the use of the transmission capacity for a variety of TV service configurations, including sound and data services. All service components are Time Division Multiplexed (TDM) on a single digital carrier.

Modulation and channel coding system for the distribution of digital multi-programme Television (TV)/High Definition Television (HDTV) by Multipoint Video Distribution Systems (MVDS) below 10 GHz. The System described in this ETS is based on that described in ETS 300 429 on framing structure, channel coding and modulation for cable systems. The frequency band may/will be harmonized within CEPT. The System is suitable for use on different MVDS transmitter bandwidths. Compatibility with Moving Pictures Experts Group-2 (MPEG-2) coded TV services (see ISO/IEC IS 13818-1), with a transmission structure synchronous with the packet multiplex, is provided. Exploitation of the multiplex flexibility allows the use of the transmission capacity for a variety of TV service configurations, including sound and data services. All service components are Time Division Multiplexed (TDM) on a single digital carrier.

Transmission System proposal for Digital multi-programme Television by suitable for distribution in Satellite Master Antenna Television (SMATV) systems [12]. This ETS is complementary to ETS 300 429 (Cable) and it is aligned with ETS 300 421 (Satellite). The system described in this ETS is compatible with the modulation and channel coding systems used for digital multi-programme television by cable and satellite transmission. The system is based on the MPEG-2 System Layer, with the addition of appropriate Forward Error Correction (FEC) technique. The system allows for further evolution as technology advances and is capable of starting a reliable service as of now.

DVB has worked in developing a set coherent practical specifications for an Interactive TV System. This system is divided in two standards:

DVB-Data is seen as one of the new applications for DVB. Data broadcasting involves file downloading, data carousels, and protocol tunneling for a wide range of applications including INTERNET delivery and the broadcast of secure private data [19]. This standard is in drafting status. This specification is designed to be used in conjunction with the DVB-SI.

Support for use of scrambling and Conditional Access (CA) within digital broadcasting systems [14], meant as a marketing tool to attract interested parties for applying/receiving the "CS system specification", which will be released to qualified applicants on receipt of a non-disclosure agreement. This ETR does not include "Security" issues.

The Common Scrambling Algorithm was designed to minimize the likelihood of piracy attack over a long period of time. The specification, prepared by the Conditional Access Specialists' Group, is lodged with the European Telecommunication Standards Institute (ETSI) as custodian. The technical details are

distributed, and the technology licensed, by the custodian to appropriate organizations upon signature of a license agreement.

By using the Common Scrambling Algorithm system in conjunction with the standard MPEG data transport and selection mechanisms it is possible to incorporate in a DVB transmission the means to carry multiple messages which all enable control of the same scrambled broadcast but are generated by a number of different CA systems. This 'Simulcrypt' technique allows both the delivery of one Programme to a number of different decoder populations that contain different CA systems, and also for the transition between different CA systems in any decoder population, for example to recover from piracy. The 'Multicrypt' option is also available, facilitated by the Common Interface (DVB-CI) specification proposed for standardization by CENELEC (European Committee for Electrotechnical Standardization).

The Common Interface operates at the MPEG Transport Stream level, and although specifically intended for conditional access, it may also prove useful in other applications, such as for Electronic Program Guides. The CA functions of decryption and descrambling take place in a small plug-in module, similar in size and appearance to a PCMCIA card as popularly used with lap-top PCs. This CA Module may operate in conjunction with a smartcard through an associated smart-card reader [14]. The DVB Project has agreed to an Important procedure, or Code of Conduct, which should help service providers gain access to markets where decryption systems other than their own are in use. The DVB Project has also adopted a series of Recommendations on Antipiracy Legislation, which it has submitted to the European Commission and other European bodies. These most notably call for criminal penalties for the commercialization and use of pirated decoder devices. The area of Conditional Access has received particular attention within DVB. Discussions have prove to be difficult and lengthy, but common consensus has yielded a package of measures.

The DVB systems Cable (DVB-C), Microwave (DVB/MVDS) and Satellite (DVB-S) are very similar, using the same parameters, little changes at the RF transmission system and in the Error Correction System [7]. A model of digital satellite system is shown in the Figure 1 "DVB-S Transmission System". In this example, the system translates the output of the MPEG-2 transport stream multiplexer (see Section 3.1) to a signal appropriate for a satellite channel. Although the following system description, based on European Telecommunication Standards, assumes a single carrier per Transponder with multiple services carried as packetized data in the time-division multiplex (TDM) of the transport data stream, multicarrier systems using frequency-division multiple access (FDMA) techniques, each carrying a transport data stream, are also possible. This technique was also used for the Skyplex system but added some changes in the ground stations.

