27.1 Introduction

With the advent of the CD (compact disc) and higher-quality distribution channels for music, the need for radio broadcasting of digital audio with CD or CD-like quality arose. One immediate problem for enabling such radio broadcasting services was limited radio bandwidth. The CD format uses PCM, which requires a bit rate of about 1.3 Mb/s for stereo. Thus, for successful digital audio broadcasting, efficient high-quality audio compression was required. Since people listen to radio predominantly in cars and not so much at home, another requirement was reliable delivery to moving receivers in cars and on high-speed trains.

Digital Audio Broadcasting (DAB) is more spectrally efficient than analog FM. In addition to the improved quality requirements, this spectral-efficiency advantage of digital over analog was one of the driving forces behind the development of DAB. The increased spectral efficiency of digital transmission (with respect to analog FM) effectively means that a larger number of radio programs can be broadcasted over a given bandwidth. Another key advantage was the flexibility in transmission offered by the digital technology. Thus, auxiliary services (data services) can be obtained in addition to high-quality audio. Examples of such services are traffic information, weather information, stock quotations, image and even video services, and so on.

The first system that was developed was the European Eureka-147 or DAB system. The term DAB is used as a general term for digital audio broadcasting and it is also used as a synonym for Eureka-147. It was an all digital system and required new spectrum. The European Broadcasting Union (EBU) played a leading role in the development of Eureka-147. The system was launched in many European countries, with mixed success, and was also adopted by some countries outside Europe. A key disadvantage with DAB is “old” audio coding (audio compression) technology, which leads to poor spectral efficiency and far from optimum audio quality. A new more efficient version, DAB+ has recently been developed. This system utilizes more efficient audio coding technology at lower bit rates. Unfortunately, it is not backward compatible with DAB.

In the United States, digital audio broadcasting developed along a different path. Since no new spectrum was allocated for this service, it was difficult if not impossible to adopt the all-digital Eureka-147 system. Instead, new “in-band” systems were developed for both the FM and AM bands. The broadcasters wanted an evolutionary path with initially hybrid systems with a digital service added to the existing analog FM and AM broadcasting services. In the future, the analog transmission would be abandoned, freeing bandwidth for additional digital services, including not only extra music and talk show channels, but also multichannel audio and data broadcasting services. The technical name for these systems was hybrid IBOC (in band on channel) and all digital IBOC, both for the FM and AM bands. There is now digital audio broadcasting services in all states in the US, referred to as HD Radio (high definition radio). Companies like Lucent Digital Radio, and USA Digital Radio, later merging into iBiquity Digital Corp., played a major role in the development of digital audio broadcasting in the United States. HD Radio is currently available throughout the US and, although the technology was solely developed for the US market, it is now being considered and tested in many other countries all over the world. IBOC yields increased spectral efficiency and the flexibility in terms of providing a backward compatible way to augment existing systems with data services.

In a separate development, the Digital Radio Mondiale (DRM) consortium, has developed systems for digital shortwave broadcasting as well as systems for the AM and FM band transmission of digital audio including hybrid schemes.

In parallel, there has also been a convergence with digital cellular. Systems like MBMS (Multimedia broadcasting multicast services) have been developed both for 3G CDMA and for OFDM LTE 4G as well as MediaFLO and DVB-H. Finally, other systems worth mentioning include DMB in Korea and China, and audio as part of digital video broadcasting systems like ISDB in Japan and DVB-H.

MBMS is a digital audio broadcasting system integrated into a cellular system. Thus, low-power transmitters are used with small cells. In contrast, all the other terrestrial systems mentioned above, including HD Radio and Eureka-147, as well as traditional analog FM and AM, use high-power transmitters and tall towers with large cells. The MBMS system uses only one radio receiver for cellular and broadcasting while the other systems use two separate receivers for cellular and broadcasting.

In a different development, digital satellite audio broadcasting systems were developed for the United States. CD Radio, later Sirius, and XM Satellite Radio were licensed and they developed similar but somewhat different technologies. Sirius uses geosynchronous satellites and fewer terrestrial gap filters while XM uses geostationary satellites and a large number of gap filters. Later the two companies merged into one company, Sirius XM. The business model is pay radio services with coverage throughout the 48 contiguous US states. The service is also available in countries south of the United States and in Canada.

