If you are not a regular reader of our comments section you may have missed a learned and well informed debate has been raging (well perhaps ‘raging’ is an overstatement) about DAB+ in response to a recent article on ten years since Australia’s digital radio launch.
It has built up a good cross section of opinions about the success of DAB+ in Australia and about the audio quality of transmission.
Here’s your summary of the debate so far:
Anthony the Koala, a regular contributor to various topics we report on, started it all by posting a response to this article about ten years of digital radio:
The DAB+ system is an improvement on the test transmissions conducted in the early 2000s using MP3 streaming instead of HE-AAC streaming. The latter system permits more stations to be broadcast for a given ‘fidelity’.
However, while the DAB+ has a superior sound, some of the ‘permanent’ auxiliary stations associated with the main FM or AM station are transmitting on 32kbs stereo. I don’t count Macquarie’s NTS which is mainly broadcasting speech in mono. Nevertheless, 32kbs stereo streams have a “metallic sound”. The sound is acceptable at rates of 48kbs in stereo. ABC-FM, RN and Metropolitan ABC (in Sydney 702 (2BL)) broadcast at 70kbs. In Sydney the best signal is 2GB and 2CH at 128kbs stereo.
He also raised the issue of transmission bands (currently the VHF Band in Australia) and alternative transmission systems for regional areas.
Raymond responded that he was worried that DAB+ has not expanded across the whole Australian continent yet:
That in itself is a worry about the real commitment of the industry to adopt the medium in the way it trumpeted on launch day a whole decade ago.
He also drew attention to New Zealand’s approach to DAB and why NZ won’t be introducing it any time soon. “It’s a revealing read,” says Rayond, linking to this document.
Christer then weighed into the debate saying: “After 30 years still few countries are on the DAB trail. The unique national FM switch-off in Norway is a fiasco.”
GerardW proposed that community radio should move off the FM band to make more space:
A few days later StJohn responded to Anthony, taking issue with his belief that the sound is tinny, and raising the DRM question for large coverage areas:
Don’t convert the country to FM, the coverage areas become smaller than AM. Use DRM which can sound as good over larger coverage areas, but also transmit pictures and multipage indexed text as well.
That led to a few back and forth comments from both Anthony and StJohn, politely differing from each other’s opinions and adding other interesting technical points.
Anthony has just added a further comment today about India’s DRM transmissions and more about the quality of transmission (Read more about radio in India in our sister publication AsiaRadioToday). One of the questions raised in that post is about Indian commercial radio transmitting in DRM, the answer is that only the national broadcaster AIR currently has some transmissions in DRM, there are no commercial stations broadcasting in digital.
We are delighted that our site stimulates informed, polite debate from all viewpoints. We don’t necessarily agree with all the points made, but we are pleased that readers feel that our publication is the place to have a sensible discussion about issues.
We verify people who post, but we never reveal their identities and we allow them to post under pseudonyms if they wish.
We always moderate comments, based on our policies, and read and sometimes edit them before we make them live.
When you register, you can opt in or out of our three newsletters: twice weekly, daily or job alerts, just tick the appropriate boxes.
Anthony,
Many AM stations use processors to try and sound like FM by boosting the high frequency audio because of the very poor high frequency response of AM receivers. This will particularly affect the SBR signal. X Digital in Perth only transmits music and some advertising but not much speech. There is plenty of bass.
The problem for broadcasters is that they can preview MP2, MP4 files and they sound fine, but once transmitted are poor sounding regardless of the bit rate.
As far as India goes, India does not have its population evenly distributed. The Government is trying to cover the whole of the country because it’s All India Radio is the only broadcaster allowed to transmit news and current affairs. They do separate the networks like the ABC local radio does. There FM commercial broadcast industry is tiny. http://www.asiaradiotoday.com/news/drm-association-radio-operators-india’s-agm shows that DRM+ is likely because it can transmit 3 programs on one transmitter. Two DRM+ transmitters can fit into a single FM channel. A main driver for DRM in India is the Emergency Warning System will save millions of lives from the floods of the dual monsoons per year.
As far as cars goes, they have only just started installing DRM in cars and have done it to 1.5 million cars in 18 months and they have only just started. An app has been written for the FM tuner in most Android phones which are much more popular than iphones.
The ACMA has already stated that they are only converting the remote and a few country AM licences to FM as the band is too crowded elsewhere. You can fit 6 DRM+ audio programs onto one FM transmission channel.
As far as Australia goes the cost difference between FM and DRM+ is no where as big as you claim as the single transmitter is used for both of each broadcasters’ programs. The only difference is the modulator and a content server to combine their pair of audio programs to fit into one digital stream. They can add coloured pictures for advertising and could also radiate an electronic text “newspaper” using Jounaline.
