| French radio groups want to adopt DAB+ & DVB-H27th September 2006 All of the main French radio broadcasters have formed a new Digital Radio Group, which has issued a press release saying that they plan to launch digital radio in France in 2007, and that they plan to employ the new DAB+ system, along with transmitting digital radio on the DVB-H system, and on DVB-T for in-home reception. Although the press release doesn't mention DAB+ by name, it does say that the capacity of the old DAB system (the one that the UK uses) is insufficient to allow all existing analogue stations to transmit on digital radio, and that it will be necessary to use the upgraded version of the DAB system, which is precisely what DAB+ is. The following radio groups are participating in the newly-formed Digital Radio Group:
Together, 90% of all radio listeners listen to stations owned by these groups. One significant addition to the radio groups that spoke out against using the old DAB system in a previous public consultation about digital radio is the NRJ Group, which is one of the largest French radio broadcasters. The press release also mentions that they plan to transmit radio stations at high audio quality; that data services will be transmitted; and that many existing analogue radio stations will be able to extend their coverage, for example to regional or national level. It should be said, however, that it has not been officially confirmed that France will use DAB+ and DVB-H for digital radio, and the French communications regulator, the CSA, is set to launch a public consultation on the subject of digital radio in the near future. However, considering that all of the main French radio groups have cooperated to form this new Digital Radio Group, and they've said that they want to use DAB+, I doubt the CSA will turn round and tell them that they can't use it!
The old DAB system is now effectively dead Australia has already committed to using DAB+, and New Zealand is apparently going to adopt it as well; but with France now almost certainly adopting it this markedly changes the outlook, especially concerning European countries, because France is such an important market. Also, the fact that the French broadcasters want to launch the new service as soon as 2007 implies that the new DAB+ standard will be released some time next year (the standard must be released before receivers can be put on sale, and they need receivers in the shops prior to marketing the new system), so it is hard to imagine any European country adopting the old DAB system when they could wait literally a matter of months for the next standard to be released — and, importantly, receivers will be available for it. Furthermore, the German commercial radio industry also recently issued a press release saying that the old DAB system should be upgraded or switched off, and they favoured using DMB and DVB-H instead. So combining the weight of the French and German broadcasters' views, this surely spells the end of the old DAB system as far as new countries adopting it goes. That would mean that the only countries using the old DAB system will be the UK and Denmark, and the head of engineering from DR, the Danish public service broadcaster, even said a couple of weeks ago that people will have to buy new DAB radios in 5 years' time, so the old DAB system might actually be switched off globally much sooner than many would have envisaged just a matter of months ago. It will be interesting to see what the Republic of Ireland and Norway do now. Ireland said that they wanted to launch DAB (i.e. the old DAB system) later this year or some time next year, and the Norwegian public service broadcaster NRK has suggested that it will stick with the old DAB system, despite sales having been low in Norway. The claims coming from NRK seem to be more to do with the professional pride of a handful of senior people at NRK, and they just can't bear to admit that they were wrong to adopt a dead-end system so late in the day.
DAB+ As mentioned above, the press release says that due to capacity reasons it will be necessary for France to use the upgraded version of DAB, i.e. DAB+. The reason for this is simply that the old DAB system (the one that the UK is using) is very inefficient due to it using technology that was designed in the 1980s, such as the MP2 audio codec, and very few radio stations (only 4 or 5 stations) can be transmitted at high audio quality in a multiplex using the old DAB system. Taking this into consideration, as well as the fact that there are a large number of FM radio stations transmitting in France (more than in the UK), and the amount of spectrum allocated to France at the Regional Radio Conference in the summer, it will be necessary for France to use the DAB+ system, or else the same problems would be faced as in the UK, i.e. low audio quality and less than 50% of all existing FM stations being able to transmit on DAB. The following link is to a more in-depth article about why it is necessary for countries to move to DAB+, and in particular adopt a system that uses the AAC+ audio codec. The new DAB+ system will use the AAC+ codec, which is the most efficient audio codec in existence, and together with more modern error correction coding, the DAB+ system is far more efficient than the old DAB system, and this translates into radio stations being able to transmit at high audio quality instead of low audio quality; more radio stations being transmitted; and far more robust reception quality — DAB+ will not suffer from the awful 'bubbling mud' effect we get on DAB in the UK on multiplexes where the signal is not very strong. Reception quality with DAB+ will be very similar in nature to the reception quality you get on Freeview (reception on DAB+ will actually be far more robust than reception on digital terrestrial TV though, because video is extremely intolerant of data errors, whereas AAC+ audio is far more forgiving), where it is very much a case of either having good reception, with very rare interruptions (such as pixellation on TV channels) caused by data errors, or you don't get any reception at all. The reason reception quality will be similar to that on Freeview is simply because the error correction coding that will be used on DAB+ is virtually identical to that used on Freeview — an inner layer of convolutional coding is combined with an outer layer of Reed-Solomon coding, with the latter being used to catch errors that the inner layer missed. DAB doesn't have an outer layer of error correction coding, so data errors that the error correction decoder (Viterbi decoder) cannot correct are passed through, and this is the main reason why we hear 'bubbling mud' when the multiplex signal is weak.