The PES packets are organized into a one or various Transport Packet in the Transport Stream Multiplexer (see Figure 2). A transport packet is always 188 bytes (1504 bits) long. The content of each transport packet (see Figure 3) is a 4-bytes Header followed by an Adaptation Header (used to fill the 188 bytes of the transport packet, if is necessary) and a Payload (184 Bytes without Adaptation Header).

The output data of the MPEG-2 Transport stream Multiplexer is randomized to ensure adequate binary transitions to comply with ITU Radio regulations. The data are randomized by a pseudo-random sequence polynomial generator (X¹⁵ + X¹⁴+1) as shown in Figure 4, with the generator initialized to a value 00A9H ((MSB)000 0000 1010 1001) every eight transport packets [1,2].

Figure 4. Pseudo-Random Sequence Generator.

All bytes of the transport stream are randomized with the exception of the sync bytes (see Figure 5). The receiver descrambler is provided with an initialization signal by forcing an inversion of the sync byte in the first transport packet in the group of eight to a value of B8_HEX(is equal inverse of the sync byte (47_HEX)). This inverted sync byte is followed by the first (MSB) bit of the first byte from the pseudo-random sequence generator. The randomization process continues even when the bit stream is absent or does not conform to the MPEG-2 Transport stream format, to ensure that the carrier is properly modulated.

Error Corrective Coding is needed to preserve the integrity of received data. The basic concept is to add redundancy to a message in order to eliminate the need for retransmission of the data. Shannon's work showed that dramatic gains in performance could be achieved by intelligently adding redundancy to a message. This addition of redundancy to the message is called Forward Error Correction (FEC). There are many methods of adding redundancy to a message, but one of the most efficient and popular methods is Reed-Solomon outer coding combined with convolutional inner codes.

including magnetic and optical storage systems, space and mobile communications, etc.. It is important to be familiar with some of the terminology for describing Reed-Solomon codes. The mathematics underlying RS codes is very complex and it is beyond the scope of this paper. A Reed-Solomon code is described as an (n,k) code, where the codewords consist of n symbols of which k are message symbols. The following parameters are important in the characterization of the codes:

The outer coding and interleaving shall be performed on the input packet structure (see figure 3). Reed-Solomon RS(204,188, t=8) shortened code, derived from the original systematic RS(255,239, t = 8) code, shall be applied to each randomised transport packet (188 byte) of figure 5 to generate an error protected packet (see figure 6). Reed-Solomon coding shall also be applied to the packet sync byte, either non-inverted (i.e. 47_HEX) or inverted (i.e. B8_HEX) [9].

The RS code has symbol size m=8 (one byte per symbol), which means that the symbols are elements of the Galois Field GF (28) and the maximum codeword length will be 28-1=255. It has 16 symbols of CRC and thus can correct 8 symbols of errors per codeword. The codeword length is programmable by the user. The implemented RS code has generator (Code Generator Polynomial) G(x)=(x-a⁰)(x-a¹)...(x-a¹⁵), where a (a=02_HEX) is a root of the binary primitive polynomial (Field Generator Polynomial ) P(X)= x⁸ + x⁴ + x³ + x² + 1.

Each RS codeword consists of k message bytes (Mk-1,Mk-2,...,M0) and 16 CRC bytes (C15, C14,...,C0). The check polynomial C(x)=C15x15+C14x14+...+C0 is obtained as the remainder when the message polynomial M(x)=(Mk-1xk-1+Mk-2xk-2+...+M0) * x15 is divided by G(x).

It is divided into four steps:

The polynomial division in the encoding process is implemented by a 16-stage 8-bit shift register. Between any two consecutive shift registers, there is a 8-bit XOR for the binary finite field addition. The feedback path containing the quotient for each division step, is broadcasted to 16 constant finite field multipliers. The message bytes are pipeline from input, and the first k bytes of output are same as input, while last the 16 CRC bytes are from the data stored in the 16 shift registers after all the message bytes have been computed. Figure 7 shows the block diagram of the encoder architecture.

Following the conceptual scheme of figure 1, convolutional byte-wise interleaving with depth I = 12 shall be applied to the error protected packets (see figure 6). This results in the interleaved data structure (see figure 8).

