PPress

Press

Red

To

Select

0000055555

DVB - S

CENE

Tune in to Digital C

onvergence

Ed i t ion No. 05 March 2003

This issue’s highlights

> MHP sparkles in the US with GEM

> DVB 2.0 Making Progress

> What is DVB-X?

> Advanced Audio & Video Coding

> DVB-S2 - Next generation

> Service requirements for DVB-T

> Copy Protection

The Standard for the Digital World

10

Peter M

acAvo

ck,

Execu

tive Directo

r

A word from the DVB Project Office

YEARS AFTER

The views expressed in this newsletter

are those of the individual DVB members

or guests and are not necessarily the

views of the DVB Project Office or

Steering Board.

Published by the DVB Project Office,

c/o European Broadcasting Union,

17a Ancienne Route,

CH-1218 Grand Saconnex, Switzerland.

www.dvb.org & www.mhp.org

Editors: William Daly, Harold Bergin

Editorial and Advertising enquiries to:

WHD PR

E-mail: news@whdpr.com

Telephone: +44 (0)20 7799 3100

All rights reserved. No part of this publication may

be reproduced without prior consent of the

publisher. All content correct at time of print. (c)

DVB-SCENE 2003. DVB an MHP are registered

trademarks of the DVB Project. Certain other

product names, brand names and company names

may be trademarks or designations of their

respective owners.

Design by GyroGroup, Geneva, Switzerland.

Printed by Pims Print Limited, Yeovil, UK.

Force UK Ltd

Fujitsu-Siemens Computers GmbH

Techno Trend AG

NEWMEMBERS

Cover - DVBDVBDVBDVBDVB-SCENE-SCENE-SCENE-SCENE-SCENE wishes tothank the following companies forcontributing screenshots of MHPservices: Alticast, BetaResearch, ICTEmbedded BV, MTV3 Finland, OrtikonInteractive OY, Philips, Premiere,Sony, TVC Multimedia, and YLE.

DVB turns 10 in 2003. One oftenwonders if the Project continues to haverelevance in the world of broadcasting,especially since there have been manychanges in television. Digital televisionis a reality, but the harsher realities ofthe economic situation have meant thatinteractive television, once thought of asthe panacea of the broadcaster andmanufacturers, hasn’t yet delivered onits promises.

This hasn’t stopped DVB preparing forthe day when we all “use” our TVs. MHPis the cornerstone of these preparations.It is now well under way, with thespecifications maturing, theconformance process in place anddelivering conformance test basedlogos. But what of the other initiativesaround the world to develop openmiddleware systems? Well one of these,CableLabs, announced some time agothat they would adopt MHP and workwith DVB to produce a specificationessentially permitting MHP to run ontheir particular environment. The resultis the Globally Executable MHP (GEM)specification developed by the MHPUmbrella Group (MUG). You can blamethe group’s chairman, Bill Foote (Sun)for the name, but it has raised a feweyebrows.

But that’s not all. There are other groupsworking on open middleware systemswho are following the GEM process veryclosely and we hope to be able to makeannouncements in this area shortly.

Regardless of this activity, MHP offers aunit set of conformance arrangementsand a licensing regime which breaks themould. Designed as a truly opensystem, it is delivering on its promise.Now the market must decide when andhow to implement MHP with new andexciting applications.

DVB-T is often seen as the flagshipDVB specification. It has been around

for quite a while - since 1995 to beprecise. It still continues to be at theleading edge of technology thanks to thelatest developments in receivingtechnologies - evidence of a truly future-proof specification. All DVB-T eyes areon Berlin at the moment as they proceedtowards analogue switch-off at anunbelieveable pace. Since the launchthere in November 2002, 70,000receivers have been sold which is asignificant number for a smallgeographic region completely dominatedby cable television. In fact, the firstmajor switch-off date comes at the endof February, just as we go to press.Germany is key, because many of thesocial issues around switch-off will bedealt with some 10 years before themajority of us see analogue switchedoff.

This issue of DVB SCENE also reportson the developments in the UK whereFreeview has taken everyone bysurprise, and pleasant surprise at that.Freeview was launched to time with theChristmas season which is all importantin Western European countries. Theidea was fabulous, especially as the“converter boxes” (the low-end set-top-boxes available in the UK) were freelyavailable at the type of price which madethem appropriate Christmas presents.But the shops ran out of stocks of theseconverter boxes in mid-December.

So 2003 promises to be interesting forDVB. In these tough economic times,standardisation is more important thanever, as organisations can ill affordpoorly targeted development investmentas they prepare for better times to come.Will DVB be around in another 10years? Well nobody knows, but you canbe sure that DVB-T and MHP certainlywill.

DV

B-S

CE

NE

: 03

Bill Foote, JavaTV, iTV and Standardization Engineer –Sun Microsystems, L.A. & Chairman MHP Umbrella Group

In January 2002, the DVB technicalmodule created a new working group,called the MHP Umbrella Group, orMUG. Our mission? To build on theworldwide momentum of the MultimediaHome Platform (MHP) by enabling it inother markets. The end goal was noless than the adoption of the core ofMHP as a universal, global standard,even in NTSC markets such as the US.One year later, we are well on our waywith GEM, a DVB specification that is a“blueprint” for building an MHP-basedspecification that is tailored to therequirements of another market.

MHP is the DVB specification forbuilding interactivity into digital TV. Itspecifies all layers of the interactivecomponent of a set-top box or digital TVset. This includes a core Java virtualmachine, a number of interfaces usedby applications as well as details ofscreen resolution, signalling, serviceinformation and modulation schemes(DVB-T, DVB-S and DVB-C). The MHP

specification is backed by acomprehensive conformance test suitethat ensures that all MHP compatibleequipment works the same way. Thisenables a horizontal consumer marketfor receiving equipment. It also providesa common platform for enhanced andinteractive television programming.