What does the future hold? There is obviously the option of using (existing) high-quality audio coding schemes yielding superior quality to CD quality. However, it is not clear as to whether or not this is a viable option, since it requires an increase in source bit rates and thus its bandwidth requirements. Multichannel stereo (spatial audio, surround sound) could be of interest for broadcasting in the future. In the further convergence with cellular and wireless broadband, internet radio with a very large number of programs is gaining increasing support, especially with the increasing use of smart phones. This case represents an example where unicasting services are instead providing services traditionally provided by broadcasting.

In the remainder of this chapter we give a brief technical description of many of the systems mentioned in the introduction.

27.2 Eureka-147, DAB, and DAB+

Eureka-147 was an EU (European Union) research project that produced a digital audio broadcasting standard in the 1980s with a start in 1981. The first commercial broadcasts took place in 1995 in the United Kingdom. This system is often referred to as DAB, [F02]. ETSI produced the original specification for DAB, [ETSI_1]. DAB later evolved into DMB (Digital Multimedia Broadcasting), [wDAB] and DAB+, [ETSI_2]. DAB, DAB+, and DMB are currently in operation in many countries in Europe, Canada, China, Korea, and Australia, [wikiDAB]. DAB, DMB, and DAB+, are all-digital systems for audio broadcasting in new frequency bands. Here we focus on audio services, although, in the case of DMB and DAB+, there is also video broadcasting.

The DAB system uses a wideband transmission technology format in Band III (174–240 MHz) and the L Band (1452–1492 MHz) although in principle the scheme allows operation almost on any band above 30 MHz. DAB offers a number of modes, allowing different countries to use different modes and various frequency bands.

The original DAB format uses the MPEG-1 Audio Layer 2 audio coder. This is also referred to as MP2, [F02]. At 192 kb/s it provides less than adequate audio quality at this high bit rate due to the fact that this audio coder represents “old” technology today. In contrast, the new DAB+ system uses the HE-AAC version 2 Audio Codec (AAC+), [wikiHEAAC,wikiAAC]. This audio coder yields very good stereo audio quality at 64 kb/s.

The DAB system uses punctured convolutional codes with unequal error protection (UEP), [LC04,C91]. Parts of the bit stream are provided with higher levels of error protection than others, to account for the fact that audio quality is much more sensitive to errors on those bits, [ETSI_1]. Interleaving is used to spread out the effect of burst errors and thereby improve the performance of the decoder. In the new DAB+, an additional layer of Reed Solomon outer nonbinary block coding is also used for improved error protection, [ETSI_2].

The modulation system used in DAB is orthogonal frequency division multiplexing (OFDM) with differential quaternary phase shift keying (DQPSK), [ETSI_1]. The OFDM format is used today in many mobile applications (like cellular 4G advanced LTE), [D08]. DAB was actually pioneering the use of this technique. No equalization is used to compensate for multipath propagation effects. As an example, in Transmission Mode I, 1536 sub-carriers (tones) are transmitted in parallel. The overall OFDM symbol duration is 1.246 ms and the cyclic prefix is 246 µs.

With OFDM, DAB can use the concept of a Single Frequency Network (SFN), [ETSI_1]. This is equivalent to a frequency reuse factor of 1. (In contrast analog FM has a frequency reuse factor of 15!). With SFN, the spectral efficiency is dramatically improved. Provided that the transmission delays for signals from different transmission antenna towers are within the cyclic prefix length, OFDM can resolve this multipath, [D08], and actually readily exploit it to provide macro diversity.

The services provided by DAB and DAB+ include primary services, such as audio for main radio stations, but also secondary services, including sports commentaries, data services, video, etc., [ETSI_1, ETSI_2]. The older DAB standard is not forward compatible with the new DAB+, [ETSI_1, ETSI_2].