The convolutional interleaving process is based on the Forney approach which is compatible with the Ramsey type III approach, with I = 12. The interleaved data bytes shall be composed of error protected packets and shall be delimited by inverted or non-inverted MPEG-2 sync bytes (preserving the periodicity of 204 bytes). The interleaver may be composed of I = 12 branches, cyclically connected to the input byte-stream by the input switch. Each branch j shall be a First-in, First-out (FIFO) shift register, with depth j ´  M cells where M = 17 = N/I, N = 204 (see Figure 9). The cells of the FIFO shall contain 1 byte, and the input and output switches shall be synchronized. For synchronization purposes, the SYNC bytes (47_HEX) and the bytes (B8_HEX) shall always be routed in the branch "0" of the interleaver (corresponding to a null delay) [9].

The DVB system allow for a range of punctured convolutional codes, based on a rate 1/2 convolutional code with constrain length K=7 (see Figure 10). This allow selection of the most appropriate level of error correction for a given service data rate.

The system allows convolutional coding with code rate of 1/2, 2/3, 3/4, 5/6 and 7/8 as shown in the table below.

The different rates are constructed by periodically deleting specified bits from the bits contained in the consecutive blocks of the original 1/2 rate code (see Figure 11).

Prior to modulation, the I and Q signals (mathematically represented by a succession of Dirac delta functions spaced by the symbol duration Ts=1/Rs, with appropriate sign) are square-root raised cosine filtered. The roll-off factor a is 0.35. The Baseband square-root raised cosine filter have a theoretical function defined by the following expression:

for

for

where:

is the Nyquist frequency and = 0.35 (roll-off).

Finally in the DVB-S, the coded bits are Gray mapped in the QPSK constellation ( see Figure 12) with absolute mapping (no differential coding).

Figure 12. QPSK constellation

The available Transponder bandwidth is defined by the combined effect of the input (IMUX) and output (OMUX) filters. The Figure 13 shows, as an example, the satellite amplifier characteristics of an INTELSAT VIII transponder.

The ratio BW/Rs (Bandwidth/Symbol rate) determine the symbol rate transmitted in the satellite transponder for any particular bandwidth. Decreasing BW/Rs means that the symbol rate is increased and therefore the capacity available to transmit Programmes. However, there is a BW/Rs lower limit caused by the acceptable distortion (ISI, Inter Symbol Interference) introduced by the satellite IMUX and OMUX filter. The problem of choosing an optimum BW/Rs value is equivalent to estimating what is acceptable as a C/N degradation compared to the gain in symbol rate and useful bit-rate. Figure 14. gives an example of the Eb/No degradation at BER=2.0E-04, due to bandwidth limitations (IMUX and OMUX). The reference 0 dB degradation refers to the case of a satellite transponder without bandwidth limitations (BW=), and with saturated TWTA (OBO=0 dB). The inner coding rates 2/3 and 7/8, associated with QPSK modulation, have been analyzed. For example, with the adopted IMUX and OMUX filter responses, BW/RS value between 1.20 and 1.14 (see Figure 14, QPSK rate 7/8), corresponding to a ISI degradation of about 0.2 dB and 0.5 dB, respectively. When a 7% margin is allowed on BW to protect against transponder instabilities, more conservative figures of BW/Rs=1.28 and 1.22 are achieved.

The required Eb/No and C/N to achieve "quasi error free" (QEF) bit-stream at the MPEG-2 demultiplexer input, corresponds to a BER = 1E-10 to 1E-11 after RS decoding, and BER = 2E-04 after viterbi decoding (see Table 2). The Eb/No values refer to the useful bit rate Ru before RS coding and include the noise bandwidth due to the outer code (10 log 188/204 = 0.36 dB). In the case of satellite transmission, the TWTA was operated in saturation ( Output Back Off (OBO)= 0 dB) and a typical Eb/No and C/N degradation of about 1dB (including TWTA, IMUX and OMUX distortions and power losses) has been obtained for all the code-rates and BW/Rs=1.28.

The DVB System has very steep failure characteristic (about 0.9 dB of C/N variation to pass form QEF to service outage). So the satellite power design should not be based on the conventional criteria used for TV/FM services, PAL, SECAM, NTSC (e.g., quality grade 3.5 of the ITU-R 5-level quality scale for 99% of the worst month at the coverage are contours), but on extended service continuity targets (e.g., more than 99.7% of the worst month, corresponding to about 99.9% of the average year) . Assuming a typical propagation condition at 12 GHz in Europe (i.e., Geneva (climatic zone K)), a C/N increase of about 2 dB is required to extend the service continuity form 99% to 99.7% of the worst month.