In markets such as the US and Japan,it’s not practical to adopt all details of theMHP specification, because there is asubstantial investment in equipment thatis not compatible with MHP’srequirements. The most obviousexample may be video, where the USand Japan have a 30Hz frame rate at asomewhat lower vertical resolution than

and work in all markets.

• Adoption of MHP’s core acrossmarkets promotes economies of scaleand competition. This hold the promiseof driving the price of set-top boxes anddigital TV sets down further, faster.

• Enhanced commonality of broadcastinfrastructure similarly promoteseconomies of scale and competition,thus generating cost savings.

• The considerable number of man-years of effort invested in MHPconformance testing can be applied inother markets.

The MUG has made great progresstoward achieving its goal. A single,worldwide iTV standard that unifiescontent, receiver manufacture andconformance testing across Europe, theUS and the rest of the world nowappears to be within reach.

Shortly after its formation, the MUGbegan work on a specification calledGEM, which stands for GloballyExecutable MHP. GEM lays out thoseelements of MHP that must besupported in any iTV specification thatis based on MHP. It’s sort of acookbook for creating an iTVspecification: It describes thefoundation and required technicalessentials, such as the base virtualmachine and a comprehensive set ofapplication interfaces. It then includesrequirements for those elements to bedefined in a market-specific way, toadapt to local signalling or otherelements of the local environment.

“...MHP conformance testingcan be applied in other

markets.”

DVB’S GEMBringing MHP’s sparkle to the US and beyond

that required by MHP. The differencesgo beyond that, however: serviceinformation is sent differently, themodulation schemes are different,subtitles are handled differently, andother details differ.

Despite all of this, most of the MHPspecification applies perfectly to theseother markets. Indeed, the vast majorityof the MHP guarantees that impactnetwork-independent broadcastapplications can be preserved in otherenvironments. Keeping theseguarantees in all markets has a numberof important benefits:

• Applications can be authored once,

DV

B-S

CE

NE

: 04

At the November 2002 meeting of theDVB Steering Board, Carter Eltzrothwas named as DVB’s Legal Director.Although a new position within DVB,Carter comes well prepared. He hasacted as the chairman of the IntellectualProperty Rights Module since theinception of the DVB ten years ago.

Carter will continue working actively withthe IPR Module and its new chairman,Anthony Dixon of Pace Microsystems,and he will see through to completionprojects he helped launch over the pastyears.

“Over the next months I’ll help the MHPcompliance and licensing regime find itsoperational cruising speed,” Carter toldus. “I want to make sure that the regimeadministered by ETSI and the DVBProject Office is as efficient as possiblefor distribution of the MHP Test Suiteand the MHP Mark. All the while ofcourse the other key objectives arerespect for the IPRs of the licensors andinteroperability among MHPimplementations.

“Also our MHP ambitions extend now toproviding the core of the technology,

DVB’SLEGAL EAGLE

“A GEM application will run unaltered on an MHP digital TV in Finland, an OCAP set-top inNew York, or any other GEM-compliant device.”

It’s worth noting that GEM is not amechanism for subsetting the MHPspecification, to make somethingsmaller. Aspects of MHP that are notdirectly required by GEM usually requirethat a “functional equivalent” be defined.For example, GEM does not require the

In January of 2003, the GEMspecification was approved forpublishing by the EuropeanTelecommunications Standards Instituteas ETSI TS 102 819 V1.1.1. Work onthe GEM terminal specification fromCableLabs is nearing completion, and

through our GEM process, to sisterstandards forums in other broadcastingenvironments. At the Steering Board’sNovember meeting, we adopted anapproach to ensure reciprocity with ourIPR rules and coordination with theMHP conformance regime.”

Carter will also be following the progresson the initial steps in creating an IPRlicensing regime for MHP and otherrecent DVB specifications. Working withTheo Peek, Chairman DVB, and PeterMacAvock, Executive Director, he’ll betackling the full range of other legalquestions that a standards forum has toaddress. He’s an Oxford graduate (“adegree in literae humaniores,” he tellsus, “that’s Greek and Latin studies”) anda member of the New York bar. He’s adual Belgian/US national.

Carter was one of the initial signers ofthe DVB’s memorandum ofunderstanding in 1993, when he wasacting for pay TV operator FilmNet. Hehas sat on the Steering Board since,until becoming Legal Director. He’ll bedrawing on this experience, and hisyears in European broadcasting and

new media technologies, as he beginshis new work.

L-R: Carter Eltzroth and Anthony Dixon

exact color map of MHP, but it doesrequire that a functionally equivalentcolor map, adapted to the localrequirements, be defined. Generallyspeaking, GEM-compliant devices builtfor other markets will be about the samesize as their MHP equivalents.

A specification built on GEM is called a“GEM terminal specification,” and twoare under development. CableLabs inthe US is creating a specification calledOCAP (OpenCable ApplicationsPlatform), which is the basis of thefuture iTV plans for cable television in

North America. ARIB in Japan isstudying the creation of a Japanesestandard based on GEM for theirterrestrial, satellite and cable markets.

A GEM Application is an application thatis written only to the guarantees andinterfaces that are present in GEM. The

majority of broadcast content that isn’ttied to a particular network can bewritten to just the GEM guarantees.GEM includes support for MHP-stylegraphics, service selection, tuning, dataaccess, time-based and “do it now”stream events (or “triggers”), remotecontrol interaction, and access to anoptional interactive return channel forconnection to the Internet. A GEMapplication will run unaltered on an MHPdigital TV in Finland, an OCAP set-topin New York, or any other GEM-compliant device.

