Wireless Systems Design

Shortwave Radio Jumps On The Digital Bandwagon

Like many of you, I attribute my career in electronics and wireless design to shortwave listening (SWL) and amateur radio. My father gave me an old S-38 Hallicrafters receiver when I was in junior high and I learned to listen to world wide shortwave (SW) broadcast stations. That soon lead to ham radio and so on.

Most of you probably don't think much about shortwave radio because we have so many other radio services today at our fingertips. We have had the usual AM and FM for some time, but now we also have digital versions of these stations in the form of HD Radio. And, of course, there are the newer Sirius and XM satellite digital radio services and Internet radio (which is also digital).

But while the average person doesn’t give it too much thought, shortwave radio is a big deal globally. In fact, in many countries shortwave radio is the ONLY radio available. There is also a large SWL hobbyist following in this country and throughout the world. I guess we all should have expected it, but now SW radio is going digital. They call it Digital Radio Mondiale (DRM).

DRM 101

DRM is a digital wireless technology designed to work in the well-established Long Wave (30-300 kHz), Medium Wave (300 kHz-3MHz ), and Short Wave (3-30 MHz) broadcast bands. Most digital modulation methods really eat up spectrum, but at these low frequencies there is not much room. As a result, digital techniques usually involve digitizing voice and music, compressing the digital signal, and then using some kind of spectrally efficient modulation method to transmit the result—just like in most cell phones. The result is that you can transmit voice and music in the common 9 kHz and 10 kHz bandwidths available in the broadcast bands. The digital method is designed to be compatible with the AM broadcast transmitters already in use throughout the world.

So why do this? I guess the short answer is because we can. But it’s about more than that. Most other radio services are going or have already gone digital, so we must also. But even better is that there are some benefits to going digital. Better signal quality is one. That means FM radio-like frequency response (up to 15 kHz) in the AM bands. It also means better reception quality. Propagation characteristics at these frequencies vary widely from the predominantly ground wave transmission of LW and MW transmission to the extreme multiple hop sky wave propagation of SW. SW reception, if you have ever experienced it, suffers from lots of static, fading, interference, and other maladies, as the signals are refracted from the ionosphere back to earth multiple times. Digital techniques can help mitigate all of those problems, reducing noise and fading to a minimum.

If you have already listened to satellite or digital broadcast radio (HD Radio in the U.S.), you may have already experienced these benefits. There are nearly a thousand AM and FM stations now transmitting their traditional content and other new content in the HD Radio digital format right along with the regular analog AM and FM signals using a digital method developed by iBiquity. Not too many people know of this or use it but it does work pretty well. I have a Boston Acoustic HD Radio, and I listen to the FM version here in Austin, TX. None of the AM stations here have adopted it, but it does a good job on FM.

The DRM digital standard will not be used in the U.S., especially in the AM and FM broadcast bands, because the FCC has blessed iBiquity as THE digital standard for the US. However, the FCC may consider DRM for digital broadcasts in the SW bands for worldwide broadcasting. As of now, it does not exist, but it is at least being discussed. In the rest of the world, especially Europe, DRM is already under way with some stations already transmitting and receivers finally becoming available.

How It Works

DRM is an international standard established and maintained by the European Telecommunications Standards Institute (ETSI) and the International Telecommunications Union (ITU). It uses the audio compression techniques of MPEG-4, mainly the high efficiency-advanced audio coding (HE-AAC) codec that is good for both voice and music. If voice only is preferred, the MPEG-4 CELP or HVXC codecs can be used to narrow the bandwidth or reduce transmission errors. Compressed bit rates are in the 8kbit/s to 20 kbit/s range for a standard 10 kHz broadcast bandwidth. Versions are also available for bandwidths of 9 kHz, 18 kHz, or 20 kHz.

As for modulation, DRM uses coded orthogonal frequency division multiplexing (COFDM). No surprise here, as what new digital wireless standard has not adopted OFDM? It is used in Wi-Fi, WiMAX, HD Radio, European DAB, DVB, and ISDB-T, and is expected to be the method of choice for 4G cell phones. Although not wireless, DSL uses a form of OFDM called discrete multitone (DMT). It is a modulation and access technique that is hard to beat for its spectral efficiency and multipath robustness.