So, the cost is reduced because the AM transmitter is switched off as well as the FM transmitter. It can be replaced with a DRM+ transmitter of 1/ 10th of the power for the same coverage area. In remote areas such AM transmitters can be modified to DRM, however there is currently one commercial station on High Frequency (Short Wave) which can economically cover large remote areas. Converting it to DRM will give good stereo sound and it is easily tuned.
Your comment about tuning ranges is exaggerated. FM radios tuning below 87.5 MHz Lower FM frequencies were only used in Japan and the USSR. USSR is now testing DRM+ at a frequency around our analog TV channel 2. All digitally tuned AM receivers are selectable for 9 kHz spacing for 531 – 1602 kHz or is 530 – 1700 on 10 kHz spacing which is only used in the Americas. Much of Europe has no AM at all. No mobile phones virtually no DAB+ radios will receive AM at all.
Over 67% - 100 % of the power of an AM signal is wasted in the carrier. DRM does not have a carrier resulting in drop in electricity consumption. It uses COFDM just like our digital TV.
St John
Dear StJohn,
Thank you for your reply.
You have confirmed what I said about using lossy compression files in the transmission/reception chain sounding poor.
Thank you for elucidating on the number of 'services' on one DRM channel being three. If two DRM channels can fit onto one FM frequency, given one FM frequency occupies 200kHz, then one DRM channel is 100kHz. It follows that in the AM band typically 1602 to 522 khz has a bandwidth of 1122kHz can fit approximately 10 DRM services. On the FM channels, 108MHz-88MHz = 20Mhz, which gives approximately 20/0.1 = 200 theoretical DRM transmitters. With 3 services per DRM transmitters, that's the equivalent to 600 DRM services. Theoretical because that does not take account of any co-channel interference whether it's on MW or VHF.
HOWEVER, I do note that with DRM and DAB+ transmissions, the same frequency can be used in different areas PROVIDED that all DRM and DAB+ are synchronized using accurate clocks. Synchronization shouldn't be hard because atomic clocks have been used in broadcasting for decades.
My calculations were based on the third paragraph of three programs on one transmitter, NOT the fifth paragraph, where you say that one can fit 6 DRM channels on one FM channel. Please clarify the 3 channels and 6 channels distinction.
Nevertheless, that does not account what for what is final government policy on spectral allocation of DRM services.
One question I would like to ask you is when the DRM signal is on MW or SW bands and due to the properties of the ionosphere such that MW and SW signals skipping long distances is: what is the quality of the received signal over long distances? In addition is there enough error correction information sent with the DRM signal that the recovered signal at the receiver is 'perfect'?
I noted in your previous answers (plural) that DRM has the facility to transmit pictures and text. The questions about pictures and text services raises non-engineering questions but economic: will the radio stations be able to afford providing pictures and text services OR will the radio stations want to monetise and and let other parties pay to use the text and pictures services. A corollary to auxiliary text services is the teletext service, text services on analogue tv as well as DVB. Teletext services providing news, weather, market information was abandoned by the Seven network nearly 20 years ago. The justification for the abandoning of teletext services was that such information was available on the web! Similarly the potential of DRM to provide text and picture services, one asks will it be utilized when the information is available on the web, especially the radio station's web page.
REGARDLESS of technology: whether DRM or DAB+ or web pages, it is an issue for the broadcaster to keep information up-to-date. For example, a former 2UE personality who left 2UE nearly three years ago is still publicised on a rural radio station's web page http://www.hayfm.org/wp/show/stuart-bocking/ . Nevertheless, I do note that DRM's capability to send pictures and text - regardless of DRM or WEB technology, the information needs to be up-to-date.
I must disagree with you about the alleged 'exaggeration' of the use of the FM band. My description of the FM band is EXACT. Yes I am aware for example that VHF FM bands are not the same world wide. For example Japan's VHF FM band is between 75MHz and 95MHz. I was talking about the 88MHz to 108MHz band as used in the Australian market place. All my 88MHz to 108MHz FM receivers can receive frequencies beyond this band. There are services used below 88MHz for example in Sydney for multicultural languages such as Campbelltown's Spanish language station "Radio Austral" 87.6MHz, dance tracks on "Raw FM" to name a few. Source, https://www.acma.gov.au/-/media/Licence-Issue-and-Allocation/Information/Word-Document/High-power-and-low-power-radio-open-narrowcasters-programming-and-broadcasting-data.doc?la=en - definitely not an exaggeration.
