The last time we spoke to Peter MacAvock the DVB Project executive director, he put us off the scent for DVB-T2 and said that it would be ready some time this autumn, but at a June meeting of the DVB Steering Board the DVB-T2 standard as ratified and published as a specification.
Sure that doesn’t make it a standard but it means that chip makers first and then equipment manufacturers everywhere will be able to begin creating the building blocks that will be needed for what might be deployments as early as 2009 instead of the mid-2010 that we had all been led to expect.
We wondered how, two weeks ago, the BBC was able to begin tests on a DVB-T2 service which largely had not been defined, but given the way the UK regulator Ofcom has already committed to using DVB-T2 in the near future for HD transmissions, it doesn’t surprise us that the two major UK transmission companies National Grid Wireless and Arqiva have already got the standard up and running for BBC tests. Ofcom wants to start the new HD Multiplex some time during 2009.
In the end the DVB-T2 draft standard was approved on June 26. In essence it promises 30% to 50% TV broadcast capacity improvement, and is expected to allow new transmitter multiplexes which can squeeze more existing SD channels in to existing Digital Terrestrial transmissions, freeing up space for at least one new HD multiplex. Similar initiatives are expected all over Europe.
DVB-T2 continues to use OFDM modulation the same as DVB-T and DVB-H (though that was never guaranteed and we understand that other schemes were looked at including the Chinese Time Domain Synchronous TDS-OFDM ). Better error correction means that modes such as 256 QAM can now be catered for as well as QPSK, 16QAM and 64QAM, and a huge variety of guard intervals can be specified and the number of OFDM carriers now rises from 8,000 to include both 16,000 and 32,000.
But the new system has opted for the same Low Density Parity Check error correcting codes used in DVB-S2 (satellite) combined with BCH (Bose-Chaudhuri-Hocquengham) rather than those used in DVB-T because it offers better performance in high noise. Completely new features include Physical Layer Pipes, Multiple-Input-Single-Output (MISO) and Rotated Constellations. The standard has also been defined in such a way that it can be enhanced in a backwards compatible manner through Future Extension Frames.
Rotated Constellations is a new techniques which provides additional robustness in difficult channels which also allows transmissions to be tailored such that a receiver can save power by decoding only a single
program rather than a whole multiplex of programs. This would be useful for mobile TV and is analogous to how DVB-H works.
One of the big breakthroughs is to use a diversity method, known as Alamouti or Space-time block coding, which in essence is a way of sending two signals and combining them on a single antenna. This breaks a data stream into blocks, which are distributed among spaced antennas and across time to provide multiple copies, which are then combined at the other end to produce a better image, under more difficult conditions. Whether this might create some cost issues for transmitter antennas or not we are unclear.
The specification was the result of more than 30,000 person hours from member bodies and will now be presented to ETSI for publishing as a standard.
It leaked out during May that the mobile TV element of DVB-T2 would not initially address mobile TV despite the fact that it was in the original spec. However now we can see how later a backwards compatible Future Extension Frame could address how this should be implemented, on top of the standard.