The minimum bandwidth of Eureka-147 in Mode 1 is 1.536 MHz with 1586 tones. The typical system uses a multiple of such 1.536 MHz bands. Each such band carries a multitude of audio channels in a TDM format, [wikiDAB]. The typical single-frequency network Eureka-147 system has a signal delay of about 2 seconds [wikiDAB].

Countries where DAB, DAB+, and DMB are in use (in 2011) include Canada, Singapore, Australia, China, South Korea, Taiwan, Sweden, Denmark, Norway, United Kingdom, Germany, Switzerland, Netherlands, Belgium, Spain, Portugal, Monaco, Greece, and Croatia, [wikifDAB]. It is interesting to note that a number of countries are in a category labeled “DAB no longer in use,” among them Finland. It is also worth noting that in Sweden DAB had nationwide coverage just a few years ago. Today (2011), however, DAB is only available in four large cities while the transmitters in the rest of the country have been switched off, due to low listener rates. The future is indeed uncertain.

27.3 HD-Radio, IBOC-FM, HIBOC-FM, IBOC-AM, and HIBOC-AM

The development of the digital audio broadcasting in the United States took a distinctly different path from the one in Europe. In contrast to Europe where new frequency bands were provided for digital audio broadcasting, no new frequency spectrum was allocated for this service in the United States. Therefore, adopting a standard like Eureka-147 would have been difficult, if not impossible, in the United States. Furthermore, broadcasters also requested a digital audio broadcasting system for replacing/upgrading analog AM radio. The services provided by AM radio have a significantly large market in the United States, in contrast to Europe, and Eureka-147 offered no solutions for the AM band.

A standardization effort started in the 1990s to develop systems in the existing FM and AM bands. These were labeled In Band on Channel, or, IBOC systems. The idea was to first keep the analog signal and add a digital audio signal in a hybrid in band on channel system both for FM and AM. In the future a migration to an all-digital system in both the FM and AM bands would take place by closing down the analog transmission and utilizing the freed-up spectrum for providing additional digital programming. Other systems like IBAC (in band adjacent channel) schemes were also explored, involving a hybrid system with a digital signal in the channel adjacent to the host analog FM. The IBOC schemes were however preferred.

The commercial label for these systems is High-Definition Radio (HD Radio). Currently (in 2011) both services in the FM and AM bands are offered throughout the United States and a multitude of radios are available. Several car manufacturers have been offering HD radio as an option.

The requirements for digital audio broadcasting were stereo CD-like quality in the FM band and stereo analog FM quality in the AM band with coverage for the digital systems comparable to that of the analog transmission. Thus far (in 2011), only hybrid systems are in use, but both the FM and AM band systems have transmission modes that include all-digital IBOC modes, [wIBQ,wIBQ_P,wIBQ_J], that is, both IBOC-FM and IBOC-AM systems.

Early contributors to the standardization efforts were AT&T and later Lucent Digital Radio, as well as CBS/Westinghouse and USA Digital Radio. Lucent Digital Radio and USA Digital Radio later merged to form iBiquity Digital Corporation, [wIBQ].

Federal Communications Commission (FCC) emission masks for FM and AM regulate the emission levels allowed from an analog FM and AM station [wFCC]. This combined with allowed interference levels for co-channel and adjacent channels form the bases for the IBOC designs. In North America, carrier separations of multiples of 200 kHz are used in the FM band and multiples of 10 kHz are used in the AM band (these numbers are different in, e.g., Europe). The digital signal is added to the analog FM signal in a frequency multiplexing fashion at a lower level (25 dB below the host analog FM signal) on both sides of the analog FM spectrum. In particular, letting f_c denote the carrier frequency of the host analog FM signal, the digital signal is added over bands spanning, for example, the spectrum from f_c − 200 kHz to f_c − 130 kHz, and from f_c + 130 kHz to f_c + 200 kHz, [wIBQ_P].