ARIB’s technical work is progressing.We look forward to publication, and tothe day when all interactive digital TVsets in the US and the rest of the worldhave MHP’s GEM at their core.

DV

B-S

CE

NE

: 05

DV

B-S

CE

NE

: 06

In May 2001, a commercial andtechnical strategy was defined for DVB -known as DVB 2.0 - which set out apath for the current phase of DVBswork. We are rapidly approaching the10th birthday of DVB - a decade sincethe first members signed the originalDVB MoU. How far have we progressedwith the DVB 2.0 strategy over the last 2years? How accurate have ourpredictions been?

“DVB’s vision is to build a contentenvironment that combines the stabilityand interoperability of the world ofbroadcast with the vigour, innovation andmultiplicity of services of the world of theInternet.”

The vision for DVB outlined aboveremains a valid statement of DVB’sintent. The business climate hasbecome considerably more hostile sinceit was formulated. However, DVB hasmade very substantial progress againstmany of the objectives we set ourselves.Sometimes, it has taken longer thanexpected to resolve issues and deliverfinal specifications, processes ordocuments. In other instances,unexpected opportunities have emergedthrough DVB that are new and exciting.Many of these have fitted well within theDVB 2.0 strategy and brought additionalstrength to the overall work programme.

Back in 2001 we identified a number ofcommercial drivers for the next phase ofDVB work.

1. Leveraging DVB’s ProvenProcesses

Technology has moved forward in anumber of areas since the initial DVBspecifications were put in place. TheDVB-S satellite modulation scheme is agood example. It is now possible to carrysignificantly higher data rates on asatellite transponder using the latestmodulation schemes. The DVB processis being used very effectively to define

DVB 2.0

the DVB-S2 specification. This is anexcellent example of the value in DVB’swell proven processes.

2. Convergence

In 2001 a prime convergence path wasthought to be DVB broadcast servicesbeing delivered over IP networks to PCs.It now appears that PC convergencewith TV broadcasting is happening lessrapidly. The emerging market forhandheld and portable devices isemerging as much more significant.DVB is working on a new specification,known as DVB-X to be delivered laterthis year; to support PDA devices andadvanced mobile phones by delivering aone-way always-on broadband datastream to complement their two-waynetwork connection based on GSM,GPRS or UMTS. The DVB-T terrestrialbroadcasting standard is expected to bethe basis for this solution.

3. Migration Routes

MHP is a sophisticated, feature richsolution for applications running on DVBreceivers. It is very pleasing to see MHPbeing adopted in a number of marketsaround the world, including in the UScable TV market through DVB’srelationship with CableLabs. There hasalso been a constructive approach todefining migration paths to MHP in theGerman market and in identifying someminor enhancements to the MHPspecification that would assist operatorswishing to migrate smoothly andeconomically from MHEG5 to MHP.

4. A Global Approach

The initial DVB specifications are clearlythe most widely used specifications fordigital TV worldwide - by a substantialmargin. However, they have not beenadopted in every major market. For MHPto achieve its full potential, we shouldstrive for it to be adopted as thepreferred open solution in all major

markets. This requires that MHPfunctionality should be available with allthe major digital TV transmissionformats. Provided that this approach isimplemented in a way that retains DVB’scontrol of the core MHP specification -now and in the future - and thatapplications are fully portable betweenmarkets then we will have truly achievedour goal.

In conclusion, DVB is continuing to drivethrough new, valuable and innovativespecifications that are well tuned to theneed of the industry, in Europe andglobally. Although there are manyunexpected external challenges that DVBhas had to address, it remains a vibrantand effective forum for the industry tocollaborate on pre-competitive activitiesin a commercially focussed way.

Graham Mills, Chairman DVBCommercial Module has made asubstantial contribution to the DVBindustry group since its formation,setting the internationalspecifications for digital TV andinteractive services.

.

Making Progress

“...this new standard wouldmean a more inexpensive

and richer contentexperience...”

DVB-XAcronyms like DVB-S, DVB-C or DVB-Tare familiar to most people working inbroadcast engineering. However, manymay not yet have heard of DVB-X. Whatis it?

The terrestrial broadcast transmissionsystem (DVB-T) is known to fulfill allknown requirements of a broadcastsystem targeting receivers in homes,offices, cars, etc. DVB-T is perfectlysuited for reception by mobile receiverseven at high driving-speeds, like forinstance 170 km/h. However, if batteryoperated devices with single antennareception at high speeds in large singlefrequency networks are needed thenDVB-X steps into the scene. DVB-X isto some extent unkown as hinted by theX. It is based on the current DVB-Tstandard and is purposed for the newservice scenarios where small,pocketable mobile terminals receivebroadcast services in various places onthe move. These terminals could besmall handheld devices that may includeintegrated mobile phones and providereception inside buildings, in cars,maybe even in trains.

The origins of DVB-X lie somewherelate in the year 2000 when a small groupof interested parties started to studywhether there is a need to changeanything in DVB-T concerning theemerging new service scenarios.During the spring of 2002 thecommercial module of DVB defined thecommercial requirements of the possiblynew standard for these servicescenarios. This later led to the start ofthe official technical work within theDVB technical module in summer 2002and the ad-hoc group DVB-X wasformed. The first outcome of this group

was the evaluation of whether DVB-Tcould do the job without anymodifications. The conclusion was thatDVB-T has excellent performance inmany aspects and the vast majority ofthe commercial requirements aredirectly fulfilled as such. The battery-operated small devices in some networkscenarios, however, could not be fullysupported. That is why a Call forTechnology was released in January2003 and the inputs will be evaluatedduring spring 2003. The schedule is toget the first draft of the future standardready in June 2003.