DVB-H The press release also mentions that radio stations should be carried on DVB-H and on DVB-T for home listening. Following this comment, the press release goes on to request that manufcturers "ensure the interworking at the level of the terminals". It is a bit too vague to know precisely what is being proposed here, but it could mean that they want receivers to support the DAB+, DMB and DVB-H systems on the same receiver. Significantly, numerous radio stations are still being carried on the Paris DVB-H trial, and in the previous French public consultation on digital radio a number of radio groups were in favour of transmitting radio on DVB-H, so it could well be that the proposal is for receivers to support all these standards.
What are the consequences for the UK? One thing several people have asked me is: will the change to DAB+ make all DAB receivers in the UK obsolete? The simple answer is, yes, but only once we've switched over to it, which won't happen for a number of years. Ofcom is in favour of changing over to using DAB+ once the vast majority of DAB receivers in the UK can receive both old DAB and new DAB+ formats (in that article, DAB version 2 refers to the same thing as DAB+, i.e. the upgraded — and backwardly compatible — version of the DAB system). The likely scenario will be that within a couple of years the vast majority of DAB receivers sold in the UK will be capable of receiving DAB+ (due to economies of scale, it makes more sense to make one chipset or one design of receiver that can be sold in any country rather than having separate DAB and DAB+ versions sold in different regions), and after a few more years the vast majority of all DAB receivers will be able to receive DAB+, so the UK will be able to make the switch to the new standard. This is not as extreme a course of action as you might at first think, because if you think about analogue radios you've bought in the past, many people discard them or replace them every few years anyway, and when DAB sales are high (in, say, 2-3 years), the receivers being sold will be able to receive DAB+ anyway, so the time when the "vast majority" of all DAB receivers can receive DAB+ will probably come sooner than many think. Furthermore, DAB portable radios that are expensive today should be cheap in a few years' time, so it won't cost much to replace older receivers. Another incentive that will make people replace current DAB receivers is the added functionality of DAB products in the future, such as having the EPG, the pause and rewind feature, and the ability to record to SD card — the vast majority of receivers don't have these features at the moment, but some of them at least may become commonplace in future, and they will increase the rate at which the older, more basic DAB receivers will be discarded. As an example of how quickly the number of older receivers diminishes, Ofcom estimated that the number of ITV Digital set-top boxes reduced by 70% (from over 1 million to under 300,000) in the 3 years following the launch of Freeview in October 2002. So, when will the UK be able to switch to DAB+? An aggressive estimate would be in 5 years' time, and a worst-case scenario estimate would be 10 years' time — I'll go for 7 years from today.