In addition, though the MW AM band is officially 522kHz-1602kHz, there are radios sold and marketed in Australia that have its AM band go to 1701kHz. Pioneer A/V receivers, for example VSX-932 and Toyota Camry's radio goes to 1701kHZ to name a few. You can receive in Sydney "The Voice Of Charity" on 1701kHz.
An observation, this is not numerology. Nevertheless, without mathematical proofs, adding the digits of the frequency of a radio station, if divisible by 9 is indicative of a 9kHz-spaced radio frequency. To illustrate for Sydney frequencies, RN on 576 kHz, adding the digits 5+7+6 =18 is divisible by 9. Another example 1701kHz, add the digits, 1 + 7 + 1 = 9, is divisible by 9. One final example 2GB at 873kHz, 8 + 7 + 3 = 18 which is divisible by 9.
One final point. I agree with about the carrier in an AM mode being wasted. I already stated that the carrier remains at a constant level regardless of the amplitude of its side bands.
I thank you for enriching the public at large about the capability of DRM. Please clarify whether a DRM transmitter can carry 3 or 6 signals on one tx, and the quality of a decoded DRM signal over long distances especially on the MW and SW bands.
Thank you,
Anthony of exciting Belfield
Anthony,
I left out a couple of answers at the end of your post.
If DRM30 ie below 30 MHz is used the bandwidth is limited to 20 kHz or less.
For music only one program per transmitter. Maximum for 18 kHz bandwidth is 64 kbit/s For speech stations you can have atleast 2 because you can use a lower data rate, infact, the DRM consortium has specified the latest most efficient compression system xHE-AAC It can produce excellent speech at 12 kbit/s. DRM+ because of its wider bandwidth can have 3 stereo programs as mentioned previously.
You can have near FM stereo quality from a DRM transmission in the medium and high frequency bands! You can also have annoyingly noisy AM reception particularly at high frequency, but replace it with DRM and it is a reliable noise free stereo with the high pitched sound in tact. There is also no phasing effects caused by multiple reflections from the ionosphere.
Dear StJohn,
Thank you for your reply, we are better informed about DRM. Two points, the ubiquity of DRM (or even DAB+), the disposal of e-waste and patent law regarding the HE-AAC codecs.
(1) Ubiquity of AM and FM radios and making DRM as ubiquitous. As a matter of 'common-sense' policy, to switch off the analogue services requires both simulcasting the analogue services and a new DRM (or DAB+) service for a period of time until the market place is saturated with DRM receivers.
(1a) Disposing of e-waste from non-functioning analogue receivers and transmitters. There has to be a policy about what to do with non-functioning analogue radios and transmitting equipment. This has to be a genuine policy where parts can be recycled and toxic compounds prevented from entering the environment. Not just landfill!
(1b) I need more knowledge on whether current AM (MW and HF) and FM (VHF) transmitters can be retro-fitted with digital exciters.
(2) Given the DRM is open-sourced, the implementation of the HE-AAC codecs (coders for tx, decoders for rx) is subject to paying a licensing fee to the "Via Corporation". There will be an eventual expiry of the patent, but given that HE-AAC is a suite of compression technologies, source http://www.via-corp.com/us/en/licensing/aac/faq.html it may take a few years until the the whole suite of technologies is patent-free.
Thank you,
Anthony of really exciting Belfield
Dear StJohn,
Thank you again for the information regarding DRM.
Just a couple of points on analogue and digital transmissions and policy and IP and not getting ripped off.
Analogue and digital transmissions - policy and updating new codecs.
(a) Simulcasting both analogue and digital. When I mentioned simulcasting of analogue and digital radio transmissions, I did infer an analogue transmission cut-off date. You hit the "nail on the head" with the analogy of the introduction of digital TV from 2001 and the analogue cutoff date of 2013 and that we did not look back since adopting digital TV.
(aa) In regards to analogue TV. I recall our house was near the former AGL gasometers in Enfield/South Strathfield which have since been demolished. When the gasometers were operable, the received analogue transmissions contained a lot of ghosting. During electrical storms, the video's AM signal would deteriorate resulting in a noisy picture. I don't look back to analogue signals. Though I did miss watching dx signals on WIN 4 Wollongong, NBN 3 Newcastle, and the occasional glimpse of TVNZ on VHF Channel 0.
(aa) Technological change in codecs. In the analogue days, the analogue codecs for TV was a monochrome compatible PAL, NTSC or SECAM and its variants for nearly 60 years (starting with NTSC in 1954). It was stable. Similarly in radio, we've had AM for 100 years and FM for nearly 80 years (invention, adoption 60 years). Manufacturers, broadcasters and users had a reliable method of transmission.