The new DVB-X standard, while by noway changing the current digital TVbusiness models for fixed reception,could provide new business possibilitiesfor a variety of players from broadcastand cellular operators to chip andequipment manufacturers. For thebroadcast operators it would provide anew vehicle to reach their customerswhile they are on the move, for cellularoperators it could provide a cost efficientway to implement data broadcast type ofservices and so on. For the end userthis new standard would mean a moreinexpensive and richer contentexperience in the mobile environment.

This would be very good news for thebroadcasters. This would allow them toserve their current customers in a totally

new way: service anywhere, anytime.This would even mean the possibility togain millions of new customers amongcurrent mobile phone users. This is alsovery good news for mobile phone users.Logically, TV service is the main servicethat is missing from today’s mobile

handsets. In addition to voice,messaging, internet browsing, radio, stilland video camera, TV wouldcomplement the service offering thatdigital convergence makes possible.According to field studies people arevery keen on being able to receive TV inthe device they already carry with themdaily.

DV

B-S

CE

NE

: 07

Dr. Jukka Henriksson chairs thead-hoc group DVB-X within TM.He holds 14 patents and is aResearch Fellow in NokiaResearch Center, Helsinki and aSenior Member of IEEE.

DV

B-S

CE

NE

: 08

The past twelve months has been agloomy period in the TMT sectorscreated by the collapse of dotcomdelusions, economic recession,geopolitical uncertainty and the harshwinds of commercial realism blowingover many digital television venturesaround the world. With severalspectacular DTT pay-tv collapses,mooted mergers of a number of rivaldigital satellite platforms around theworld, comprehensive smartcard swap-out programmes in many countries dueto endemic piracy, and advertisingsupported television taking a hammeringdigital broadcasting now must prove itseconomic worth.

We are seeing the plateauxing ofsubscribers to pay financed digital cableand satellite platforms around the worldsuggesting the pay-tv phase of digitaltelevision growth may now be peaking.The FT reported Germany’s cablenetworks might not be digitised for manyyears because new owners do notbelieve viable business plans can bedeveloped. Ambitious plans to deliverVOD via cable and telecommunicationsnetworks have been quietly sidelined bymany telecommunications playerspreferring instead to focus on upgradingdialup customers to highspeed Internetservices. And, almost de rigeur, HDTVlaunched as a technical success in anumber of countries has failed to break

About the author: Dermot Nolan is aDirector of Telecommunications andBroadcast services, a strategyconsultancy specialising in themedia and telecommunicationssector with clients worldwide.Contact details:dmenolan@compuserve.com,+44 (0)20 7286 5570

out from its narrowcast niche.

Interesting future trends are nowemerging in the UK, the world’s largestdigital market but where digital disastershave also hit hardest: ITVDigital foldedand the two cable MSOs have neededrefinancing to the tune of almost $30billion after racking up debts inoverambitious growth plans. The UKnow has 40% digital televisionpenetration, DVD penetration has shotto 25% penetration in three years, it isthe largest widescreen market in theworld, 80% have GSM phones andbroadband connections are nowexpected to reach 25% within four years.

The Freeview-led renaissance of UKDTT has attracted the greatest attentionleading to the record-breaking sale of300,000 DVB-T adapter boxes priced at£99 in the first two months of service.This spectacular turn around, enabledby the DVB-T system flexibility, hasconfounded the sceptics with manyleading agencies now predicting UKDTT will achieve mass-marketpenetration within five years. Freeviewhas around 2.0million TVHH and is thelargest DTT service globally. The otherdigital Cinderella, DAB, has finally comegood in the UK with availability of £99portable digital radios which sold outbefore Xmas, with predicted salesexceeding 500k units in 2003.

Free-to-air digital TV is also emerging in

countries like Germany where a portableand mobile DVB-T service launched tocritical and commercial success inOctober 2002.

Global international DTT standards warscontinue but DVB-T’s rivals, eitherimplemented or planned, exist largely aslaboratory demonstrators/paperconcepts or are plagued by severetechnical problems, minimal consumertakeup, and abject failure to export thesestandards across national borders. Inuncertain economic times broadcasters,consumers, Governments andregulators worldwide are more likely toendorse a tried and tested DVB-Tsystem leveraging global economies ofscale and scope and minimisesbusiness risk throughout thebroadcasting value chain.

In My Opinion - Dermot Nolan

GLOBAL DIGITAL BROADCASTING IN 2003:FROM HYPE TO HARSH REALITIES?

DV

B-S

CE

NE

: 09

Over 300,000 digital terrestrial adapters have been sold in the UK in the first two months following the launch of freedigital terrestrial television service Freeview. The BBC’s news correspondent John Simpson, pictured above, togetherwith a number of British celebrities featured in an televison commercial promoting the new service.

Over the same period, almost two million requests were received for information on Freeview, with approximately700,000 calls being made to the telephone information line and 1,200,000 unique visits recorded to the website.

Andy Duncan, BBC Director of Marketing & Communications said: “Of course it’s still early days for the launch ofFreeview, but these figures show that there’s undoubtedly a major opportunity out there for the idea of simple, freedigital television. “We’re delighted that Freeview is making the BBC’s digital channels available to many more peopleand bringing in a new audience who’ve not considered digital television before.”