Remember that OFDM takes the serial compressed digital signal and divides it up into many parallel slower bit streams then modulates them onto multiple adjacent, but orthogonal, carriers within the broadcast spectrum. Orthogonal, of course, is just a complicated way of saying that the very closely spaced carriers (called bins in some cases) do not interfere with one another. Modulation is by quadrature amplitude modulation (QAM). QAM-4, QAM-16 or QAM-64 may be used. And, as you might expect, DRM uses a forward error correction (FEC) scheme called multi-level coded modulation.

The standard actually defines four different profiles, depending on the level of robustness desired in the applications. Profile A has little or no multipath or Doppler problems and uses 228 carriers with a carrier spacing of 41.66 Hz. Profile B assumes some multipath propagations in the MW bands. It uses 206 carriers with 46.88-Hz spacing. Profile C assumes more Doppler problems associated with longer distance transmission. It uses 138 Hz carriers spaced 68.18 Hz apart. Profile D is for the really long distance SW transmissions with Doppler and delay spread problems. It uses 88 carriers spaced 107.14 Hz. Those figures are for a 10 kHz bandwidth, but other values are used for 9 kHz channels.

In any case, the good news is that all this is generated with an inverse fast Fourier transform (IFFT) in a DSP chip. Then any old standard, high power AM broadcast transmitter can be used to transmit it. The bad news is that you need a really special DSP receiver implementing the FFT to recover the signal. Not too many of those around.

How to Design a DRM Receiver

A DRM receiver is going to be some kind of software-defined radio (SDR). It will have the usual front end with LNA, mixer, synthesized local oscillator, and one or more IF stages. From there, the signal is digitized and sent to a processor running the OFDM software. All the demodulation and final filtering and signal recovery is DSP.

The easiest way to make a DRM radio is to take the IF output of a good communications receiver and send it to the sound card of a PC. Then get some of the open source DRM software available. A good number of engineers exploring this technology have successfully modified the AOR AR7030 general coverage receiver for DRM. AOR is the successful communications receiver manufacturer in Japan. The TenTec RX320 and RX340 radios made by TenTec, the amateur radio manufacturer in Tennessee have also been successfully converted to DRM reception. In most cases, the DRM software is the open source code developed by the Fraunhofer Institute in Germany. Another software package is DREAM software created by University Darmstadt also in Germany. Diorama is another German output written in MATLAB. And there are a few others.

As for commercial DRM receivers, there are a few. Current manufacturers include Fraunhofer, Kenwood, Morphy Richards, Sangean, and Starwaves. Others are no doubt on the way, as broadcast services become available and as today’s high receiver prices drop.

Radioscape, a UK manufacturer of radio modules, has a combined DAB/DRM module ready for receiver manufacturers. That will ease the design process which is not for the faint hearted because of the OFDM challenges.

The Fate of DRM

It is still too early to determine whether DRM will be successful. It depends, of course, on what "successful" means to the various players. I think it will be a winner worldwide once more broadcasters adopt it and a reasonably-priced receiver is available. I am not sure it will be much of a factor in the U.S. market, though (except for the SWL niche which is small, but reliable).

I am anxious to try DRM out. I still listen to SW once and a while; it’s easy to get lost for a couple of hours tuning in the huge number of foreign stations and listening to them. It is easy with an analog SW radio. I cannot imagine how easy a DRM signal will be to tune. Because the CODFM requires pilot carriers, channel estimation, and synchronization procedures, it may take a few seconds of dwell time on a station to get it to come in. It might be that instead of continuous tuning as is common in SWL, listeners may have to tune to specific known DRM stations with a synthesized receiver. That's OK too. We shall see.

For more about DRM, go to the official Web site at www.drm.org.

Hide comments


  • Allowed HTML tags: <em> <strong> <blockquote> <br> <p>

Plain text

  • No HTML tags allowed.
  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.