The 'DAB debate' is presumably now over — thankfully! I mentioned above that the professional pride of a few people at the Norwegian public service broadcaster NRK was stopping them from doing what's actually in the best interests of the country — i.e. to ditch the old DAB system and adopt DAB+ instead. Unfortunately, this situation has happened in numerous European countries, and there has been a very bad-tempered debate about which digital radio system should be adopted over the past 2 - 3 years, with one group in favour of using the old DAB system, and the other side wanting to adopt the more modern systems instead. With very few exceptions, it has been people from the respective national public service broadcaster that have been arguing in favour of sticking with the outdated DAB system but, thankfully, in every decision taken in the last couple of years, their views have been overruled, and the old DAB system now looks decidedly Dead And Buried. Another characteristic of the the proponents of the old DAB system is, I'm sad to say, how dishonest they have been in the debate. With the emergence of new systems, such as DVB-H and DMB, and the forthcoming DRM+ system, the old DAB system was well and truly shown up to be exactly what it is: an extremely inefficient and outdated system that was designed in the 1980s and should never have been commercially launched in any country. The efficiency of a digital radio system is the ultimate measure of how good or bad a digital radio system is — everything important measure of how good or bad a digital radio system is is directly related to the efficiency of the system, such as the audio quality provided; the number of radio stations that can be carried; the cost of transmitting the radio stations (and thus the likelihood that broadcasters will actually provide sufficient bit rate levels necessary to provide good audio quality). And as I've shown on this website, DVB-H is around 6-times more efficient as the old DAB system, and DAB+/DMB is around 4-times as efficient as DAB in terms of how many radio stations can be carried at a certain level of audio quality and in a certain amount of bandwidth. Even more staggering was the comparison between the old DAB system and DVB-H when the transmitter powers were made equal (this will mean that the transmitter network costs of the two systems will be approximately equal), because then DVB-H could carry 13-times more radio stations than DAB, which would equate to the radio stations being 13-times cheaper to transmit, which would obviously help avoid the issue of the broadcasters reducing the bit rate levels to save money on transmission costs. And although the DRM+ system is still under development, its efficiency figures are even more impressive than DVB-H's!... And yet despite this evidence that clearly demonstrated just how awful the old DAB system is in comparison with modern systems, the proponents of the old DAB system — those handful of people from each of various European public service broadcasters — still tried to claim that the old DAB system was the best system to use! This is what I've found most disgraceful about the whole debate, because these people must have known that they were backing a system that was far worse than the modern systems they didn't have a leg to stand on in terms of the technology used, and that it would invariably lead to a very poor service in comparison to what would have been possible if one of the modern systems were adopted instead, but none of that put them off, and they persisted in disingenuously trying to suggest that the old DAB system was the right system to use. For example, every time I saw a document written by a DAB supporter that compared DAB with DVB-H, the number of incorrect assumptions made was astonishing — examples included using mobile phone antennas for DVB-H and 75cm antennas for DAB (approx 10 dB power difference in DAB's favour); using indoor reception for DVB-H and outdoor mobile reception for DAB (approx 8 dB power difference in DAB's favour); assuming that the bit rate required on DVB-H and DAB was the same (AAC+ is approx 3-times as efficient as MP2); using DVB-H with half-empty multiplexes to make it look more expensive to run; using carrier-to-noise (C/N) ratio figures for video reception for DVB-H and audio reception for DAB (approx 4-5 dB power difference in DAB's favour); and so on and so forth, all of which were methods that were used to make DVB-H look far worse than if comparisons were made on a fair, i.e. like-for-like, basis. But needless to say, they were always careful to avoid addressing the issue of efficiency, because that was something that no matter how hard they tried, they couldn't make DAB look any better than the incredibly inefficient system that it really is — multiplex capacity figures are there in black and white, written in the specifications themselves, and there are numerous results of listening tests available that shows that AAC is twice as efficient as MP2 and AAC+ is 3-times as efficient as MP2 — MP2 is the old DAB system's major Achilles' heel, and yet they even tried to spin it that MP2 was better for broadcast applications, because they claimed it was more robust, which is a total fallacy, although I genuinely think they simply didn't understand why they were wrong on this issue (the simple technical answer for why they were all wrong is that DAB uses UEP (Unequal Error Protection) error correction coding rather than EEP (Equal Error Protection), and if UEP is not used then MP2 is no more robust