CANAL+TECHNOLOGIES haslaunched its new MEDIAHIGHWAYDevelopment Kit (MHDK), a standard-based development environment forprogrammers wishing to createcompelling interactive televisionapplications for digital TV platformsworldwide.

MHDK is an effective environment fordeveloping applications to be deployedon Java-based digital TV platforms,including those compatible with DVB-MHP 1.0.2 and all MEDIAHIGHWAYmiddleware profiles.

MHDK enables a developer to create,edit, test and debug an application on aPC, simulating the complete interactiveTV technical chain. Thus, developerscan easily check how an application willbehave from its broadcast at the head-

CREATIVE ENVIRONMENTS

end to the final display on the TVscreen.

By reducing equipment costs andaccelerating development time, MHDK

leads to faster time-to-marketimplementation for innovative iTVapplications.

UK’S DIGITAL TERRESTIALBACK ON TRACK

:

New audio and video compressionalgorithms are now emerging thatchallenge the dominance of MPEG-2.These new algorithms offer animprovement in compression efficiencyof more than two; in other words it willbe possible to obtain comparable videoand audio quality with less than half thebit-rate needed for MPEG-2 video andMPEG-2 Layer II audio. Theimprovement is partially due to a betterunderstanding of compressiontechniques and partially due to theprogressive effect of Moore’s Lawallowing a level of complexity to beincluded today that could not have beencontemplated in consumer-pricedequipment 8 years ago.

DVB does not develop compressionalgorithms itself, but it does evaluate theperformance of candidate algorithmsagainst commercial requirements. DVBthen specifies parameters andconstraints to give a good trade-offbetween performance andimplementation cost for DVB services.In 1995, DVB published the documentthat became TR 101 154, theimplementation guidelines for the use ofMPEG-2 systems, video and audio.Work is now well underway on a similarexercise to include the advanced audioand video algorithms in the DVB suite ofstandards. This will allow the newH.264/AVC video and MPEG-4 HighEfficiency AAC Profile audio to be usedfor DVB services delivered directly overIP as well for more traditional broadcastservices where the audio and video areembedded within an MPEG-2 TransportStream.

In addition to being more efficient than

any previous video compressionstandard, H.264/AVC also has thedistinction of having more names thanany previous compression standard!The work began in ITU-T under theworking name H.26L. At the same time,ISO/IEC was considering AdvancedVideo Coding (AVC) within the MPEG-4standard. The ITU-T and MPEG expertsthen agreed to form a Joint Video Team,referred to as JVT. Within ITU-T it willbe published as H.264, whilst ISO/IEC

was first published in 1997 and itoffered about twice the coding efficiencyof Layer II. However, initial take-up waslimited, as at that time the market wasnot really ready for a new audiocompression scheme. MPEG-4 AAC isclosely based on MPEG-2 AAC butincludes some further enhancementssuch as perceptual noise substitution togive better performance at low bit-rates.The new MPEG-4 High Efficiency AACProfile adds spectral bandwidthreplication, to allow more efficientrepresentation of high-frequencyinformation by using the lower harmonicas a reference.

Of course, the increased sophisticationof the video and audio compressionalgorithms does not come withoutrequiring increased complexity of theencoder and decoder. In terms of thenumber of computer cycles required fora software implementation, the newalgorithms have been estimated to bearound 2 to 3 times more complex thanMPEG-2.

So what are the likely early applicationsof the new algorithms? They areunlikely to replace MPEG-2 in existing

Ken McCann, ZetaCast - Chairman ofTM AHG on Audio-Visual Content(DVB-AVC)

“...new algorithms offer animprovement in

compression efficiency...”

“...also has the distinctionof having more names thanany previous compression

standard!”

will publish it as Part 10 of the MPEG-4specification, 14496-10.

H.264/AVC is more of an evolution fromMPEG-2 video than a radical change.As with all ITU-T and ISO/IECalgorithms since H.261 and MPEG-1,the architecture is based on a motion-compensated block transform. LikeMPEG-1 and MPEG-2, H.264/AVC hasintra-coded pictures, predicatively codedpictures and bi-directionally codedpictures (known as I-, P- and B-frames).However, H.264/AVC has smaller,dynamically selected block sizes to allowthe encoder to represent both large andsmall moving objects more efficiently. Italso provides multiple reference framesto allow the encoder to find the bestmatch over several frames and itsupports greater precision in therepresentation of motion vectors. Thevariable-length coding used to compressthe picture and motion information iscontext-adaptive to give greaterefficiency.

MPEG-2 Advanced Audio Coding (AAC)

services for many years to come, but willprobably be used first for totally newservices. One possible example isHDTV, where the potential bit-ratesaving may significantly alter the viabilityof the business model. Another possibleexample would be low-resolution DVBservices delivered over IP to mobilereceivers.

ADVANCEDAUDIO AND VIDEOCODING WITHINDVB

DV

B-S

CE

NE

: 10

DVB-S2

In 1994, DVB introduced the DVB-Sstandard (EN 300 421) and today thespecification is used by satelliteoperators worldwide for televisionbroadcasting services and datatransmissions. It is based on QPSKmodulation and convolutional / Reed-Solomon FEC (Forward ErrorCorrection). In 1997, DVB-DSNGstandard (EN 301210) was developedfor satellite newsgathering and TVcontribution services and specified theuse of 8-PSK and 16-QAM modulationand a “pragmatic” reuse of DVB-S FEC.

Recently developed coding schemesbased on iterative decoding, takentogether with higher order modulation,promise more powerful alternatives tothe coding and modulation schemes ofthe current DVB standard.

