For additional news items on Digital TV - CLICK HERE - this will access news and information that has appeared in the News section of the auspaytv site

Digital Video Broadcasting Project
A URL for those with an interest in DVB (Digital Video Broadcasting). The project is an industry-led consortium of over 300 broadcasters, manufacturers, network operators, software developers, regulatory bodies and others in over 35 countries committed to designing global standards for the delivery of digital television and data services.
DVB systems Blue book reference


ABA INFO, etc    

ABA digital tv channel plans for Sydney, Newcastle, Wollongong, Brisbane, Toowoomba and Darwin

The Australian Broadcasting Authority has released its first digital channel plans. The plans, for the Sydney, Newcastle, Wollongong, Brisbane, Toowoomba and Darwin television markets, set out the channels existing broadcasters will use for their digital transmissions from 1 January 2001.

Full details available on the ABA release ... ... HERE   

LAST week the newspapers reported the International Telecommunications Union had "adopted and endorsed" a new global digital television standard.

What they've actually done is develop an electronic production standard which is better for the production of movie films for large-screen theatrical projection. This is important, because a lot of prime-time programs are still shot on expensive 35mm or 16mm film, and this meant the benefits of electronic techniques aren't being exploited to the full.

So the ITU has created a "common image format" (CIF) which is 1080 pixels in height, and 1920 pixels in width (a wide-screen ratio of 16:9). And, unlike the old interlace analogue system and some of the new digital transmission formats, these images will be scanned progressively from top to bottom, one line at a time, at a rate of 24 frames each second – the same rate used for motion pictures.

This produces images which are usable both for TV broadcasts and the cinema screen, and gives us a storage and production standard which is transferable to both without the need for complex conversion gear. Until now, HDTV cameras used television's 50Hz or 60Hz image scan-rates, and ignored the requirements of cinema's celluloid projection process.

Hollywood movies now include long sequences created (at least partly) in the computer, so this intermediate one-size-fits-all production format makes a lot of sense. It is being supported by most of the major networks around the world and many of the equipment makers.

But does this mean higher image quality in the future?

Personally, I doubt it, except perhaps in the cinema. However, most people seem to associate digital with quality, and they've done that since CD-Audio replaced the old LP vinyl disc record player in their lounge room.

Digital TV is not really being introduced for reasons of quality, despite the propaganda. For standard image sizes, we will probably just replace analogue ghosts and artefacts, with a whole range of new digital audio and video artefacts. MPEG decompression produces some weird and wonderful results on occasions.

However, digital does mean bandwidth reduction, and this is an important issue in Europe, Asia and America where channel capacity in the VHF and UHF bands is fairly tight.

Bandwidth isn't a problem in Australia, but . . . what the heck! If the Yanks and Poms are going to do it, we need to prove we can do it quicker and better by introducing digital HDTV . . . even if we don't have a need, or the programming.

The standard 7MHz of channel bandwidth we reserve for an analogue television channel, consists of 5MHz which is set aside for video, and the rest for audio and a bit of slack. But, in reality, any good moving image needs about 10MHz.

However, back in the 1960s they managed to jam the video stream into 5MHz of bandwidth in a very clever way. They scanned the image 625 times, but only sent every second line; then repeated this pattern, alternating between odd and even line-numbers, 25 times a second. The home TV set reconstructs the full 625-line image without much flicker.

That's what TV engineers mean by "interlace scanning", which is an analogue compression scheme.

The problem with interlace is that, when lines 1 and 2 are sent at different times, minor changes in electronic timing (in terms of millionths of a second) can cause random horizontal image shifts and lower the picture quality. Such "time-base" changes become noticeable when you play a tape from your VCR, which is why the replay quality is markedly inferior to the original.

When colour arrived, they embedded the "chroma" information into the same analogue video bandwidth by highly compressing it, and transmitting it at the beginning of each line – and that produced some of those rainbow and herringbone artefacts. This was a complex way of adding colour at that time, but it meant that viewers with old B & W television sets could watch the same channels as those with new colour sets – a remarkable achievement.