The Antenna size is another factor very important. The next table (Table 3) shows the impact on the usable bit-rate (Ru) and inner code-rate, in the range 1/2 to 7/8, on the antenna size required at 12 GHz for 99.9% (average year) service availability (Quasi-Error-Free quality), in climatic zone E (in Europe, Berlin). The main link budget assumptions are:

-Antenna pointing losses = 0.5 dB

-External Interference = 1.0 dB

-Antenna Efficient = 70%

-Antenna Noise Temperature = 35°K

-Coupling Losses = 0.2 dB

-Satellite EIRP = 51 dBW

-LNB noise figure = 1.1 dB.

The phrase "Conditional Access" describes the system by which service providers control viewer access to different Programmes and channels, so that only those who paid the appropriate charges can actually watch the services. Conditional Access is achieved by the service provider scrambling its available TV services, whilst providing subscribers with an Integrated Receiver Decoder (most commonly known as a "set top box"), that allows the subscriber to unscrambler the picture, if he or she has the correct smartcard available and has paid to receive the services. A conditional access (CA) system comprises a combination of scrambling and encryption to prevent unauthorised reception. Scrambling is the process of rendering the sound, pictures and data unintelligible. Encryption is the process of protecting the secret keys that have to be transmitted with the scrambled signal in order for the descrambler to work. After descrambling, any defects on the sound and pictures should be imperceptible, i.e. the CA system should be transparent [14].

The basic process of scrambling and descrambling the broadcast MPEG-2 transport stream [2] is shown in Fig. 15. The European DVB Project has defined a suitable, highly-secure, Common Scrambling Algorithm.

Figure 15. Conditional Access Scrambling, Encrypted Control Word, EMM and SMS.

The generation, transmission, and application of Entitlement Management Messages (EMMs) by the Subscriber Authorisation System (SAS) is illustrated in Fig 15. The card supplier provides the CASS (Conditional Access Sub-System) (usually a smart card). The SAS sends the EMMs over-air or by another route, e.g. via a telephone line. To retain the confidentiality of customer information, it is best that the card supplier delivers the smart cards direct to the Subscriber Management System (or another business centre which guarantees confidentiality) for mailing to the viewer (see Section 4.5.). It is possible to supply the cards through retail outlets as well, provided the retailer can guarantee confidentiality. In this situation, considerable care has to be taken if the cards have been pre-authorised by the SAS, because such cards will be a worthwhile target for theft. When using the CA Common Interface [16], in conjunction with a PCMCIA module acting as the CASS, the descrambler is also situated in the CASS.

As shown in Figure 15, the reference model is completed by the addition of a Subscriber Management System (SMS), which deals with the billing of viewers and the collection of their payments. The control word need not be a decrypted ECM; it can be generated locally (e.g. from a seed) which means that the control word could be changed very quickly.

The SMS is primarily responsible for sending out bills and receiving payments from viewers. It does not need to, and should not, be specific to a particular CA system. The SMS necessarily holds commercially-sensitive information such as the database of subscribers names and addresses and their entitlement status. Sharing of the SMS between rival broadcasters is possible if, and only if, it is operated by a trusted third party and only if adequate ‘firewalls’ are provided so that any one service provider can access information only about subscribers to his or her own services. Although sharing an SMS may be seen as undesirable, it must be recognised that setting up and running an SMS is expensive, perhaps prohibitively so for services with a small number of subscribers at the outset. The work of the SMS can be contracted out to a trusted third party (TTP), e.g. a secure and reliable organisation such as a bank. There should be a subscriber database and the system must deal with changes to subscription details, installation difficulties, marketing, billing and card distribution. The SMS also manages the system installers and sends Entitlement Management Messages (to authorise viewers) to the Subscriber Authorisation System queue. To ensure the privacy of customer database information, the SMS could mail replacement smart cards and CA modules to the viewers. These could be provided pre-authorised by the conditional access system operator or could be authorised over-air by the Subscriber Authorisation System using virtual addresses provided by the SMS.

The Subscriber Authorisation System (SAS) is primarily responsible for sending out the over-air entitlement messages and for validating security devices. The SAS needs a unique serial number (address) for each IRD security device but should not need access to commercially-sensitive information such as the names and addresses of subscribers. Hence it should be easily possible for rival broadcasters to share an SAS, although there are issues to be resolved concerning the queuing times for messages. Smart cards can be indirectly authorised over-air by the SAS.