A similar arrangement is being made for AM over the f_c − 15 kHz to f_c + 15 kHz band, with the analog AM occupying the middle of the spectrum. There is, however, one conceptual difference between the AM and FM systems. In particular, unlike the FM case where the digital signal is added at the edges of the spectrum, in the AM case the analog AM signal is transmitted simultaneously with the digital signal in the center of the f_c − 15 to f_c + 15 kHz band, since, in the AM case, the digital signal also occupies the center of the band together with the host analog AM signal. This is straightforward to do in the case of analog AM, since AM amounts to linear modulation. In principle it can also be done with nonlinear analog FM, but it requires more complex receiver structures, [PS08,PS98,CS00a], and is thus not included in the current standard.

The HD Radio FM system is described in some detail in [F02,wIBQ_P]. A number of hybrid and all-digital transmission formats or modes have been developed. A control channel is used to inform the radio receiver about which mode is being used.

The audio coding used for HD Radio FM is AAC+ with spectral band replication (SBR) typically at 64 kb/s, [wikiHEAAC,wikiAAC]. The channel codes are complementary punctured pair convolutional (CPPC) codes, which constitute a special class of punctured convolutional codes of a type especially designed for hybrid IBOC FM, [F02,CS00b]. The signal is transmitted in two separate frequency bands on both sides of the carrier and host analog FM signal. The interference situation is such that a first adjacent signal can sometimes wipe out all or most of one of the two digital-signal sidebands. In this case the receiver is to reconstruct the digital signal based on only the surviving sideband. Sometime both sidebands are received and in this case the receiver softly combines them for more reliable reception, [LS01]. Sometimes a sideband partially damaged by interference can be useful in the context of soft combining.

For the described transmission scenarios so-called classic code combining is a possibility, [CS00b]. However, CPPC codes offer a superior option, [F02,CS00b]. Punctured memory 4, 16-state, rate 2/5 codes are used based on combining two different rate 4/5 codes on the upper and lower sidebands. The pair of rate 4/5 codes is obtained by an optimized complementary puncturing procedure on the rate 2/5 code. In contrast, classical code combining would employ identical rate 4/5 codes on both sidebands. As an example, with memory 4 codes, the optimum rate 4/5 codes would in all cases have a free distance of 4, while for the full rate 2/5 case the CPPC code has a free distance of 11, while the code combining code only have a free distance of 8 (i.e., twice the component-code free distance of 4). This constitutes a rare case of gaining in performance without increasing complexity. The modulation system used in HD Radio FM is OFDM with QPSK. Interleaving, both in time and frequency (tones), is also used, [wIBQ_P].

Initially only one digital program was envisioned, with content identical to the one carried by the host analog FM signal. However, the system evolved into a number of possible digital programs, HD1,HD2, and HD3, depending on the transmission mode employed. HD1 is the original format, with only one program in digital form, while HD2 and HD3, carry two and three programs, respectively. This is achieved by dividing the digital bit stream into separate streams. Different audio coder rates can be used for different programs. This also enables some trade-offs between the number of multiplexed programs and audio quality. Data services are also provided, [F02,wIBQ_P]. In the case that the content carried by the digital program is identical to the content of the host analog FM signal, blending from digital to analog can be used, [F02,wIBQ_P].

Among future possible improvements, the use of a list Viterbi algorithm (LVA) for decoding presents an attractive option for increasing robustness [SS94] at the expense of a small increase in receiver complexity. Audio systems often use error mitigation techniques to reduce the perceptual impact of transmission errors. This is achieved by using an error detecting code on the most important bits. When an error is detected, the audio segment is replaced by predictions from already received segments. By using the LVA, the prediction errors in this operation can be significantly reduced, resulting in improved audio quality. The use of the LVA only modifies the receiver. Another noteworthy option that would allow increasing the digital data rates that can be provided involves simultaneous analog and digital transmission by exploiting “writing-on-dirty-paper coding” schemes [C83]. These rely on inserting a weak digital signal on top of the dominant host analog FM transmission in such a way that the weak digital transmission can be reliably decoded even in the presence of the dominant analog FM interferer [PS08]. Since analog FM has an interference suppression capability, the added weak digital signal does not noticeably distort the demodulated host analog FM signal.

