Why the energy cost or benefit of switching to DAB digital radio is hard to assess

One of the things that the current, terrible situation in Ukraine has taught us -- if we didn't already know -- is that we use a lot of energy. It often feels as if manufacturers and promoters of electronic devices think they're living in a world with an unlimited energy supply. In reality, UK domestic fuel consumption has decreased over the last few years, but there's still a long way to go.

The UK shut down its analogue television stations some time ago and, frankly, they aren't missed. The prospective shut-down of analogue radio, however, has raised concerns, one of which is whether digital (DAB) radio will use more energy overall. This wasn't really a concern with television -- most of the energy used by a television goes into the display; it doesn't make a huge different whether it uses analogue or digital broadcast technology. The same is not true for radio receivers -- not, at least, for the conventional "tabletop" receivers that are found in almost every household in the UK.

It's astounding -- at least, it astounds me -- that BBC radio services account for fully 1% of the total UK energy consumption. It's still a small amount, when compared to heating the spaces we live and work in, but it still seems an awful lot of energy. I guess it just shows how much power it takes to fill the air with radio waves.

With this in mind, it's not surprising that there is concern about the power consumption of DAB digital radio. DAB is a popular technology -- a tabletop DAB radio is as easy to use as an analogue one but, in most regions, offers a larger range of stations, and is largely immune to electrical interference and atmospheric effects. The BBC says that about 60% of all radio listening is done with digital devices of some sort.

There's no doubt that a tabletop DAB receiver uses more energy than an analogue FM receiver with similar features. The same is true of portable devices. My pocket radio has a battery life of about 50 hours when playing FM, but only six hours for DAB. I do not know if the same sort of findings affect tabletop, mains-powered radios. I think they do, but probably not to the same extent. A widely-quoted figure is 8 Watts (W) for a DAB radio, 2W for an analogue radio. It's worth bearing in mind that these figures apply only to small radios -- if you're listening to radio broadcasts on a full-sized hi-fi system, at ear-splitting volumes, the overall power consumption is not going to depend significantly on the broadcast technology: all the energy will go into making noise.

I suspect that younger people might be surprised at just how prevalent tabletop radios are. Many of my parents' generation -- those who are still with us -- have a radio on all day. I listen to the Radio 4 news for half an hour in the morning, and that's about all. My kids, in their twenties, don't know what a radio is. Be that as it may, according to Radio Today, 50 million people in the UK listen to radio broadcasts, averaging about 20 hours per week. I'd guess that much of that radio listening is on equipment that will show very different energy consumption between analogue and digital technology.

Based on these figures, it's not hard to see why the switch-off of analogue radio, originally planned for 2015, has been postponed until at least 2032. If a digital radio really does consume 8W for every 2W of an analogue radio, and radio already uses 1% of the UK's entire energy supply, that would be catastrophic.

 

The situation is, of course, not that simple.

According to research carried out by the BBC in 2020, the transmission of broadcast radio accounts for nearly 80% of the total energy budget. The energy demands of receivers are not really significant. So when we talk about the energy usage of radio, we really have to be clear about whether we're talking about the personal or the national impact of any planned change. There's no doubt that, at the personal level, DAB radios are more expensive to buy and run that analogue ones. At the national level, though, we need to be more concerned about the energy required for transmission.

So does it use more, or less, energy to transmit DAB? This is where things start to get messy.

An FM transmitter consumes essentially the same amount of energy whatever it is transmitting. The carrier wave is always present, even if what's being carried is silent. If the radio program is human speech, probably 10%-50% of the time what's being transmitted is silence. It goes without saying that running a transmitter at full power to broadcast nothing is wasteful.

A DAB transmitter behaves in much the same way -- the power consumption will still be high if there is nothing being broadcast but silence. However, DAB stations are multiplexed, that is, many different stations will use the same tramsmitter. The multiplexing means that, although a DAB transmitter is just as wasteful as an FM transmitter when it's broadcasting silence, it's never broadcasting silence.

Of course, broadcasting silence is an extreme example. A music broadcast will contain very little silence. So does the use of multiplexing reduce the power demand with music as well?

This is where things get really messy.

The use of multiplexing means that the available transmission bandwidth (capacity) of the transmitter is shared between different radio sources. Part of the multiplexing process involves data compression -- using mathematical tricks to reduce the bandwidth required by a particular signal. DAB radio compression is lossy, in the same sense that, for example, MP3 audio files are (and the soundtracks of DVD movies). The compression process loses some detail, which can never be regained. The cleverness of the compression scheme -- and it is undeniably clever -- is that it attempts to lose only information that cannot actually be perceived.

Of course, the more lossy the compression, the more likely it is that the signal will be degraded to an audible extent.

Now, the amount of loss in the compression depends on the number of stations that are sharing the same transmitter, and the content of each station's broadcast. In the UK, a few (very few) stations are protected from having their signals over-compressed. BBC Radio 3 is one example. All other stations have seen the lossiness of their compression increaed over the years, to accomodate more stations in the same amount of transmitter bandwidth. It's no longer true, if it ever was, that DAB radio offers CD-quality sound. Many stations don't even achieve FM-quality.

The effect of multiplexing, combined with lossy data compression, means that DAB radio makes much more effective use of the transmitter power than FM radio does. All this means that, at current levels of technology, the total, national energy consumption associated with radio will probably reduce when FM radio is switched off.

The reason that assertions like this are so hotly contested, particularly by radio enthusiasts, is that comparison is not like-with-like. The improved energy consumption is achieved at the expense of greater levels of multiplexing, with a consequent loss of audio quality. It could be argued that we should be comparing energy consumption of DAB radio when used with the same number of stations, at the same quality, as FM radio. However, DAB technology simply doesn't work that way -- this isn't a comparison that can actually be made.

 

So where does that leave us? With the best figures we have today, and assuming the same operating conditions, it would appear that having everybody move from FM to DAB radio would save a modest amount of energy -- at the national level. At least, that's what the BBC tells us. Each individual would still experience greater energy costs but, as noted above, the receivers only play a small part, when it comes to energy efficiency.

But this gain in energy efficiency comes at the cost of a reduced sound quality. The scale of that reduction is extremely difficult to assess, depending as it does on how stations are routed to multiplexes, and what kind of content is being broadcast.

Another way to reduce the energy requirements of digital radio would be to allow fewer stations, or reduce power (which would leave some households out of range), or broadcast only part of the time. Nobody is proposing any of these approaches as ways to reduce the energy needs of radio. Reducing quality, however, is seem by many as an acceptable cost, if the benefit is improved energy efficiency.

All of this means that assessing the costs and benefits, in energy terms, of switching to digital radio is not merely a technological problem -- it's also an aesthetic and social one. This is why it's so difficult.