The SAS generates a scrambler control word, encrypts the conditional access data, queues and prioritises the Entitlement Management Messages from the Subscriber Management System, and scrambles the pictures and sound. New messages from the SMS join the immediate queue, to be transmitted as soon as possible, and also join a regular cyclic queue where they stay until they expire. The frequency of transmission depends on the length of the queue and messages are expired by category. There will be a maximum limit on response time beyond which the system will not be usable. Disable messages may have an indefinite lifetime whereas enable messages may have a lifetime of a month or so. For security reasons, communication between the SMS and the SAS can be encrypted although this may not be necessary if all systems are in a secure environment.

The European DVB Project has designed a Common Interface for use between the Integrated Receiver Decoder (IRD) and the CA system. As shown in Figure 16, the IRD contains only those elements that are needed to receive clear broadcasts. The CA system is contained in a low-priced, proprietary module which communicates with the IRD via the Common Interface. No secret conditional access data passes across the interface [15,16].

Figure 16. DVB Common Interface

In the case of SimulCrypt, each service is transmitted with the entitlement messages for a number of different proprietary systems, so that decoders using different conditional access systems (in different geographic areas) can decode the service. SimulCrypt requires a common framework for signalling the different Entitlement Message streams. Access to the system is controlled by the system operators. Operation of the system requires commercial negotiations between broadcasters and conditional access operators. A code of conduct has been drawn up for the operation of SimulCrypt.

The philosophy behind the system is that in one geographical area, it will only be necessary to have a single smart card or CA module and a single decoder to receive the local service. If one wanted to descramble the service of a neighboring area, one could subscribe to, and use the smart card/module for that service. Consequently, it is only necessary to have a single Subscriber Management System for a given area. When a viewer wants to watch services from two neighboring areas, it is necessary for both services to carry the entitlement messages for that viewer. Therefore it is necessary to have secure links between the different Subscriber Management Systems of the different operators to allow transfer of the entitlement messages between operators.

MultiCrypt (receiver equipped with more than one CA system) is an open system which allows competition between conditional access system providers and Subscriber Management System operators. MultiCrypt uses common receiver/decoder elements which could be built into television sets. The Common Conditional Access Interface can be used to implement MultiCrypt. Conditional access modules from different system operators can be plugged into different slots in the common receiver/decoder, using the common interface.

DSS makes use of several standards, but is not fully compliant with the European Digital Video Broadcasting (DVB) satellite convention. DSS was designed before the ratification of DVB. However, DSS should be compliant with the most difficult Interoperability aspect of DVB: video and audio representation. In fact, the video and audio bitstreams are unaffected by the transport layer protocol. An

MPEG-2 video bitstream transported over DSS packets will demultiplex identically to a bitstream transported by DVB means. The DSS system utilizes a concatenated encoding scheme. The first section is a Reed-Solomon encoder, followed by an interleaver, a Convolutional encoder, and a QPSK modulator. No randomization function is performed [13].

DSS IRDs receive the satellite signal after amplification and block down conversion at the receive antenna. The signal is transmitted via coaxial cable to the IRD at L band. Either polarization may be selected by changing the level of the block downconverter supply voltage. This power is supplied to the downconverter by the IRD via the coaxial cable.

To illustrate the makeup of the DSS System, a block diagram of the video Signal path is provided below. Note: this diagram is generic enough to apply to many other set top-box designs.

[Dish/LNB] --> [Tuner/IF demodulator] --> [A/D]--> [QPSK demodulator] --> [Viterbi FEC]

--> [Reed-Solomon FEC] --> [Systems demultiplexor] --> [Video decoder ]

--> [NTSC modulator] --> [D/A] --> [RF modulator] --> [TV]

Dish/LNB: Low Noise Block converter maps 500 MHz of the Ku band (BSS) (RF carrier around 12 GHz) onto coax cable (RF carrier in the more friendly neighborhood of 1 GHz). In the dual-LNB deluxe DSS receiver system, both polarities (set of 16 transponders) are mapped onto the same coax cable. (total ~1 GHz).

Tuner/IF Demodulator: essentially isolates one data carrier out of the total 16 carriers contained within the 500 MHz spectrum.

A/D: samples the IF analog Baseband signal for processing by the QPSK decoder.