In the 1960s, the analogue transmission process designed for the VHF band (30 to 300MHz) was also translated successfully to the next higher band, the UHF (300MHz to 3GHz). Television now uses some of this bandwidth at the lower end for channels numbered between 12 and 69, and at the high end for 19 MDS channels. MDS channels need a set-top box with its own tuner.

Australia had so many free channels at this time that the spectrum managers didn't bother giving us the channels between 11 and 28, and for a few decades they maintained the myth that the 19 MDS channels couldn't be used for television.

With digital techniques there will be more than enough programming channels in the UHF band for the programming that will be available, and the only reason television networks can retain a hold on VHF is through the political strength of the media owners. They want to retain VHF because it provides the farthest reach and best coverage from a single TV tower.

But the spectrum managers would dearly love to recover all the VHF band and make it available for mobile communications. It is only marginally better than UHF for television, but very much better for mobile links.

If the media moguls can retain their VHF channels and add digital techniques, they can simultaneously pump out four standard-image programs with a better population spread, in exchange for the one old analogue channel. This then has potential both for a free-to-air channel and a couple of pay-TV channels – and maybe a bit of datacasting on the side if they can see a way to make profits.

What makes digital techniques of value to customers, is much less obvious. The 6.5 million households in Australia will eventually cough up $6 billion to $12 billion in direct costs for digital TV conversion by progressively buying new sets and video recorders. They will also indirectly pay for the lot, of course, including studio and transmitter conversions, and any increased production costs, because that's how the economy works.

The way to understand the development of digital television is to visualise a box with four quadrants. On one side we have video technologies, and on the other side audio. In the vertical plane we have compression and decompression techniques for both – and then radio transmission and detection/selection techniques.

The breakthrough came in 1993 when the MPEG (Motion Picture Experts Group) committee developed a standard way to compress video and audio using very-fast acting, large-scale, integrated computer chips. In effect, these were computer-processors-on-a-chip, but, by designing them specifically for one job and including the instructions in the wiring, they could be made extraordinarily fast and foolproof while still being cheap to manufacture.

Fortunately MPEG is "asymmetrical" in its requirements; it is easier to decompress than compress, which is why the system works effectively with low-cost home equipment. Television stations can afford to pay a million dollars to buy digital compression systems, but television distribution can't work unless the home-decompression equipment costs a few hundred dollars at the most. That was the breakthrough.

At the same time, and as part of the chip-set, MPEG also introduced three standard "levels" of audio compression. So we can now have multiple channels of sound for each image (up to eight), and these sub-channels can be of different qualities. In Europe they'll broadcast sports with different language commentary on different low-quality sub-channels, and combine this with high-quality background sound of the match.

The American company Dolby had a rival audio standard called AC-3. And in Australia, our chips will need to handle the decompression of both MPEG and AC-3 sound because the Europeans are using MPEG audio, while the Americans are using AC-3.

Incidentally, the highest MPEG audio quality has since become the MP3 standard that allows music to be sent over the Internet.

The second component of digital television is the transmission technique, and this has nothing to do with MPEG or AC-3. The transmitters and home receivers just transfer data of any kind reliably from one end of the radio link to the other.

The modulation standard we are using comes from Europe and is called Coded Orthogonal Frequency Division Multiplexing or COFDM. The best way to think of it is as a very wide road with a few thousand lanes for slow-moving traffic. The more lanes you have, the more traffic can be sent down in total, even if each lane only moves at pedestrian pace. The use of slow data-rates on 8000 FDM channels means that not much can happen to cause serious interference.

So with COFDM, a single old analogue channel can carry 20Mbps of digital data, which is enough for one wide-screen high-definition image with six-channel surround sound. Alternatively, it can be split into two wide-screen standard-definition channels, or four PAL-equivalent channels with images and sound of about the quality we have now.

The beauty of this approach is its extreme flexibility. Provided the home receiver is intelligent enough it can find and select the sub-channels it wants, and perhaps direct the other sub-channel to a VCR, or another to a PC for data-cast delivery.

So while the benefits of digital may not be readily apparent to most consumers in the near-term, there will be some value derived from the conversion in the long-run.