HD Radio AM actually presents a larger design challenge than FM, because of its stringent spectrum limitations, [F02]. A detailed description of the modes is given in [F02,wIBQ_J]. The audio coding used in HD Radio AM is AAC+ with SBR, typically at 24–32 kb/s, [wikiHEAAC,wikiAAC]. Trellis-coded modulation (TCM) and Reed Solomon coding are employed, and multi-stream transmission [S99,L02] (over multiple frequency bands with OFDM and various modems, ranging from BPSK to 64QAM) are used on different carriers. Unlike the FM case where the fading channel can be adequately modeled as a Rayleigh channel, the channel in the AM case behaves more like an additive Gaussian channel with added interference, [F02]. The HD Radio AM system represents a substantial improvement in audio quality over analog AM. Furthermore, stereo is achieved with HD Radio AM. Finally, we remark that, although the HD Radio schemes were developed in the United States for domestic use, these schemes are being used in and tested for use in several other countries, [wIBQ].

There are currently (in 2011) nearly 2000 FM and AM radio stations throughout the US broadcasting in HD, plus more than 1100 new local FM HD2/HD3 stations, [wHDR]. The Website www.hdradio.com provides a complete list of all radio stations transmitting in HD in the United States.

Portable radios with HD as well as home stand-alone radios with HD are now (in 2011) available at reasonable prices, [wHDR]. Car radio with HD for upgrades in existing vehicles and options in new vehicles for a large number of models are also now available, see [wHDR].

There is an increased adoption and testing/advanced interest for HD radio in many countries outside the United States, [wIBQ_I]. There are now (in 2011) adoption and nationwide operation in the US, Philippines and Puerto Rico. There are adoption and regional use in Mexico, Brazil and Panama. There is limited operation in the Dominican Republic, Jamaica, Switzerland, Ukraine, Thailand, and Indonesia. There is testing and advanced interest in many other countries including Canada.

27.4 Digital Radio Mondiale, DRM, and DRM+

Digital Radio Mondiale (DRM) is an international nonprofit consortium for digitization of broadcast (shortwave mediumwave and longwave) up to 30 MHz, DRM30. The DRM consortium is composed of broadcasters, network providers, transmitter and receiver equipment manufacturers, universities and research institutes. DRM+ extends operation to VHF bands up to 174 MHz, [wDRM]. DRM30, includes digital shortwave radio. The consortium was formed in 1998 and the first broadcast took place in 2003. Data Services are also included [wDRM]. Transmission is organized in a number of different modes, including hybrid adjacent modes, [wDRM].

DRM has ITU support, and the main standard has been published by ETSI. Among the audio coders used by DRM and DRM+ is MPEG-4 HE-AAC with SBR, [wikiHEAAC,wikiAAC]. The modulation used is coded OFDM (COFDM) with QPSK/16-QAM/64-QAM, [wDRM]. The channel codes used by DRM can be chosen to provide UEP with up to 4 different protection classes, [wDRM]. Layered modulation with multilevel coding is also used [RS08].

DRM (DRM 30) is the only international standard for HF (3–30 MHz) use, [wikiDRM]. DRM uses COFDM and it is designed to fit inside existing AM broadcast channels, with 10 kHz bandwidth in the US, and 9 kHz in Europe. For the European version, DRM has modes, which require as little as 45 kHz, [wikiDRM]. For DRM30 Systems with bandwidths of 9, 10, 18, and 20 kHz, see [wikiDRM]. DRM+ Systems use a bandwidth of 100 kHz, [wikiDRM]. For details of the OFDM schemes for DRM30 systems with 88–460 tones depending on mode and bandwidth, see [wikiDRM]. Current broadcasters (in 2011) include ALL India Radio, BBC World Service, Radio Canada International, Deutsche Welle and Voice of Russia, [wikiDRM]. Until recently, a DRM receiver has typically been a personal computer. However, stand-alone DRM receivers are now available, [wikiDRM].