In the last 20-30 years there have been changes in digital audio codecs, for example MP2, MP3 and HE-AAC. Each codec was an improvement on the other such that more 'relevant' (remember it's lossy and redudant info is removed) information could be transmitted for a given fidelity.
Similarly in digital TV, a digital TV sold between 2001 and 2012, despite being digital may not receive the new high definition channels and auxiliary channels because the codecs were not available. But the new codecs such as H.264 enable more relevant (recall it's lossy) video and audio information to be transmitted for a given video and audio quality, sources: https://www.smh.com.au/technology/hd-tv-new-channels-may-not-be-supported-by-your-television-20151201-glccvr.html and https://www.acma.gov.au/theACMA/engage-blogs/engage-blogs/Interference/MPEG-4-video-Unravelling-the-standard . Today, TVs have the HEVC = HE.265 to decode 4k transmissions whether by broadcast or VOD.
The result is that the digital standards for codecs change resulting in some receivers not being able to decode pictures and audio. Put it another way, unless the TV receiver can adopt the new codec, one will have to purchase a set top box capable of decoding the latest decoder.
Could the same problem occur with digital radio or TV transmissions where a clever bunch of scientists devise a new and efficient codec for transmitting more information? What if for example if there is an improvement to the HE-AAC format for DAB+ or DRM?
The rhetorical question, is it good policy to allow for changes in newer codecs making modern receivers inoperable? Alternatively, should there be provision for transmitters and receivers to enable any future adoptable codec via a software update?
IP for codecs and don't get ripped off.
(b) IP for codecs. StJohn, while Fraunhofer was responsible for the MP3 format, it is not responsible for the development of the HE-AAC format. As I mentioned in the previous posting, the HE-AAC was developed by many developers not just Fraunhofer.
HE-AAC is not one codec but a suite of technologies that are developed by a number of entities that enable the transmission of audio via broadcast and IP streaming. To purchase a suite of HE-AAC technology, it is licensed through one 'consortium' "Via Corporation", sources: http://www.via-corp.com/licensing/aac/overview.html and see "What technologies are covered ...." heading under http://www.via-corp.com/licensing/aac/faq.html
For digital TV whether broadcasting or VOD, it is not a matter of "it's just an HEVC/H.265 codec..." it's a suite of codecs, see heading "Patent holders" at https://en.wikipedia.org/wiki/High_Efficiency_Video_Coding.
My last remark is if any 'small person' develops any technology that improves on the previous technology that the 'small person' is not ripped off by large corporations with abundant financial resources to litigate. Let's not forget the tragedy of the late Herbert W Armstrong, the inventor of FM who was ripped off by larger corporations.
However what happened to Mr Armstrong shouldn't stop a 'small person' from developing a new method to improve the previous art base/technology. The issue of IP protection and risk management of ideas leaking into the marketplace is another issue.
Regards
Anthony of exciting and dynamic Belfield
Anthony,
Yesterday, I have written to the ACMA to bet them to publish on their website when they will publish the submissions for "The future delivery of Radio in Australia".
Analog TV were ubiquitous and we switched to digital, why not radio? Just remember that there is a lot less electronics in a radio than a TV. Remember that analog radio particularly AM radio is causing lots of carbon dioxide to be produced in Australia except Tasmania, who use hydropower.
The reason e-waste is recycled is the value of the recovered metals. Remember that many phones are replaced not only to change the generations, remember that CDMA, 2nd generation phones have been recycled because the telco's stopped running those networks as well as fashion.
As far as upgrading transmitters is concerned, there are still plenty of old transmitters which use a lot more electricity than new ones even in analog modes. Also old transmitters it is hard to get parts. So compare the damage to the environment of carbon dioxide caused by increased electricity consumption over many years.
The through the air loss of a signal for DAB+ is 12 times that for band 1 DRM+ The comparison between DRM+ in the FM band is 4 times that for band 1.
Remember that the patent cost decreases per receiver, depending on the number of receivers produced by a company. This is why an analog switchoff date needs to be set to get the public to buy many receivers quickly.
The patent has runout on COFDM modulation so the CSIRO no longer gets payments from manufacturers including Wifi manufacturers. The CSIRO sued computer manufacturers for not paying patents, and were paid more than $400 million.