In Spring 2002 the DVB CommercialModule sub-group on BroadbandSatellite Services approved thecommercial requirements for asuccessor to the current DVB-Sstandard that may be introduced for newservices and allow for long-termmigration. The primary objective of thenew specification is to enable delivery ofa significantly higher data rate in a giventransponder bandwidth than the currentDVB standard.

The new specification will cover threesatellite application areas: video andaudio broadcasting; interactive services(direct path), such as Internet access,for the consumer market, andprofessional links (SNG, contribution,Internet Trunking). Furthermore, forbroadcasting applications there areplans for two modes to be available:DVB-S2a, a non backwards-compatiblemode, and DVB-S2b, a backwards-compatible mode, allowing “old” DVB-Sreceivers to decode at least part of thetransmitted bouquet.

For unicasting applications, the DVB-S2system will include Adaptive Coding andModulation techniques (ACM), in orderto differentiate the FEC error protectionaccording to the service requirementsand the propagation conditionexperienced by each specific customer.

On the basis of the Commercial

Alberto Morello is director of theResearch and Technical InnovationCentre of RAI-RadiotelevisioneItaliana.

He was Chairman of the DVB ad-hocgroups which defined the technicalspecifications for the DVB-S andDVB-DSNG systems, and is nowresponsible for DVB-S2.

NEW GENERATION SYSTEM FORBROADBAND SATELLITE SERVICES

LDPC (Low Density Parity CheckCodes), with or without a concatenatedexternal block code (BCH or Reed-Solomon).

The first results are very promising. Forbroadcasting applications a capacity

increase of 30-35% versus DVB-S looksfeasible, assuming the sametransponder characteristics and link-budget. The achieved performance ofthe best systems are only 1 dB from theShannon capacity of the modulation,therefore it is expected that the DVB-S2physical layer will be the state of the artfor a very long period. Furthermore, forunicasting applications, the DVB-S2FEC system combined with ACMtechnique promise a capacity increaseof the satellite system between 100%and 200%, depending on the climaticcharacteristics of the coverage area: thismay open the doors to cost effective fastInternet services by satellite, in directcompetition with ADSL and cableterrestrial solutions.

Requirements, the Technical Module ofDVB set-up the DVB-S2 ad-hoc groupin May 2002, with the task to define thetechnical specification. The milestonesof the work plan are the following: thespecification for the non-backwardcompatible broadcast profile should beready in April, and for Interactiveservices (including ACM) in June, whilethe backward-compatible broadcastprofile should be defined before theAutumn 2003.

At present, seven technical proposalsare under investigation by computersimulation. They are based on fourfamilies of codes with recursivedecoding: TPC (Turbo Product Codes),SCCC (Serially ConcatenatedConvolutional Codes), PCCC (ParallelConcatenated Convolutional Codes),

“...objective of the newspecification is to enabledelivery of a significantly

higher data rate...”

“...seven technical proposalsare under investigation...”

DV

B-S

CE

NE

: 11

DV

B-S

CE

NE

: 12

PLAYING

SkyLife, the Korean satellitebroadcaster, will begin its

commercial launch in March and isusing the Alticast authoring tool,

AltiComposer, to deliver MHPinteractive services to its

subscribers.

SkyLife’s interactive servicesinclude: TV-Shopping, T-mail, SMS,Polling, News, Games, Quiz, Events

and Weather.

GAMES

There has been a great deal ofconsternation about the lack of MHPreceivers causing delays in the roll outof MHP. Now that manufacturers haveovercome this hurdle, why are we notseeing millions of receivers in stores?What is the problem? Where are theMHP markets? Well as far as DVBmember ADB is aware, they arewherever signals are being broadcaston-air (not as a field trial), which to dateis Finland, Germany (partially onsatellite) and Korea (satellite). This doesnot however, constitute a mass market,which is a drawback for those withproduction lines ready to rumble intoaction.

At ADB we are constantly asked, “Whenwill you have your fantastic MHP boxesin the shops?” Well the answer is simple- if there are no MHP consumer services

on-air - there is no set top box marketavailable. The STB horizontal market isonly the ‘Channel Marketing’ model ofFactory - Distributor - Retailer -Consumer. There is nothing the STB/iDTV manufacturer can do if thedistributors will not buy product that theycannot move on. To have any fullcommitment from MHP distributors, CEvendors, software houses, applicationsdevelopers and the likes require MHPon-air commitment. In the meantime,MHP companies are in limbo, waiting,hoping and encouraging thebroadcasters and operators to get onwith it and switch it on as fast aspossible before their enthusiasm, andfor some their cash reserve, runs out.

Finland will have an MHP horizontalmarket because it has gone about it thehard way and has been a pioneer andmodel of how to digitize and how tolaunch MHP services. If you have anMHP box you get the added services, ifyou want ‘plain vanilla digital services’you can have them too - off the samenetworks with different types of boxes

VERTICALLY CHALLENGED - DIAGONALLYANSWERED - HORIZONTALLY ACHIEVED

Anthony Smith-Chaigneau - Director of Marketing & Communication, ADB.

available to the consumer. MHPreceivers were virtually on the market atthe same time as Zapper boxes and theconsumer was allowed to make aneducated choice. Some made it onprice, some on technology, however, TVhad to advertise itself on TV - PublicService announcements were a keyfactor and reduced advertising spots forDistributors and Retailers also helped.Some cable and all terrestrialbroadcasts have MHP.