27.5 ISDB, DVB-H, DMB, and Other Systems

ISDB is an ARIB Japanese standard for digital TV and audio. The 1seg part of ISDB is used for digital audio broadcasting to mobiles, [wikiISDB]. The technology is very similar to that used in Eureka-147 and DVB-H. DVB-T (Digital Video Broadcasting, Terrestrial) and DVB-H (Digital Video Broadcasting, Handheld) are digital TV standards developed under ETSI, [wikiDVBH] and as part of these standards there are also a digital audio broadcasting component, [wikiDVBH]. OFDM technology is used in these systems. As mentioned in the introduction, DMB [wDAB] is in use in Korea and China. It includes both digital video and digital audio broadcasting, and employs technology that is very similar to that of DAB and DAB+. All these systems in this section use high-power transmitters for transmission in large cells.

27.6 MediaFLO

This is a proprietary technology developed (and operated) by Qualcomm, [wikiFLO]. FLO stands for Forward Link Only. It is designed for digital video and audio broadcasting in the 716–722 MHz band. MediaFLO is a competitor to Eureka-147/DAB, DAB+, DMB [wDAB] and DVB-H, [wikiDVBH]. The technology is encrypted COFDM with QAM (QPSK and 16QAM) and Layered modulation with multilevel coding [RS08,C07]. High-power transmitters with tall antennas and large coverage areas are used for broadcasting, just like in conventional analog FM broadcasting. A detailed technical description of the physical layer in MediaFLO is given in [C07], [wQC].

27.7 MBMS and Convergence with Cellular

Multimedia Broadcast Multicast Services, or MBMS, is a multimedia broadcasting service standard [w3GPP] that has emerged under the umbrella of cellular and contains digital video and audio broadcasting over the cellular network. It has been developed for 3G UMTS (CDMA) [w3GPP] and there are also proposed versions for 4G LTE [D08]. Systems such as MBMS point to a convergence between digital audio broadcasting and cellular. Since MBMS is part of the cellular system, a single radio is used by the receiver, and only low-power transmitters are used. Additional details about the MBMS system are given in [BH05] and [I08].

MediaFLO and DVB-H are also offered as broadcasting services on mobiles. They however constitute a different (lower) level of convergence, as separate radios are used for cellular and broadcasting radio, and broadcasting has larger coverage areas (large cells) than cellular transmission. In contrast, a single radio is assumed with MBMS, serving both broadcasting and cellular transmission modes. MBMS is flexible enough so that it can be used for broadcasting, as well as for multicasting (transmission only to a selected group of users) and even unicasting [BH05,I08].

27.8 Sirius XM Satellite Radio

Sirius XM is the result of a merger between Sirius and XM in 2008. Sirius (originally Satellite CD Radio) was formed in 1990 and XM in 1992. The joint company provides satellite radio broadcasting in the United States (actually in North America), [wSXM]. The business model is based on the subscription pay radio principle. Nine Satellites are in orbit, 5 from XM, and 4 from Sirius. The frequency band for the satellite segments is the S band, 2.320–2.3325–2.345 GHz and the terrestrial gap filler segments use the same band, [wSXM]. Originally, the two companies Sirius and XM were licensed each in half of the allocated band.

The technologies used by the two companies Sirius and XM are slight variations of each other. Sirius uses geosynchronous satellites, which requires fewer terrestrial gap fillers. The audio coding is a multi-program PAC audio coding [N05] which has some statistical multiplexing benefits. The satellite segment has single carrier QPSK modulation and the terrestrial segment uses OFDM. XM uses geostationary satellites and HE-AAC audio coding, [wikiHEAAC,wikiAAC]. The satellite segment has single carrier QPSK modulation and the terrestrial gap fillers use OFDM. A great advantage with satellite radio in the United States is coverage. Indeed the same set of programs can be received throughout the 48 states in the continental United States.