to bit errors than any other audio codec). They couldn't twist the figures for efficiency, so they simply avoided the issue if at all possibe, but when it was brought to their attention, they reverted to their usual disingenuity. So why did these European public service broadcasters want to use the old DAB system? I think they stuck with it for two main reasons: Firstly, the old DAB system was primarily designed by the European public service broadcasters themselves, so they basically didn't like to see their baby criticised, and they wanted to persevere with it even though they knew it was the wrong system to use — or they had simply collectively deluded themselves into thinking that it really was the best system to use. Also, most of the "DAB supporters" are senior technical people in their 50s or approaching retirement (this is actually a possible reason why they displayed such a poor understanding of the digital technologies of which DAB is comprised, because they will have spent most of their working life working on analogue systems, and I'm afraid digital communications systems are very different to analogue communication systems, and being an expert in analogue comms definitely does not mean you're automatically an expert on digital comms), so people in such positions will likely be the same people that recommended to senior management that DAB was the right system to use, so they simply didn't want to be shown to be wrong. It would also seem that the EBU was pivotal between all the public service broadcasters thoughout the debate, because the EBU represents the European public service broadcasters collectively. Now, it may again be just a coincidence, but the Director of the Technical Department at the EBU, Philip Laven, was the Controller of Engineering Policy at the BBC between 1993 and 1997, and it just so happens that the time between 1993 and 1997 was the time when the BBC should have noticed that DAB couldn't deliver all that was required of it and that it should have been upgraded so that it could — the BBC had launched its national DAB multiplex in 1995, the multiplex was already almost full, the BBC wasn't going to get another national DAB multiplex, and yet it had always intended from 1992 onwards that new digital stations would be launched. (If you're wondering, the technology needed to upgrade the system did exist in the mid-1990s: Reed-Solomon coding has been around since the 1980s (it's used for the error correction on CDs), and development of the AAC codec began in 1993.) The BBC did eventually launch five new digital stations in 2002, which led to the bit rates plummeting, which is why the audio quality of the BBC stations on DAB is so poor. That is, the current situation with very poor audio quality of the BBC stations on DAB was entirely predictable from the early to mid 1990s onwards, and that was precisely when Philip Laven was in charge of Engineering Policy. Another case of professional pride getting in the way of what action should have been taken, perhaps? Another reason why I suspect that the EBU had a large part to play in trying to stick with the old DAB system was the number of articles it published in its quarterly Technical Review publication that were heavily biased towards DAB — in one EBU Technical Review article comparing DAB with DVB-H I found 14 gross errors, and all of these errors favoured DAB and/or made DVB-H look bad (I fondly call this article: "The most inaccurate engineering article I've ever read in my entire life"). In another EBU Technical Review article, again comparing DAB with DVB-H, I found two incorrect assumptions that invalidated the results of the entire article, and by another amazing coincidence, the article (written by a proponent of the old DAB system from Swedish Radio) concluded that it was better to use DAB than DVB-H, yet when both the assumptions were corrected, DVB-H was by far the better system to use. And it's probably just another amazing coincidence, but Philip Laven is on the Steering Board of WorldDAB — the organisation in charge of promoting DAB around the world! The other reason why, in my opinion, the European public service broadcasters favoured using the old DAB system is simply protectionism. Moving from analogue to digital radio moves from a position where the public service broadcasters were dominant — or at least as far as single broadcasters go they usually have a disproportionately high share of the audience — to a position where the commercial radio industry becomes greatly dominant in terms of the number of radio stations transmitting. Therefore, the public service broadcasters stand to lose out from the move to digital radio as listeners migrate from their stations to the new commercial stations that launch, and the more efficient the digital radio system used the more the public service broadcasters will lose out, because there will be a larger number, and thus a wider choice, of new commercial stations for listeners to migrate to. So the logical thing for the public service broadcasters to try and wrangle is to use a system that is as inefficient as possible, which is why I think they favoured using the old DAB system, despite the fact that they knew that supporting it was not in the best interests of radio listeners. Thankfully, it seems that truth has prevailed over dishonesty, and the old DAB system is effectively dead in the rest of the world and dying in the UK and Denmark. Good riddance to bad rubbish.
| |||||||||||||||
Digg
del.icio.us
Reddit
Facebook
Google
Stumbleupon
Slashdot