We have to encourage countries,Broadcasters/Operators/Networks tomove to digital as soon as they possiblycan. They should include MHPEnhanced, and Interactive Services atthe same time and allow the customer tochoose the type of STB that suits theirneeds. If the services are compellingenough consumers will pay good moneyto access them. We will then and onlythen have a fabulous Zapper/MHPReceiver/iDTV Horizontal Market.

DV

B-S

CE

NE

: 13

At DigiTAG (the Digital Terrestrial TVAction Group) we have often been askedby broadcasters or network operatorsnew to Digital Terrestrial TV and who arethinking of starting tests or field trials toadvise them on which of the 30+ modesin the DVB-T standard they shouldchoose. Of course our answer ‘It alldepends what sort of service you wantto offer....’ doesn’t satisfy them, but thenwe point out that their question is a bitlike ‘how long is a piece of string?’There are plenty of variables which needto be considered on the service featuresyou want to offer, and a particularlycritical question is how much frequencyspectrum may be available.

DigiTAG brings together people from allaspects of the broadcasting industry -broadcasters, operators, regulators andmanufacturers - so when we set up ourSpectrum Matters group in 2001 wewere able to call upon experts frommany disciplines to tackle the issue in athorough way. An important part of thestudy was a survey of servicerequirements, which sampled thethinking of 34 targeted experts from 21countries. The questionnaire probedtopics such as Type of service(traditional TV, video-on-demand,Internet access, Datacasting, shopping/banking services, IP-based servicesetc.), Coverage (the geographical areaor population over which the service is tobe available including regionality),Capacity, Quality and Reliability,Reception mode (fixed, portable ormobile), Type of Reception Equipment,Interactivity, Transmitter networktopology (using an existing TV network,a dense network of low-power sites, or acellular network for interactiveapplications), the DVB-T mode,Frequency Planning model (multi-frequency or single-frequency networks)and Convergent technologies- the use ofmultiple technologies (for example,DVB-T and UMTS) to provide theservice. This produced very interestingresults with some positive signs ofconsensus but plenty of variety in thedetail.

But we also wanted to look to the all-

digital future, for some countries at leastten years from now. To allow people toescape from the constraints imposed bythe real world, and the present day, weheld a brainstorming workshop wherewe invented three mythical countries.These countries were broadlyrepresentative of three major distinctionsbetween typical situations found today -the countries where terrestrial deliverydominates, where cable deliverydominates or where there is a highlycompetitive mix of cable, satellite andterrestrial delivery. This resulted ininteresting conclusions showing howinfluential the present market conditionsare to the future service requirementsand the development of a DTT offer.The typical demand found was for 100-150 Mbit/s total capacity, equivalent tobetween four and fifteen multiplexesdepending on the system variant to beused.

The spectrum planners of the ITU,CEPT and EBU have been lookingcarefully at the DigiTAG Report as partof their preparations for the marathonseries of meetings which will lead to anew Frequency Plan for Europe in 2005

or 2006. The new Plan will replace thenow 42 year old Plan fixed at a RegionalRadio Conference in 1961 and known bythe cognoscenti as ‘Stockholm ‘61’.Preliminary conclusions from their initialstudy to convert reception requirementsand the coverage targets into precisefrequency spectrum requirements seemto suggest that they have plenty of workto complete between now and 2006 ifthe are to squeeze DTT into theavailable spectrum and simultaneouslyreach the target population coverageand provide the required number ofmultiplexes!

The full report can be downloaded fromwww.digitag.org.

Ed Wilson has worked in the NewTechnology Division of the EBU inGeneva since 1989, and is presentlyalso Project Manager for DigiTAGand Secretary of the DVBCommercial Module.

SERVICEREQUIREMENTSFOR DVB-T: THE

FINAL REPORT

COPY PRCOPY PRCOPY PRCOPY PRCOPY PROOOOOTECTIONTECTIONTECTIONTECTIONTECTION WORK IN THE DVB

Digital distribution of video and audio(content) into homes and the advent ofaffordable consumer digital recordingand processing equipment and softwareis increasing both the ease of copyingand re-distribution, and the quality of theresulting copies and re-distributedcontent. This is causing concernamongst rights owners and theirlicensees about unauthorised copyingand re-distribution and the resulting lossof revenue.

The DVB is working towards producinga specification for an open,interoperable, Copy Protection andContent Management (CPCM) systemfor use in digital broadcasting,consumer products and in-homenetworks to provide end-to-endprotection for content in all processesfrom the point of initial distribution to theend-user’s point of consumption(viewing and listening).

The DVB CPCM system specification isintended to provide a means to preventunauthorised copying and re-distributionof content whilst also enabling thewidest range of new business models,including new methods for the legitimatedistribution of content, brought about bythe introduction of new compellingconsumer products, such as thepersonal digital recorder (PDR).

DVB Commercial Approach

In September 1999 the DVB establisheda new sub-group, DVB-CP (CopyProtection) of its Commercial Modulewith a mandate to prepare commercialrequirements for a CPCM system to

provide a common framework for theprotection and management of contentbeyond the traditional boundary points ofDVB compliant Conditional Access (CA)systems and for the protection ofcontent delivered by free-to-viewbroadcasters. In particular, the scope ofthe required system includes in-homedigital networks and PDRs. In commonwith all DVB standardisation work, thegoal of DVB-CPCM is an open,interoperable, flexible, industryconsensus and market-led specification.

The Stakeholders

The first step in the requirementscapture process undertaken by theDVB-CP group was to identify the keystakeholders and constituencies ofinterest amongst the DVB members.These comprised:

• rights owners - includingrepresentatives of the major moviestudios;

• consumer electronics manufacturers;

• broadcasters and platform operators(including both pay and free-to-viewservices);

• technology providers - including chipmanufacturers and CA securitytechnology providers.