27.9 Future of Digital Audio Radio

A very simple and inexpensive way to supply digital audio broadcasting is to replace the analog FM transmitter with a GSM type radio [wikiGSM]. This means that the analog FM modulator is replaced by a constant-amplitude GMSK modulator, [AAS86,MH81], yielding an all-digital in-band system using the same high-power transmitter. A significant advantage with such a system is that it uses a lot of the existing analog FM infrastructure. A new receiver is of course required and no obvious hybrid scheme is available. This is the reason why such schemes have not yet been adopted. A gross bit rate of up to about 150 kb/s before channel coding would be available for one transmitter. Different modes are possible, including multiple audio programs (e.g., one music and one talk program), or a single high-quality music program, and/or data channels. With more advanced Continuous Phase Modulation (CPM) constant-amplitude modulation schemes even higher data rate would be available at the expense of complexity, [AAS86]. For countries where the FM band is only sparingly used, digital transmission can be added on new frequencies. A system along these lines based on CPM has actually been built and field-tested, [BKB03,KAB08].

In Section 27.3, we described two potential future upgrades of the HIBOC-FM system. We described an improved receiver based on the concept of the List Viterbi Algorithm, [SS94]. (In Principle, this would also work for HIBOC-AM, IBOC-AM, and IBOC-FM). The second potential system improvement is related to simultaneous transmission of the host analog FM and digital, [PS08]. The net result of this is an increased data rate for the digital system.

What does the future hold for digital radio broadcasting? Audio coding schemes with higher quality than CD-quality [wikiDVBH,wikiDD] have been developed. They are based on higher audio source bandwidth and higher sampling frequency. As a result, higher audio quality than CD can be achieved. Radio broadcasting schemes with such audio coders would be much bandwidth demanding, since the source bit rate would be higher than in current systems. Multichannel (surround sound) schemes with spatial audio have also already been proposed for inclusion in the current standards. Perhaps unicasting may prove a preferred mode in the future for delivering audio. Indeed digital audio radio is already distributed as internet radio over wireless broadband on smart-phones and other terminals. In this case, it is no longer broadcasting but unicasting, that is, individual delivery instead of transmission to many. Finally, Digital Audio Radio is perhaps a good application area for Software Defined Radio (SDR), [B10].

Acknowledgment

The author wishes to thank Haralabos Papadopoulos for technical discussions and for reviewing the manuscript.

References

[F02] C. Faller , Technical advances in digital audio radio broadcasting, Proc. of the IEEE, 90(8), 2002, pp. 1303–1333.

[ETSI_1] ETSI EN300400 v1.4.1 Original DAB specification, www.etsi.org

[wDAB] www.eworlddab.org/technology/dab

[ETSI_2] ETSI TS102563 v1.1.1 DAB+ enhancement specification, webapp.etsi.org

[wikiDAB] en.wikipedia.org/wiki/Digital_Audio_Broadcasting

[wikiHEAAC] en.wikipedia.org/wiki/High_Efficiency_Advanced_Audio_Coding

[wikiAAC] en.wikipedia.org/wiki/Advanced_Audio_Coding

[LC04] S. Lin and D. J. Costello, Error Control Coding, 2nd Ed, Prentice-Hall, New York, 2004.

[C91] R. V. Cox , Sub-band speech coding and matched convolutional channel coding for mobile radio channels, IEEE Transactions on Acoustics, Speech & Signal Processing, 39(8), 1717–1731, 1991.

[D08] E. Dahlman , 3G Evolution, 2nd Ed, Academic Press, New York, 2008.

[wikifDAB] en.wikipedia.org/wiki/File:Digital_Audio_Broadcasting.svg

[wIBQ] www.iBiquity.com

[wIBQ_P] P. J. Peyla, The structure and generation of robust waveforms for FM. In Band on Channel Digital Broadcasting, www.iBiquity.com

[wIBQ_J] S. A. Johnson, The structure and generation of robust waveforms for AM. In Band on Channel Digital Broadcasting, www.iBiquity.coms

[wFCC] www.fcc.gov

[PS08] H. C. Papadopoulos and C.-E. W. Sundberg, Precoded modulo-precancelling systems for simulcasting analog FM and digital data, IEEE Transactions on Communications, 56(8), 1279–1288, 2008.

[PS98] H. C. Papadopoulos and C.-E. W. Sundberg, Simultaneous broadcasting of analog FM and digital audio signals by means of adaptive precanceling strategies, IEEE Transactions on Communications, 46(9), 1233–1242, 1998.

