The DVB-T2 standard, which is being proposed to be used in the UK from 2009, is still going through the procedures to adopt it as a standard. This is a vital, as a UK-only standard would be unworkable.
The current terrestrialdigital standard is over a decade old, which in computing terms is somewhat ancient. The widespread use of this European standard has provided many countries with an upgrade path from the old analogue broadcasting, but not without some difficulties.
An upgrade to the satellite version of the protocol, DVB-S2, has already been used to provide the higher data rates required for HD (high definition) broadcasts. However, it can be argued that satellite broadcasting is a technically a somewhat simpler proposition as an externally mounted dish and LNB is the only way that anyone expects it to work.
DVB-T, on the other hand, has been used in a wide variety of non-ideal situations.
To provide the necessary frequencies for services to be provided, most systems (such as Freeview) have be provided using the transmission frequencies that were originally intended as 'unused' buffer frequencies between the analogue transmitters.
These channels had to be left, because without them there would be areas that would receive analogue signals on the same frequency from two or more transmitters, with terrible results. Even with the gaps left as they are, at times of high atmospheric pressure the Inversion Effect can make analogue reception poor.
Therefore, for systems such as Freeview to operate in these 'analogue' buffer frequencies it was necessary for the signal levels to be at very low power levels. As viewing a digital multiplex broadcast on an analogue set just looks like snow, these low power DVB-T transmissions cause indiscernible degradation to the old analogue signals.
However, the low power levels of the current DVB-T system means that the area covered by a digital signal from a given transmitter is much smaller than for the analogue, often requires a better aerial, cables, connectors and is more prone to interference such as impulse interference from motors in fridges, freezers, central heating systems, two-stroke engines (scooters) and systems such as mobile phones, TETRA masts, wireless internet and even baby-monitors.
The use of buffer frequencies is also the reason that many Freeview transmissions are 'out of group': the grouped aerials were used in the first place to stop interference from transmitters in adjacent areas.
To complicate matters even more, not only are the Freeview multiplexes often at different transmission power levels to each other, two of them use a different mode of operation. The BBC and multiplex C and D use 16QAM mode (with 4x4 sub-carriers), the ITV/C4 and multiplex A use 64QAM mode (with 16x16 sub-carriers) with more data and channels, but with
DVB-T in common with most computer communication protocols is actually a 'stack' of component technologies. Each part builds on the functions provided by the lower part. Basically these are:
Data transmission: COFDM (Coded Orthogonal Frequency Division Multiplexing) using 16QAM and 64QAM, 2k carrier, with Viterbi/RS forward error correction.
Multiplexing and service definition: MPEG-2 multiplexing
Video coding: MPEG-2 (which is built on JPEG) at MP@ML (main profile, main level, 720 pixels on 576 lines, 25 frames per second).
Audio coding: AAC and Layer III
DVB-T2 updates these by making improvements at each level.
COFDM is improved by using the 256QAM mode, with 16x16 sub-carriers (compared to 8x8 used by 64QM and 4x4 used for 16QAM), a more efficient error-correction system, LDPC/BCH and 32k carriers
MPEG-4 video encoding.
Higher resolutions (up to 1680 pixels on 1080 lines)
Better audio (such as Dolby Digital).
UK Free has been told that "interleave will improve impulse robustness by 10 times - 10 dB ( likely 6 dB more compared to the UK 2k mode?) Better use of pilots and fewer pilots and the option for CD3 channel estimation will increase the number of useful carriers. 32k mode will allow more useful carriers or PAPR reduction tones.
The DVB-T 64QAM CR2/3 transmissions are about 4 dB less robust than the DVB-T 16QAM CR3/4 signals. This will change with the typical 10 dB ERP increase at DSO, leaving about 6dB to increase coverage. DVB-T2 will have the same (of better) robustness with at least a 30% higher capacity. The example from the RAI paper indicates more like 45% higher in MFN's and 67% in large SFN's.
Ofcom may go for a smaller capacity increase to gain an increase in robustness (compared to 64QAM).
Multipath propagation and echo is an important problem. The use of the 8k and for DVB-T2 even the 32k FFT mode will increase the guard interval from 7 to at least 28 usec - a very long time in a MFN. The use of CD3 estimation and the larger FFT modes 16k, 32k will significantly improve the possibility to use very large (nationwide) SFN's. This is what the DTG thinks is the most important feature of DVB-T2. "
Will this new mode achieve the same predicted coverage as 64QAM DVB-T?
A BBC spokesperson says: "At the moment, Mux 1 and B are 16 QAM, the others being 64 QAM. We dropped the QAM rate as it gave us slightly better coverage. But at switchover, all the powers increase, and the aim is to make the digital transmitters cover the same areas that the analogue ones do. At that point, the BBC would also revert to 64QAM on the multiplexes it controls.
The Ofcom HD proposal, albeit one that is likely to be accepted is for MPEG 4 and DVB-T2 to be used to facilitate the use of HD on Mux B. So the situation is more involved than the QAM rate for DVB-T2, which in any event is not a fully agreed standard. However it is our understanding that the coverage of the proposed HD mux should be the same as the SD ones."
A spokesperson for Ofcom says they have "considered the technical implications of DVB-T2 (on multiplex B) and expects the coverage to be the same as the improved 64QAM coverage which had been planned for the multiplex after digital switch over. It is some time since DVB-T was developed and it can be improved upon significantly by introducing new techniques, some of which are enabled by the capabilities of modern silicon chips. The satellite standard, DVB-S, has already been improved to yield DVB-S2. Likewise DVB-T2 has been developed using newer technologies.
Ofcom expects that at switchover those multiplexes changing mode from 16 QAM to 64 QAM will increase in power to compensate for the fact that DVB-T 64 QAM transmissions are slightly less robust than those using DVB-T 16 QAM. A switch from 2k carrier mode to 8k mode is also expected to improve signal robustness. DVB-T2 is different from DVB-T in a number of significant ways that increase the robustness of the received signal. These include improved error detection and correction and an increase in time interleaving, which improves the resistance to interference. DVB-T2 transmissions are likely to use 256 QAM and up to 32,000 carriers, resulting in a low symbol rate for each carrier. This combination, together with the improvements already mentioned, results in a significantly higher bitrate with the same coverage and at the same power as a DVB-T 64 QAM transmission.
I hope the above helps set your mind at rest that we expect no loss of coverage when multiplex B is re-engineered to use DVB-T2."