Digital radio systems need to support over-the-air software upgrades

As the migration to DAB+ in the UK will leave millions of DAB receivers obsolete over the next few years, and the decision in France to adopt DMB rather than DAB+ means that we'll have two incompatible digital radio standards on either side of the English Channel, this shows that the current way of doing things simply isn't working. Furthermore, as soon as one standard has been ratified someone could have a brainwave the following day that could lead to a a new and more efficient way of doing things, but we'd have to wait for years until the next time that millions -- possibly tens of millions -- of receivers are made obsolete before we could take advantage of it.

The solution, therefore, which isn't a new one, because DVB-T (the standard that Freeview uses) already uses it to a limited extent, is to use over-the-air (OTA) software upgrades. OTA software upgrades would solve or at the very least hugely improve upon the way things are done at present, because DAB+ receivers purchased in the UK, for instance, could download the software to receive DMB broadcasts in France, and vice versa. And if OTA software upgrades had been supported over the last few years on DAB, then the UK could have avoided a lot of the pain we're going to have to go through in the migration to DAB+.

There are arguments against supporting OTA software upgrades, however, such as that receivers would need to contain over-specified processors and more memory than would be required if only current standards were supported. But processing digital radio signals doesn't come close to stressing modern-day DSP processors, and memory is dirt cheap, and no doubt by the time the OTA software upgrades could be implemented another iteration of Moore's Law would have passed to make the processing yet more trivial and the memory costs cheaper still, so those arguments are no longer valid.

The other argument against supporting OTA software upgrades is that receivers might not successfully complete the upgrade process, thus leaving them being as useful as a house brick. Those who put forward this argument also say that receivers are made by a large number of different companies, so there's a large number of different architectures that would all need to work for this to be successful. Addressing the latter point first: something like 95%+ of all DAB receivers sold so far in the UK have actually contained receiver modules that were designed by just two companies -- Frontier-Silicon and Radioscape -- and the receiver manufacturers themselves don't have anything to do with the DAB software that processes the digital radio signals. So in reality there's very few architectures that need to be supported, so it would be relatively simple to test that the OTA software upgrade works correctly. And regarding the issue of receivers failing to upgrade properly and not working at all as a result, there's obviously ways to stop this from happening at all, because for example if the hard drive on your computer fails, you don't have to throw away your PC -- all that's needed is to store some software in read-only memory that's able to get the radio booted up again so that it can receive and install the upgrade again. It just needs a small amount of extra memory, and memory is cheap and getting ever-cheaper due to Moore's Law.

Using Wi-Fi Internet radios this year is what's convinced me that OTA software upgrades are the way forward, because the software that runs Wi-Fi radios can be upgraded over the Internet, and it's a painless process to perform an upgrade -- you just select an option from a meny and the rest is done by the radio automatically. But more importantly, the benefits gained from being able to upgrade the software have been huge. For example, when I first tried a Wi-Fi radio in January 2007 the number of audio codecs the radio supported was pretty limited -- MP3, WMA, Real, WAV and maybe a couple of others -- but over the course of the year support for new audio codecs, including AAC, AAC+, M3U playlists and FLAC have all been added. Also, there was an issue with playing back some Internet radio streams that used a technology called Intellistream, where 128 kbps WMA streams were only played back at 32 kbps, and that issue has been solved in later versions of the software as well.

DAB+ & DMB will themselves be made obsolete

The new DAB+ standard has only been released this year, and as I write this, France announced only yesterday that it will use DMB for its digital radio system. But they're both really just botch upgrades of the DAB system, because they're using the old DAB system's transmission scheme as a base and they've added more modern technologies on top, such as the AAC+ audio codec and Reed-Solomon error correction coding. But the old DAB system's transmission scheme dates back to the 1980s, so some of the technologies they're using are already massively outdated, and some of the new mobile systems we're going to see over the next decade will make them look ridiculously ancient in comparison.

Just to demonstrate how ancient they will look, the Japanese mobile phone network company NTT DoCoMo demonstrated earlier this year a 4G mobile phone prototype system transmitting at 5 Gbps in a 100 MHz channel with the receiver travelling at 10 km/h. Mobile digital communication systems are compared on the basis of their spectral efficiency (which is calculated by dividing the bit rate by the channel bandwidth), and the spectral efficiency of this 4G prototype system is 50 bps/Hz. In comparison, the spectral efficiency of DAB+ and DMB is 0.7 bps/Hz (1.09 Mbps in a 1.54 MHz channel), so the 4G prototype system is 71 times more efficient than DAB+ and DMB!

Therefore, I would predict that DAB+ and DMB will themselves be made obsolete by these 4G mobile systems unless they're upgraded, because whichever 4G mobile phone systems is used it is bound to include a broadcast mode, so DAB+ and DMB will have to compete against those systems. And the best way to do upgrade is to phase-in receivers that support OTA software upgrades so that only the oldest and least-specified receivers would be made obsolete when newer standards are adopted -- OTA software upgrade basically extends the useful life of receivers, so migration to newer standards can be done in a "soft" manner rather than the black and white way it happens today.

Ultimately, though, the terrestrial digital broadcasting systems are all doomed in the long-term, because there will be no point in carrying on using them, but if OTA software upgrades aren't supported on these systems very soon then it will simply bring forward the date at which they'll become obsolete.

There is an issue with the payment of licensing costs relating to OTA software upgrades, though, because for example upgrading to a newer system requires licensing costs to be paid -- DAB+ uses the AAC+ audio codec which requires a licensing cost to be paid on every receiver that uses it. However, it would only require that the user pay for the upgrade over the Internet or over the phone so that they can enter a key into the radio to allow the software to be installed, so it's hardly a difficult issue to solve.

Bookmark with:
Digg	del.icio.us	Reddit	Facebook
Google	Stumbleupon	Slashdot