OK, I've read through enough of the technical materials to convince myself I follow the standards. I'm a RF/DSP/comm guy, so hopefully I can shed some light here.
Sure, EU147 could be adapted to just about any band. It's fundamentally just a OFDM transmission scheme, which is a very adaptable digital communication technique. You could, more or less, just change the carrier of EU147 and plop it just about anywhere.
But this would require some tweaking of the format, and the transmissions wouldn't be compatible with existing EU147 receivers in Europe. In the future, it would be possible to design "dual band" receivers, analagous to the (few) GSM phones that work both in Europe and the states.
This would also be possible with IBOC, but more difficult. An IBOC radio and a EU147 radio wouldn't have a lot in common. They could share a couple components, but basically building a raddio that could handle both would be about as hard as building two radios. This is the same reason that you can't get CDMA phones that can work on GSM networks in Europe.
So, what's the advantage of IBOC? It basically comes down to an issue of backwards compatibility. IBOC is designed to fit in the USA analog radio spectrum, and allows stations to simultaneously transmit digital and analog in (roughly!) the same slot of the spectrum. This allows stations to switch over to digital without changing their station numbers, which is a big plus; this will allow a gradual evolution to digital.
By contrast, even if EU147 were adapted to the US spectrum, it would put radio stations in a funny bind. They would basically have to buy a new slot for digital bradcast. "1500 on your AM dial, 1400 for AM digital, 102.7 on your FM dial, 100.3 for FM digital." How many stations are going to try and pitch a whole new station number for the digital band? How many car buyers are going to spend the extra for a digital radio if the stations don't broadcast it? It's a viscous cycle, and it's far worse in the states than in Europe, since there are so many more stations here. Using EU147, even adapted to the US band, will really slow down the already-slow adoption of digital radio.
Furthermore, if EU147 were adapted to the US spectrum band, it would no longer have significant transmission advantages over IBOC. In fact, it would be less robust, since IBOC can fall back to analog, where EU147 would have no such (practical) fallback ability.
So, sounds like I'm a big IBOC fan, right? Well, not really. IBOC's problem is pretty obvious; it's a bandwidth hog. It basically doubles up the width that each radio station needs. Allowing IBOC's use without changing the spacing of the radio stations strikes me as stupid. It's going to cause a lot of interference in the next station over. This should have been as obvious as sunrise to the FCC folks. I guess that, given that EU147 wasn't really an acceptable standard for the US market, they decided on the next best thing avaiable, which happened to be the only thing available. Too bad it has some big drawbacks.
So, if I don't like IBOC or EU147, then what do I like? Well, I can think of a couple somewhat reasonable options. For digital to work in the US market, it has to satisfy a couple requirements:
- it has to fit in the US radio spectrum
- it has to allow broadcasters to transmit in both analog and digital, while still only pitching one station frequency
- it has to be able to fall back to analog if the digital signal breaks up
- it has to work with existing transmission equipment
IBOC passes these standards, while EU147 does not. Can you satisfy these standards without hogging spectrum like IBOC? Well, not really. You WILL need double the spectrum to broadcast in analog and digital simultaneously. So there's no easy solution. Some ideas:
- Make IBOC sit on EVEN multiples in the FM dial. No other station can use the adjacent odds. This keeps IBOC from interfering, and it doesn't waste any bandwidth. The only problem is that the dumb computer that digitally tunes most existing FM radios (like the one in your car) will want to skip over the station. The reprogramming would be easy, but somewhat impractical.
- Make IBOC bradcasters pay for 1/2 of the adjacent slots. Two IBOC broadcasters could sit next to each other (say, at 100.1 and 100.5) and no station would be allowed to use the station between them (100.3 in this case). This would be expensive for the stations and impractical when two popular stations are next to one another. (Of course, that would suck even more the way IBOC is currently set up.) Also, this only makes sense if a bunch of IBOC broadcasters end up clustered together in one part of the spectrum. Otherwise you end up with lots of wasted half-slices of bandwidth.
- Some EU147-type standard, but the digital station (on a seperate band) carries a pointer to the analog station, and vice versa. This seems very appealing at first, but then you realize that the stations need tow transmitters to handle two frequencies, and the radio needs two tuners to handle two frequencies. This is really impractical.
- Do what they're doing, and just let the poorer, less popular stations flee to cleaner slots where no IBOC broadcaster is next to them. After digital becomes popular, they can just start phasing out every other radio slot (i.e. 100.1 and 100.5 are available, but you can't use 99.9, 100.3, or 100.7 any more). I imagine that there are some folks at the FCC that understood the interference issue, and this is, more or less, how they plan to deal with it. It is the cheapest solution.
Bottom line: The US radio market is very big, higly distributed, and entrenched in existing analog technology. A standard like IBOC, which allows stations to step over gradually, is basically the only road to large-scale digital radio here. Frankly, there isn't a really elegant way around IBOC's spectrum-hogging nature. The FCC may not have shown the wisdom of Solomon in adopting IBOC, but I don't see a drmatically better way.