Anthony,
Simulcasting it is not effective in causing people to buy new receivers. A cutoff date is. Look at the our change to Digital TV. Simulcasting started in 2001 and receiver sales only took off in the 2010 - 2013 when the public were warned of an analog switchoff were occurring. I doubt a soul would want to go back to analog now! In Norway they progressively switched off all national networked FM transmitters over a year, leaving only low powered local FM stations. Within one year the ratings had returned to normal. Simulcasting, particularly AM simulcasting is wasteful in electricity and maintenance charges.
E waste recycling was covered in the ABC TV program "War on waste". The recycle these products for the metals. It is not economic to recycle components, particularly when many are old and may now not meet original specifications.
Broadcast Australia has 4 MF and 1 or 2 HF DRM compatible transmitters, most newer AM transmitters can be converted to DRM. The main criteria is that the signals must not go through a capacitor, thus loosing the DC reference.
FM transmitters can be converted to DRM+.
Remember also that a digital TV transmitter has a 7 MHz bandwidth which is full of similar signals to DAB+/DRM+. A good linear transmitter could carry more than DAB+/DRM transmission channel, using multiple content servers and modulators with sharp output filters. After all how can the full range of DAB+ programs radiated for the ABC/SBS into regional Australia. You would not wish to put 6 DRM+ transmitters on each site!
Fraunhofer needs to be paid for the development of its compression software/firmware. Hence the patent. DAB+ is already using HE AAC v2 in Australia, the latest upgrade comes at a cost to content server makers and receiver manufacturers. The scientists and engineers need to be paid to keep making these significant advances.
Dear StJohn,
E-Waste: It reminds me of the 'warning' on many electrical goods "no user-serviceable parts inside".
Reducing CO2: In the context of broadcasting, the biggest emitter of CO2 are the ABC who use transmitters of 50kW for metropolitan MW stations when commercial stations' transmitters are 5kW. The same could be said of some rural ABC transmitters at 100kW when the local commercial station is on 2kW. That is for AM transmitters only. That does not take account that ABC's FM transmitters consume more power than its commercial counterparts especially in rural areas.
HF DRM-capable transmitter: StJohn, when you mentioned that BA have at least one DRM-capable HF transmitter. Since 2017 the ABC has abolished SW transmissions via Radio Australia ('RA'): transmissions are delivered by satellite and IP streaming. With the DRM-capable HF transmitter, would DRM HF transmissions resume? That would be a matter of ABC policy. Consideration should be made especially for our Asian and Pacific neighbours in regards to the cost of DRM receivers. The corollary is that with DAB+ receivers still more expensive than the ubiquitous AM/FM despite 10 years of DAB+. Similarly how much is a DRM+ will it cost for a DRM receiver for citizens of our Asian and Pacific neighbours compared to the ubiquitous AM/FM/SW receiver?
StJohn, thank you again for informing us,
Anthony of exciting Belfield
I wish to add to the discussion against the use of lossy compression in broadcasting and the further aggravation of quality caused by recording equipment.
As I was browsing the comments archives of this site, I came across the name of renowned designer of audio mixing desks for various TV and radio stations including the ABC, Mr Poul Kirk.
One of the earliest archived articles was against the use of lossy audio files in the production of advertisements. In this article, Mr Des DeCean commented that despite the convenience of size of a lossy audio file such as MP3 and MP4, counselled against its use of emailing lossy files. Mr Poul Kirk concurred with that, source: https://radioinfo.com.au/news/avoid-using-email-digital-radio-ads. The preferred method of 'transfer' between studios is lossless compression such as WAV.
The quality of sound is further aggravated by the quality of the source recording. Mr Kirk commented on the quality of recording equipment used by 'garage bands', https://radioinfo.com.au/news/dont-play-bad-quality-music-radio-poul-kirk .
In a previous life working in a radio and TV station, I was trained to be critical of the signal going to air. Today, I am listening to CDs of supposed 'digitally remastered' recordings, one can still hear 50Hz or 60Hz hum and its harmonics and tape hiss. Here are a few examples. These are not exhaustive.
* 'Smiley' sung by Ronnie Burns. You can hear 50Hz hum and its harmonics every time the harpsichord is played. You can still hear tape hiss.
* 'Eagle Rock' sung by "Daddy Cool". Introduction to the music featuring the lead guitar and the singer vocalising "Now listen!.......", 50Hz hum, its harmonics and tape hiss.
* 'Ebb Tide' sung by the 'Righteous Brothers'. In the beginning you can hear 60Hz and its harmonics in the beginning only.
* 'Step Inside' sung by Cilla Black. First few bars before the first bass note, you can hear 50Hz hum and its harmonics. It is repeated in another segment of the song. This suggests that the loud parts of the music are masking the hum.
Just a thought,
Thank you,
Anthony from exciting and dynamic Belfield