QPSK Demodulator: converts the sampled modulated signal into a stream of discrete bits.

Viterbi decoder: performs deconvolution of the bitstream.

Reed-Solomon FEC: performs error correction using the supplied redundant 16 bytes per packet.

Systems Demultiplexor: separates audio, video, VBI (vertical blanking information data such as closed caption and teletext), program guide, and conditional access information into separate bitstreams.

Video Decoder: decodes video bitstreams into raw 4:2:0 YCbCr pictures. These pictures (whether coded as fields or frames) are output one field at a time at a sample rate of 13.5 MHz for Y (luminance) and 6.75 MHz for each of the color channels (Cb and Cr).

NTSC Modulator: converts the "CCIR 601" YCbCr format stream into a composite video stream.

D/A: separately converts the composite digital video stream and component Y/C digital video stream into analog Baseband signals. The Y/C signal is compatible with the S-Video format conveyed on the round, MiniDin 4-pin connector. The composite analog signal is additionally provided on the RCA phono-style connector. The video composite signal is also modulated on an RF carrier.

RF Modulator: modulates the analog composite video signal onto an RF carrier (channel 3 or 4), along with a (mono) analog audio subcarrier.

DVB-S specifies conformance at several levels, including channel coding (e.g. symbol rate, QPSK modulation, Reed-Solomon forward error correction, Viterbi convolution for outer error correction, packet interleaving), transport layer (MPEG-2 Systems Transport bitstreams), and elementary stream layers (MPEG-2 Video, MPEG-1 Audio) [18].

DSS elements are nearly identical to DVB. Subtle differences exist such as packet length (DSS packets are 147 bytes long whereas MPEG-2 Transport streams packets are 188 bytes long), but the more expensive implementation items are the same (modulation, error correction). The differences between the two standards are shown in Table 4 below:

Since the inception of DAVIC (the Digital Audio-Visual Council created in 1994), DVB members have understood the importance of working closely with it. DAVIC covers an extremely wide field, generally extending well outside the area of broadcasting, and DAVIC seeks to provide end-to-end Interoperability for the use of digital images and sound across countries and between applications and services. DAVIC liaison officers have been appointed in DVB to co-ordinate the efforts of both groups.

Harmonization with DAVIC has been achieved in many areas. The DVB Cable and Satellite Systems have been adopted by DAVIC. The DVB subtitling group has also reached broad agreement with DAVIC, after a good deal of discussion and a willingness to change its own draft specification. Other. harmonization agreements have been reached in areas including interactive services, receiver interfaces, and Service Information, and work will continue to ensure the greatest possible commonality.

Although DVB began as a European project, its membership has now spread around the globe, and, reflecting this, the project members now aim at achieving world standards. Liaison takes place regularly with the ITU-R and the ITU-T on the world standardization of systems developed under the DVB project, and many DVB members are long-standing members of ITU-R and ITU-T technical committees. The DVB-C cable system specification is currently being considered by ITU-T Working Party 1/9, the DVB-S satellite system specification is being considered by ITU-R Working Party 10-ll/S, and the terrestrial digital specification has been submitted to ITU-R Task Group 11/3. The DVB Steering Board has now established a Promotion and Communications Module charged with responding to the many requests being received from all parts of the world for information on the DVB systems.

Consumers worldwide paid a high price for the diversity of analogue broadcasting standards. The division of the television services worldwide between NTSC, PAL,SECAM and MAC built substantial barriers between continents and within Europe. Digital television broadcasting can change this. INTELSAT can use DVB-S standard to give to its Signatories better alternatives to non-standard systems that exist at present.

This document focuses on describing the DVB-S standard. The DVB-S standard is optimized for single carrier per transponder TDMA (Time Division Multiple Access) in order to operate the satellite TWTA close to saturation, thus optimizing the satellite power efficiency. However it could be used also for multi-carrier FDMA (Frequency Division Multiple Access) type applications, allowing better operational flexibility, by originating and uplinking the signals from different locations, but at the expense of reduced performance in terms of power and spectrum efficiency.

A follow on activity will study the new applications and standards that are being used to delivery data on top of the MPEG-2 based DVB transmission standards (Data Broadcasting specification) such as Data Piping, Data Streaming, Multiprotocol Encapsulation, Data Carousels, and Object Carousels. The most useful of these applications is the DSM-CC download protocol because this is an open protocol. These protocols will be covered in the next IOM about " Data Broadcasting".