[CS00a] B. Chen and C.-E. W. Sundberg, Digital audio broadcasting in the FM band by means of contiguous band Insertion and precancelling techniques, IEEE Transactions on Communications, 48(10), 1634–1637, 2000.

[CS00b] B. Chen and C.-E. W. Sundberg, Complementary punctured pair convolutional codes for digital audio broadcasting, IEEE Transactions on Communications, 48(11), 1829–1839, 2000.

[SS94] N. Seshadri and C.-E. W. Sundberg, List Viterbi decoding algorithms with applications, IEEE Transactions on Communications, 42(2/3/4), 313–323, 1994.

[LS01] J. N. Laneman and C.-E. W. Sundberg, Soft selection combining for terrestrial digital audio broadcasting in the FM band, IEEE Transactions on Broadcasting, 47(2), 103–114, 2001.

[S99] C.-E. W. Sundberg , Multistream Hybrid On Channel Systems for digital audio broadcasting in the FM band, IEEE Transactions on Broadcasting, 45(3), 410–417, 1999.

[L02] H.-L. Lou , Multi-stream transmission for Hybrid IBOC AM with embedded multidescriptive audio coding, IEEE Transactions on Broadcasting, 48(3), 179–192, 2002.

[wHDR] www.hdradio.com

[wIBQ_I] www.ibiquity.com/international

[wDRM] www.drm.org

[RS08] S. A. Ramprashad and C. E. W. Sundberg, Hierarchical QAM BICM MIMO systems with iterative decoding and applications to media broadcast, IEEE WOWMOM 2008, Newport Beach, CA, June 2008.

[wikiDRM] en.wikipedia/wiki/Digital_Radio_Mondiale

[wSXM] www.siriusxm.com

[N05] N. Naik , Joint encoding and decoding methods for digital audio broadcasting of multiple programs, IEEE Transactions on Broadcasting, 51(4), 439–448, 2005.

[wikiISDB] en.wikipedia.org/wiki/ISDB

[wikiDVBH] en.wikipedia.org/wiki/DVB-H

[wikiFLO] en.wikipedia.org/wiki/MediaFLO

[wQC] www.qualcomm.com

[C07] M. R. Chari , FLO Physical Layer: an overview, IEEE Transactions on Broadcasting, 53(1), 145–160, 2007.

[w3GPP] www.3gpp.org

[BH05] M. Bakhuizen and U. Horn, Mobile broadcast/multicast in mobile networks, Ericsson Review, 82(1), 6–13, 2005.

[I08] J.-A. Ibanez , Mobile TV over 3G networks—Service and enablers evolution, Ericsson Review, 85(1), 38–42, 2008.

[wikiDD] en.wikipedia.org/wiki/Dolby_Digital

[B10] D. Bodson, Digital audio around the world, IEEE Vehicular Technology Magazine, 5(4), 24–30, 2010.

[C83] M. Costa, Writing on dirty paper, IEEE Transactions on Information Theory, 29(3), 439–441, 1983.

[wikiGSM] en.wikipedia.org/wiki/GSM

[AAS86] J. B. Anderson, T. M. Aulin, and C.-E. W. Sundberg, Digital Phase Modulation, Plenum, New York, 1986.

[MH81] K. Murota and K. Hirade, GMSK modulation for digital mobile telephony, IEEE Transactions on Communications, 29(7), 1044–1050, 1981.

[BKB03] G. de Boer, C. Kupfenschmidt, D. Bederov, and H.-P. Kuchenbecker, Digital audio broadcasting in the FM band based on continuous phase modulation, IEEE Transactions on Broadcasting, 49(3), 293–303, 2003.

[KAB08] C. Kupfenschmidt, A. Ayadi-Miessen, G. de Boer, R. Patino, and H.-P. Kuchenbecker, Field tests of digital radio broadcasting in the FM band based on continuous phase modulation, IEEE Transactions on Broadcasting, 54(2), 236–248, 2008.