Inevitably, viewpoints were initiallysomewhat polarised between these fourindustry groups. For example, there wasclearly a need to balance the concernsof the rights owners to protect theirrevenues with the concerns of theconsumer electronics industry to protectthe investment made by their customersin purchasing equipment. However,despite these differences, requirementscaptured from the perspective of eachconstituency showed a great deal ofcommonality

The DVB CPCM System

Scope

Although the functionality targeted for

DVB-CPCM is much less ambitiousthan that of a full digital rightsmanagement (DRM) system, the scopeenvisaged remains end-to-endprotection for content in all processesfrom the point of initial distribution to theend-user through to the point ofconsumption. It is also intended thatDVB-CPCM shall be applicable to thewidest range of equipment and not justrestricted to DVB-specified systems:this encompasses in-home digitalnetworks and personal digital recordersas well as all digital delivery systemssuch as satellite, terrestrial, cable, andthe Internet. If widely adopted thesystem could be extended to protectcontent distributed on pre-recordedmedia.

It is recognised that other DRM andcopy protection systems already exist inthe marketplace and will continue to beused and developed. Hence to theextent possible, without compromisingits integrity and security, DVB-CPCMmust co-exist with and interoperate withother DRM and copy protectionsystems.

Functional Model

It was found to be essential to identify afunctional model describing how thesystem would operate - indicating theneed for the DVB CPCM system to beself contained and implementable in anyequipment to provide “baseline” CPCMfunctionality but with extensibility, eitherby future DVB specification or byplanned co-existence with proprietarysystems.

Recognising that future developments inconsumer markets will include homenetworking and remote access it wasdecided to design the CPCM system tobe “network agnostic” This is topreserve the system functionality overtime as it is expected that networktechnologies will be various and

By Chris Hibbert, Chairman, DVB-Copy Protection Technologies Group

DV

B-S

CE

NE

: 14

©evolving; i.e. hard wired or radio; basedon existing and developing standardsand protocols such as 1394, Ethernetand IP.The Authorised DomainThe DVB-CP recognised that toconform with traditional user experienceand expectations based on the portabilityof pre-recorded content (DVD, CD),whilst maintaining protection againstunauthorised redistribution, it isnecessary to define an “AuthorisedDomain” for the use of content in ahome networking environment. Thisrequires a means to identify andsecurely assign devices and networksinside and outside the home uniquely toeach other and to a user so that onlycontent legitimately acquired by the usercan be consumed within his/herAuthorised Domain.

It follows that a means is also necessaryto securely bind content to theAuthorised Domain if required by therights owner, or to allow content to bemoved “outside” the Authorised Domainif allowed by the rights owner. In caseswhere content is allowed to be movedoutside the Authorised Domain existingcopy control rules may still apply.

End-to-end protection

It is expected that DVB-CPCM will bedeployed in the context of a homenetwork as well as in standalone devicesand it is recognised that some of thedevices in such a network may not befully compliant with DVB-CPCM.However, to preserve the security of thesystem, trust should only be given todevices which are fully DVB-CPCMcompliant and can signal theircompliance. Furthermore, whereverpossible, trust should only be given tothe necessary nodes in the system - thecontent source and the final rendering

device. Other intermediate devices, orentities such as “dumb” storage, shouldact as a transparent pass-through ofencrypted content and USI. Thisapproach increases security by avoidingthe necessity of multiple decrypt and re-encrypt processing as would be thecase if each entity, device or linkagecomprising a home network was to beconsidered a standalone element.

The Work in DVB-CPT

A call for proposals (cfp) for CPCMtechnologies was issued in the summerof 2001. Twenty four wide rangingresponses were received, some offeringa full solution and some offering specifictechnological approaches for elementsof the envisioned system.

In analysing the CfP responses itbecame clear that further work wasneeded to develop fully the functionalmodel of the envisaged DVB-CPCMsystem to provide a coherent frameworkfor analysis.

Rather than attempt to select a solutionat the time, as the responses were wideranging, the CPT group elected todevelop Technical Requirements forDVB CPCM to further identify requiredfunctionality and against which toassess the available technologies.

A crucial stage has been to fullytechnically define the characteristics ofthe Authorised Domain. This is key workas this novel concept is the fulcrum ofthe DCB-CPCM concept and the CPTgroup has spent many monthsdeliberating a number of challengingissues which again has exposedsignificant differences between theconstituencies of interest. At the time ofwriting this difficult stage is nearingcompletion of the first pass with agrowing awareness between thecontributing parties that only aconsensus based solution will work forthe industry as a whole.

The end game?

It is likely to be some time before anagreed specification for a full DRMsystem emerges, either through formalstandardisation processes or throughthe emergence of a dominant proprietarysystem. In the meantime there is anurgent need for a basic, openinteroperable Content Protection andContent Management (CPCM) System.

Ultimately the success or not of the CPTgroup will be apparent when theresulting specification is assessedagainst the original commercialrequirements. This process will alsoinevitably re-introduce some of thecontentious, unresolved issues whichcame out of the initial commercialrequirements capture process.

The time scales are therefore difficult topredict but the long term objective is forDVB to be in a position to lodge itsCPCM specification with a Europeanstandards agency, such as ETSI, sothat the industry can benefit from theachievement of the consensus drivenopen standards approach for which theDVB is known to be successful.

The full length version can be found atwww.dvb.org.

Chris Hibbert is Vice President MediaTechnologies & Standards with WaltDisney Television International.

DV

B-S

CE

NE

: 15